0

28th Biennial Mechanisms and Robotics Conference

2004;():1-5. doi:10.1115/DETC2004-57008.

A new fully compliant linear-motion mechanism, called the XBob, is presented. The mechanism is based on the pseudo-rigid-body model (PRBM) of a system of Roberts approximate straight-line mechanisms combined in series and parallel. It can be fabricated in a single plane and has a linear force-displacement relationship. Symmetry and compliance compensate for the structural error inherent in the Roberts mechanism, resulting in precise straight-line motion. The device is designed and its motion and force-displacement relations are predicted by the PRBM. The design is validated using finite element analysis and experimental results.

Topics: Motion , Mechanisms
Commentary by Dr. Valentin Fuster
2004;():7-12. doi:10.1115/DETC2004-57009.

This paper presents an original technique to locate the center of curvature of the path traced by an arbitrary point fixed in the coupler link of a planar four-bar linkage. The method is purely graphical and the center of curvature; i.e., the center of the osculating circle, can be located in a direct manner with few geometric constructions. The advantage of this technique, compared to the classical approach using the Euler-Savary equation, is that measurements of angles and distances between points are not required. Also, it is not necessary to locate inflection points or draw the inflection circle for the instantaneous motion of the coupler link. The technique is based on the novel concept of a virtual link, which is valid up to, and including, the second-order properties of motion of the coupler link. The virtual link is coincident with the path normal to the coupler curve; i.e., the line connecting the coupler point to the velocity pole of the coupler link. The absolute instant center of the virtual link defines the ground pivot for the link and is, therefore, coincident with the center of the osculating circle. The authors believe that the graphical technique presented in this paper represents an important contribution to the kinematics literature on the curvature of a point trajectory.

Commentary by Dr. Valentin Fuster
2004;():13-21. doi:10.1115/DETC2004-57010.

In this paper, we consider the problem of cooperatively localizing a formation of networked robots/vehicles in SE (2). First, we propose necessary and sufficient conditions to establish when a team of robots with heterogeneous sensors can be localized. We then show how these conditions are analogous to well-known results in the literature on kinematics of planar mechanisms. We show how localization is equivalent to solving a system of nonlinear closure equations. Depending on what sensors are available for each robot, the multirobot formation can be modeled as a sensing graph consisting of vertices representing robots and edges corresponding to sensory information. We establish conditions that must be satisfied by this graph and show how this graph influences estimates of positions and orientations of the robots in a team through experiments and simulations.

Topics: Robots
Commentary by Dr. Valentin Fuster
2004;():23-31. doi:10.1115/DETC2004-57015.

PD control is widely used in industrial robotic manipulators because of its simple structure and acceptable performance. In this paper, the PD-based control schemes for the trajectory tracking of the robotic manipulators are addressed. The fixed gain PD control, the nonlinear gain PD (NPD) control, the adaptive PD learning control (PD-LC), and the adaptive NPD learning control (NPD-LC) are applied for the trajectory tracking of both serial and parallel robotic manipulators. The PD-LC and NPD-LC controllers can be used to improve the tracking performance for the repeatable tracking tasks in an iterative mode. The PD-LC and NPD-LC consists of a PD/NPD control as the basic feedback control and an additional feedforward control term directly inherited from the previous iteration of the same control task. A comparative study of four PD-based controllers is conducted to understand how different control schemes will affect the trajectory tracking performance, and the results are shown in this paper. Case studies are presented to demonstrate the validity of the PD-LC and NPD-LC algorithms.

Commentary by Dr. Valentin Fuster
2004;():33-41. doi:10.1115/DETC2004-57019.

Recent research in morphing wing technology has focused on complex multiple-degree-of-freedom (MDOF) mechanisms and smart structures to provide a specified shape change; single degree-of-freedom actuation concepts have generally been ignored or overlooked. In this research, the authors propose a novel design for a single degree-of-freedom (SDOF) mechanism for mission morphing of aircraft wings. A general design methodology has been developed, and this has been applied to a Hyper-Elliptic Cambered Span (HECS) wing developed by engineers at NASA Langley Research Center. The design tools developed include a synthesis procedure for determining the dimensions of the single-degree-of-freedom morphing mechanism, and a sensitivity analysis to determine the effects of manufacturing errors.

Commentary by Dr. Valentin Fuster
2004;():43-52. doi:10.1115/DETC2004-57020.

In this study a procedure to obtain direct singular positions of a 3RPS parallel manipulator is presented. If the heights of 3 spherical joints, denoted by d1n , d2n , and d3n respectively, are used as coordinate axes, then the workspace of the moving platform may be represented as an inclined solid cylinder in this coordinate system. The location of a point on the solid circular cylinder determines a configuration of the manipulator’s moving platform. The procedure to locate direct singular positions consists of two steps, the orientation of the moving platform is assumed first, from which the horizontal position of the moving platform may be obtained; then in the second step the heights that make determinant of Jacobian matrix vanish may always be determined. Results show that unless the moving platform is normal to the base, in which case there exist only one or two singular configurations, otherwise there are always three singular configurations corresponding to a moving platform’s orientation.

Topics: Manipulators
Commentary by Dr. Valentin Fuster
2004;():53-60. doi:10.1115/DETC2004-57024.

A fast, accurate and robust position controller is proposed for a planar three degree of freedom robot arm actuated by rotary Shape Memory Alloy actuators and servomotors. Servomotors are used to actuate the first link and the gripper, while the remaining two links are actuated by a combination of shape memory alloy wires and pulleys. Initially, a model of the robot arm is developed for theoretical controller development. The model combines robot kinematics and dynamics with the SMA wire heat convection, constitutive law, and phase transformation equations. The model is then used to develop several nonlinear position controllers based on the Variable Structure Control, also called sliding mode control. The controller development is of an evolutionary nature starting from a simple switching control based only on position feedback and then adding velocity and integral feedbacks, respectively. Several simulations of the proposed controllers are presented. Several experiments have been performed with a desktop prototype of the robot arm. The experimental results verify the effective and accurate performance of the controllers despite significant modeling inaccuracies and parameter uncertainties.

Topics: Robots
Commentary by Dr. Valentin Fuster
2004;():61-68. doi:10.1115/DETC2004-57026.

The dynamic characteristics of a six bar hinge mechanism as used in home cabinets were investigated using the method of equation of motion. The derived equations of motion were numerically solved and the motion of the hinge mechanism was simulated. The influence of mass and width of the cabinet door on the dynamic characteristics of the hinge mechanism as well as the effect of the hinge number on the force applied on the handle were also investigated. The experimental and simulation results showed good agreement with an error of under 2%, which validated the simulation results. The proposed approach can be used by hinge manufacturers for the design and analysis of similar hinge mechanisms.

Topics: Hinges , Mechanisms
Commentary by Dr. Valentin Fuster
2004;():69-76. doi:10.1115/DETC2004-57028.

This paper introduces the force analysis of the XYZ Micromanipulator (XYZM). The XYZM has three independent linear inputs and a positioning platform. The positioning platform remains horizontal throughout its motion and has translation in the x, y, and z directions. The design and displacement analysis for the XYZM was reported previously. This paper concentrates on the input-output force relationships and the derivation of the XYZM kinematic coefficients. Equations for the three versions of the XYZM are reported and sample results provided. Slider displacements of 45 micrometers result in a predicted out-of-plane displacement of 188 micrometers for the rigid-body XYZM and 262 micrometers for the compliant XYZM. This correlates to output force of 286 micronewtons with an input force of 150 micronewtons for the rigid-body XYZM and 261 micronewton output force with an input force of 150 micronewtons for the compliant XYZM.

Commentary by Dr. Valentin Fuster
2004;():77-85. doi:10.1115/DETC2004-57029.

A novel 5-UPS/PRPU 5-axis PMT (Parallel Machine Tool) is studied. The stationary platform is connected with the moving platform by the same five UPS actuated limbs and a PRPU passive constraining limb. Three translational DOF (degrees of freedom) and two rotational DOF can be achieved. by using kinematic screw theory and D-H parameter method, the kinematic forward and inverse solutions of the PRPU constraining limb are deduced. The rotational constraint along the vertical axis of the moving platform acted by the PRPU limb is confirmed. Moreover, the configuration of the moving platform can be acquired online by installing sensors on the joints of the constraining limb. The kinematic inverse solution equation and jacobian matrix for this 5-axis PMT are presented. Finally the workspace and the dexterity of the PMT are analyzed. Theoretical analysis is proved by the actual operation of the prototype of the 5-UPS/PRPU 5-axis PMT in our laboratory.

Topics: Machine tools
Commentary by Dr. Valentin Fuster
2004;():87-94. doi:10.1115/DETC2004-57030.

After a brief summary of the basic aspects of displacement-group theory, several applications of this theory to the kinematics of robot mechanisms are described. Then, we show in greater detail how to obtain, in a simple way, the characteristic polynomials of manipulators with special structural parameters. The goal is to find the polynomials of degree two for these special-geometry manipulators. In our paper, we apply displacement-group theory to provide an elegant and compact presentation and discussion of this subject, and we demonstrate that some geometrical conditions on the shapes of the manipulator links, as stated in other scientific papers, can be eliminated without changing the degree of the characteristic polynomials.

Commentary by Dr. Valentin Fuster
2004;():95-104. doi:10.1115/DETC2004-57031.

The authors present a new approach using genetic algorithms, neural networks and nanorobotics concepts applied to the problem of control design for nanoassembly automation and its application in medicine. As a practical approach to validate the proposed design, we have elaborated and simulated a virtual environment focused on control automation for nanorobotics teams that exhibit collective behavior. This collective behavior is a suitable way to perform a large range of tasks and positional assembly manipulation in a complex 3D workspace. We emphasize the application of such techniques as a feasible approach for the investigation of nanorobotics system design in nanomedicine. Theoretical and practical analyses of control modelling is one important aspect that will enable rapid development in the emerging field of nanotechnology.

Topics: Design
Commentary by Dr. Valentin Fuster
2004;():105-114. doi:10.1115/DETC2004-57035.

In field environments it is not usually possible to provide robots in advance with valid geometric models of its environment and task element locations. The robot or robot teams need to create and use these models to locate critical task elements by performing appropriate sensor based actions. Here, an information-based iterative algorithm to intelligently plan the robot’s visual exploration strategy is proposed to enable it to efficiently build 3D models of its environment and task elements. The method assumes mobile robot or vehicle with cameras carried by articulated mounts. The algorithm uses the measured scene information to find the next camera position based on expected new information content of that pose. This is achieved by utilizing a metric derived from Shannon’s information theory to determine optimal sensing poses for the agent(s) mapping a highly unstructured environment. Once an appropriate environment model has been built, the quality of the information content in the model is used to determine the constraint-based optimum view for task execution. Experimental demonstrations on a cooperative robot platform performing an assembly task in the field show the effectiveness of this algorithm for single and multiple cooperating robotic systems.

Topics: Robots
Commentary by Dr. Valentin Fuster
2004;():115-121. doi:10.1115/DETC2004-57036.

A wide family of parallel manipulators (PMs) is the one that groups all the PMs with three legs where the legs become kinematic chains constituted of a passive spherical pair (S) in series with either a passive prismatic pair (P) or a passive revolute pair (R) when the actuators are locked. The topologies of the structures generated by these manipulators, when the actuators are locked, are ten. One out of these topologies is the SR-2PS topology (one SR leg and two PS legs). This paper presents an algorithm that determines all the assembly modes of the structures with topology SR-2PS in analytical form. The presented algorithm can be applied without changes to solve, in analytical form, the direct position analysis of any parallel manipulator which generates a SR-2PS structure when the actuators are locked. In particular, the closure equations of a generic structure with topology SR-2PS are written. The eliminant of this system of equations is determined and the solution procedure is presented. Finally, the proposed procedure is applied to a real case. This work demonstrates that the solutions of the direct position analysis of any parallel manipulator which generates a SR-2PS structure when the actuators are locked are at most eight.

Commentary by Dr. Valentin Fuster
2004;():123-129. doi:10.1115/DETC2004-57037.

When the actuators are locked, parallel manipulators (PMs) become parallel structures, that are structures constituted by two rigid bodies (platform and base) connected by a number of kinematic chains (limbs) with only passive kinematic pairs. A set of PMs is the one collecting the manipulators (SP-2RS architectures) which become structures with one limb of type SP and two limbs of type RS (P, R and S stand for prismatic pair, revolute pair and spherical pair respectively). The analytic determination of the assembly modes of the SP-2RS structures (i.e. the solution in analytic form of the direct position analysis of the SP-2RS architectures) has not been presented in the literature yet. This paper presents the solution in analytic form of the DPA of the SP-2RS architectures. In particular, the closure equation system of a generic SP-2RS structure is written in the form of three non-linear equations in three unknowns. The solution of the non-linear system is reduced to the determination of the roots of a sixteenth-degree univariate polynomial equation plus a simple back substitution procedure. The proposed solution algorithm is applied to a real case. The result of this study is that the solutions of the direct position analysis of all the SP-2RS architectures are at most sixteen and can be analytically determined through the proposed algorithm.

Topics: Architecture
Commentary by Dr. Valentin Fuster
2004;():131-139. doi:10.1115/DETC2004-57039.

There exists a duality relationship between a twist and a wrench in the dynamics of a mechanism or a multi-body system. Using this relationship, the coordinate-free expressions for the dynamic analysis of multi-loop mechanisms is derived. In the analysis, a Jacobian matrix expressing the relationship among loops is defined, and it is clarified, by using this matrix, that both twists and wrenches can be easily analyzed based on the duality relationship among them. A parallel mechanism having connecting chains with loops is also analyzed, and it is shown that the same procedure can be applied to a parallel mechanism with connecting chains, each loop of which has a motion space that is different from that of the mechanism.

Commentary by Dr. Valentin Fuster
2004;():141-146. doi:10.1115/DETC2004-57046.

We present an experimental system for robot assisted, minimally invasive surgery that is capable of force measurement and haptic feedback. While minimally invasive surgery with robots provides several advantages like reduced tissue trauma and shorter recovery times, there are also some inherent short-comings. Referring to surgeons, the most significant issue is the lack of haptic feedback. This often causes collateral trauma and leads to prolonged operation time. Therefore we have developed an open platform, which combines the advantages of present systems with the possibility of force reflection. We have included features known from commercial available systems, like full Cartesian control of the end effector, stereo vision and ergonomic input devices. We used the system to perform basic surgical tasks (like knot-tying) and to evaluate force feedback.

Commentary by Dr. Valentin Fuster
2004;():147-153. doi:10.1115/DETC2004-57047.

The objective of this project is to develop a device to exchange interventional probes such as biopsy needles, local anesthesia, ablation tools, and lesion removal devices to perform multiple procedures for breast cancer diagnosis and treatment during single session in the MR scanner. Used as a supplement to an existing five degree-of-freedom (DOF) interventional probe positioning device, the apparatus is fitted with two additional DOF for the selection and exchange of interventional probes to be used in an automated probe positioning device. The entire system is constructed of MR compatible materials, i.e. non-magnetic and non-conductive, to eliminate artifacts and distortion of the MR images. The apparatus is remotely controlled by means of ultrasonic piezoelectric motors and a graphical user interface, providing MRI-guided planning and monitoring of the procedure while it is in progress. Based on a timing analysis, the device can quickly exchange a probe (48 seconds), thereby reducing the complexity and cost of the overall procedure.

Commentary by Dr. Valentin Fuster
2004;():155-161. doi:10.1115/DETC2004-57051.

In this paper an approximate kinematic synthesis method is presented with application to rigid-body guidance in planar multibody systems. The problem of finding the optimal dimensions in linkages with rigid-body guidance constraints has been widely studied. Many techniques have been developed and applied to numerous kinematic chains. However, some problems remain without appropriate solution, such as a large number of required poses or low computational cost. The proposed method uses exact-gradient determination to search for an optimal solution. The modelling of the mechanism uses fully Cartesian coordinates and is formulated by means of algebraic constraint equations. Furthermore, the formulation allows the use of a large number of prescribed poses giving high accuracy in the definition of synthesis conditions. Examples are included to illustrate the new approach to some synthesis specifications.

Commentary by Dr. Valentin Fuster
2004;():163-184. doi:10.1115/DETC2004-57064.

This contribution presents a more general mobility criterion applicable to parallel platforms, unlike previously employed mobility criteria based on the well-known Kutzbach-Grübler criterion that often fails to provide the correct number of degrees of freedom of parallel manipulators, the mobility criterion introduced in this contribution provides the correct number of degrees of freedom for a wider class of parallel manipulators. Furthermore, the analysis provides insight into why the criteria based on the Kutzbach-Grübler criterion often fails. Moreover, the newly developed criterion computes the passive degrees of freedom in parallel platforms and the mobility of some classes of kinematically defficient parallel platforms. Finally, it is important to note that this criterion is based on an analysis of the subalgebras of the Lie algebra, se(3), also known as screw algebra, of the Euclidean group, SE(3). As is common in many areas of kinematics, the criterion presented here can be also obtained by an analysis of the subgroups of the Euclidean group.

Commentary by Dr. Valentin Fuster
2004;():185-193. doi:10.1115/DETC2004-57065.

The instantaneous forward problem (IFP) singularities of a parallel manipulator (PM) must be determined during the manipulator design and avoided during the manipulator operation, because they are configurations where the end-effector pose (position and orientation) cannot be controlled by acting on the actuators any longer, and the internal loads of some links become infinite. When the actuators are locked, PMs become structures consisting of one rigid body (platform) connected to another rigid body (base) by means of a number of kinematic chains (legs). The geometries (singular geometries) of these structures where the platform can perform infinitesimal motion correspond to the IFP singularities of the PMs the structures derive from. In this paper, the singular geometries of the structures with topology SX-YS-ZS (S stands for spherical pair, whereas X, Y and Z stand for three generic one-dof pair which may be or may not be of the same type) are studied with a unified approach. The presented approach leads to obtain an analytic condition which allows all the singular geometries of these structures to be determined. Moreover, the geometric interpretation of the found singularity condition and the exhaustive enumeration of the types of singular geometries is provided. Finally, the use of the presented results in the design of the manipulators which become one structure with topology SX-YS-ZS when the actuators are locked is discussed.

Topics: Manipulators
Commentary by Dr. Valentin Fuster
2004;():195-203. doi:10.1115/DETC2004-57069.

This paper presents a planar parallel robot, built of spread-band elements. The robot has a workspace that exceeds its original size by far. The mass of the mechanical structure is extremely low, allowing accelerations up to 120 m/s2 . A spread-band element consists of a roll with flanges and a cambered metal tape. The arch form of the cambered band allows the transmission of tensile and compressive forces and of one moment. The flange of the roll changes the orientation of the bending radius axis by 90°. Nevertheless the combination of roll and cambered band retains the bands flexural rigidity against a moment around an axis perpendicular to the longitudinal axis of the cambered band and the axis of the roll. This leads to the possibility of transforming rotation into translation and vice versa using spread-band elements. The combination of several spread-band elements performed in the presented robot results in a rigid mechanical system.

Topics: Machinery , Robots , Design
Commentary by Dr. Valentin Fuster
2004;():205-213. doi:10.1115/DETC2004-57072.

Parallel mechanisms have drawn interest due to many advantages such as large payload ratio, high structural stiffness, low inertias and high dynamic capacities. But there are also drawbacks such as limited workspace and more singularities. Therefore many research efforts have been given to hybrid mechanisms, especially to hybrid serial-parallel mechanisms. This paper presents a new hybrid 2-DoF 7-bar linkage and several related linkages. Such mechanisms allow much larger range of movement than parallel mechanisms, yet have larger stiffness than serial mechanisms, and are desirable for robot fingers and limbs. Kinematic properties of the hybrid 2-DoF 7-bar planar linkage are analyzed. Comparative analyses between it and several 2-DoF linkages are given based on conditioning mapping, stiffness mapping, global conditioning index and global stiffness index. The results demonstrate the applicability of such linkages.

Topics: Linkages , Stiffness
Commentary by Dr. Valentin Fuster
2004;():215-222. doi:10.1115/DETC2004-57075.

Parallel manipulators are increasingly used in new applications by exploiting their better characteristics with respect to those of serial manipulators, such as higher stiffness, velocity and acceleration, payload. In this paper, experimental results are presented of a novel application of a 3-DOF CaPaMan (Cassino Parallel Manipulator) prototype to simulate point seismograms and 3D earthquake motion. The rigid body acceleration (linear acceleration, angular velocity and acceleration) has been experimentally analyzed to simulate real 3D earthquakes by using the parallel manipulator. Furthermore, first experimental results are reported to analyze earthquake effects on scaled civil structures.

Commentary by Dr. Valentin Fuster
2004;():223-229. doi:10.1115/DETC2004-57080.

To determine a suitable kinematic structure for a specific application is a complex task. Some available methods may provide a list of possible architectures for a robot wrist. However, a systematic method for selecting one among others is still an open problem. The paper presents a qualitative procedure for topological synthesis based on criteria that consider wrist design requirements and parallel mechanism peculiarities. The performance of chosen topology is evaluated according to described requirements.

Commentary by Dr. Valentin Fuster
2004;():231-240. doi:10.1115/DETC2004-57081.

The Traveling Chain Transmission (TCT) is a novel two-speed chain drive developed at the General Motors R&D Center [1]. In order to predict potential performance and develop design requirements, a multi-body dynamic model of the Traveling Chain Transmission (TCT) was developed and validated using dynamometer test data from the first TCT prototype hardware. A vehicle drivetrain model incorporating the TCT as part of a multi-speed transmission was also developed in order to study the potential performance of the TCT in a vehicle environment. The simulated transmission is a nominal 4-speed automatic transmission converted to a 5-speed transmission by replacing the fixed speed-ratio drop chain by the dual speed-ratio TCT. An upshift from 4th to 5th gear is simulated and the resulting torque and speed transients are predicted.

Commentary by Dr. Valentin Fuster
2004;():241-249. doi:10.1115/DETC2004-57092.

This paper focusses on reducing the dynamic reactions (shaking force, shaking moment and driving torque) of plane, crank-rocker four-bars through counterweight addition. Determining the mass parameters of the counterweights constitutes an optimization problem, which is classically considered to be nonlinear and hence difficult to solve. A first contribution of this paper is the proof that this optimization problem can be reformulated as a convex program, that is, a nonlinear optimization problem that still has a unique (and hence guaranteed global) optimum, which can be found with great efficiency. Because of the unique features of this formulation, it becomes possible to investigate (and by the guarantee of obtaining a global optimum, in fact prove) the ultimate limits of dynamic balancing, in a reasonable amount of time. When applied to a particular example, this results in design charts, which clearly illustrate (i) the tradeoff between minimizing the different dynamic reactions, and (ii) the fact that adding counterweights is effective, but at the cost of a significant amount of added mass. These design charts constitute a second contribution of the present work.

Commentary by Dr. Valentin Fuster
2004;():251-255. doi:10.1115/DETC2004-57093.

On the basis of first-order and second-order kinematic influence coefficient matrices, dynamics characteristics indices for robot manipulator are presented in the paper. Different from indices before, these indices include not only the first-order kinematics influence coefficient matrix G, but also the second-order kinematic influence coefficient matrix H. Then with the global index, these indices can be used to guide the dynamics design.

Commentary by Dr. Valentin Fuster
2004;():257-265. doi:10.1115/DETC2004-57100.

This paper proposes a new family of four degrees-of-freedom (dof) parallel mechanisms with two platforms and its application to a footpad device that can simulate the spatial motions of the human foot. The new mechanism consists of front and rear platforms, and three limbs. Two limbs with 6-dof serial joints (P -S-P-P) are attached to each platform and are perpendicular to the base plate, while the middle limb (P e -R e -R or P e -P e -R) is attached to the revolute joint that connects the front and rear platforms. The middle limb is driven by the 2-dof driving mechanism that is equivalent active serial prismatic and revolute joints (P e -R e ), or prismatic and prismatic joints (P e -R e ) with two base-fixed prismatic actuators. Therefore, two new 4-dof parallel mechanisms with two platforms can generate pitch motion of each platform, and roll and heave motions (1T-3R) or pitch motion of each platform and two translational motions (2T-2R) at both platforms. Kinematic analyses of the 1T-3R mechanism were performed, including inverse and forward kinematics, and velocity analysis. Based on the 1T-3R mechanism, a footpad device was designed to generate foot trajectories for natural walking. Finally, simulations of the foot trajectories in the normal gait cycle were performed using the proposed footpad device.

Commentary by Dr. Valentin Fuster
2004;():267-271. doi:10.1115/DETC2004-57103.

The ability to move at reasonable ease in all directions is an important requirement in the design of manipulators. The degree of ease of mobility varies from point to point in the workspace of the manipulator’s end effector. Maximum ease of mobility is obtained at an isotropic point, and the minimum occurs at singularities. An attempt has been made here to use a geometric approach for determining the isotropic points in the workspace of planar 5-bar linkages. The geometrical approach leads to interesting observations on the location of isotropic points in the workspace. The procedure also yields a technique for the synthesis of 5-bar linkages and associated coupler points exhibiting isotropic behaviour. Additionally it has been shown that coupler points exhibiting isotropic mobility occur in pairs.

Topics: Linkages , Isotropy
Commentary by Dr. Valentin Fuster
2004;():273-280. doi:10.1115/DETC2004-57108.

Design and analysis of a cable actuated hand prosthesis is presented. The hand has multi-actuated fingers, four with two joints, and the thumb with three joints. Each joint is designed using a novel flexible mechanism based on the loading of a compression spring in both transverse and axial directions and using cable-conduit systems. The rotational motion is transformed to tendon-like behavior, which enables the location of the actuators to be far from the arm (e.g., on a belt around the waist). It is shown that the solution of the transverse deflection of each finger segment is obtained in a general form through a Haringx model followed by an element stiffness model. Also, we use a FEM model (ANSYS) to visualize the deflection of one finger. A prototypic finger is experimentally tested, results are verified, and the hand prosthesis is built.

Topics: Cables , Design , Prostheses
Commentary by Dr. Valentin Fuster
2004;():281-290. doi:10.1115/DETC2004-57113.

We present the design and gait generation for an experimental ROLLERBLADER. The ROLLERBLADER is a robot with a central platform mounted on omnidirectional casters and two 3 degree of freedom legs. A set of passive rollerblading wheels is attached to the end of each leg. The wheels give rise to nonholonomic constraints acting on the robot. The legs can be picked up and placed back on the ground allowing a combination of skating and walking gaits. We present two types of gaits for the robot. In the first gait, the legs are picked up and placed back on the ground. In the second gait, the wheels stay on the ground at all times. Experimental gait results for a prototype robot are also presented.

Topics: Robots , Simulation , Design
Commentary by Dr. Valentin Fuster
2004;():291-298. doi:10.1115/DETC2004-57114.

The gear-teeth number synthesis of an automatic planetary transmission used in automobiles is formulated as a constrained optimization problem that is solved with the aid of an Estimation of Distribution Algorithm. The design parameters are the teeth number of each gear, the number of multiple planets and gear module, while the objective function is defined based on the departure between the imposed and the actual gear ratios, constrained by teeth-undercut avoidance, limiting the maximum overall diameter of the transmission and ensuring proper planet spacing.

Commentary by Dr. Valentin Fuster
2004;():299-306. doi:10.1115/DETC2004-57117.

A time-accurate explicit finite element code is used to predict the natural frequencies of a typical class of flexible multibody systems — automotive accessory belt-drives. The system considered consists of a belt, two pulleys, and a tensioner. Two techniques are used to find the system natural frequencies: (a) applying a sharp impulse to the system and extracting the system natural frequencies from the resulting displacement/strain time-histories via an FFT; and (b) applying a harmonic force to the system and sweeping through a frequency range, while at the same time, monitoring for large system response. In the present paper a comparison between these two techniques is presented for a typical accessory drive. Also, recommendations are offered on how to best use each technique to efficiently extract the system’s natural frequencies.

Commentary by Dr. Valentin Fuster
2004;():307-315. doi:10.1115/DETC2004-57124.

This paper presents graphical techniques to locate the unknown instantaneous centers of zero velocity of planar, single-degree-of-freedom, linkages with kinematic indeterminacy. The approach is to convert a single-degree-of-freedom indeterminate linkage into a two-degree-of-freedom linkage. Two methods are presented to perform this conversion. The first method is to remove a binary link and the second method is to replace a single link with a pair of links connected by a revolute joint. First, the paper shows that a secondary instantaneous center of a two-degree-of-freedom linkage must lie on a unique straight line. Then this property is used to locate a secondary instant center of the single-degree-of-freedom linkage at the intersection of two lines. The two lines are obtained from a purely graphical procedure. The graphical techniques presented in this paper are illustrated by three examples of single-degree-of-freedom linkages with kinematic indeterminacy. The examples are a ten-bar linkage with only revolute joints, the single flier eight-bar linkage, and a ten-bar linkage with revolute and prismatic joints.

Topics: Linkages
Commentary by Dr. Valentin Fuster
2004;():317-325. doi:10.1115/DETC2004-57125.

This paper presents a new cable-suspended robot. It is a 7-cable spatial design with a closed-form forward pose kinematics solution. Applications include automated machining, construction, and sculpting. This paper presents two new ideas. First, an independent, passive, six-string-pot-based Cartesian metrology system was built and tested since length measurements through the active drive system may be inaccurate for large-scale systems. Second, we introduce a new active cable tensioning approach wherein we control the displacement of a physical spring in-line with one of the active drive cables in attempt to ensure only positive cable tensions in all active cables. This is compared with an existing particular/homogeneous tensioning solution approach. Simulation examples are presented in this paper; we are currently building the proposed system for future evaluation work.

Topics: Robots , Cables , Metrology
Commentary by Dr. Valentin Fuster
2004;():327-335. doi:10.1115/DETC2004-57126.

Research into robotic grasping and manipulation has led to the development of a large number of tendon based end effectors. Many are, however, developed as a research tool, which are limited in application to the laboratory environment. The main reason being that the designs requiring a large number of actuators to be controlled. Due to the space and safety requirements, very few have been developed and commissioned for industrial applications. This paper presents design of a rigid link finger operated by a minimum number of actuators, which may be suitable for a number of adaptive end effectors. The adaptive nature built into the end effector (due to limited number of actuators) presents considerable problems in grasping and control. The paper discusses the issues associated with such designs. The research can be applicable to any adaptive end effectors that are controlled by limited number of actuators and evaluates their suitability in industrial environment.

Commentary by Dr. Valentin Fuster
2004;():337-346. doi:10.1115/DETC2004-57127.

This paper presents a detailed analysis of the constant-orientation wrench-closure workspace of planar three-degree-of-freedom parallel mechanisms driven by four cables. The constant-orientation wrench-closure workspace is defined as the subset of the plane wherein, for a given orientation of the moving platform, any planar wrench applied on the moving platform can be balanced by the cable-driven mechanism. Based on mathematical observations, this workspace is proved to be the union of two disconnected sets that may or may not exist. Moreover, if the constant-orientation wrench-closure workspace (WCW) exists, its boundary is shown to be composed of portions of conic sections. Then, an algorithm that determines the constant-orientation wrench-closure workspace by means of a graphical representation of its boundary is introduced. Several examples are also included.

Topics: Cables , Mechanisms
Commentary by Dr. Valentin Fuster
2004;():347-355. doi:10.1115/DETC2004-57129.

Tools that make it possible to measure the performances of manipulators are essential in many technical applications, for instance, when the optimal path to accomplish a task has to be chosen, or when different manipulator architectures have to be compared. This paper proposes new indices that fully describe the passive dynamic performances of manipulators with two degrees of freedom (dof). The proposed indices make it possible to compare the passive dynamic performances of different manipulator architectures, which can perform the same tasks, and can be used to build diagrams which highlight the effects of variations in the manipulator geometry on the manipulator dynamics. These features make them easy to be used in a design context. Finally, one application of the proposed indices will be presented and discussed.

Commentary by Dr. Valentin Fuster
2004;():357-365. doi:10.1115/DETC2004-57131.

A new control method, called adaptive nonlinear PD learning control (NPD-LC), is proposed for robot manipulator applications in this paper. The proposed control structure is a combination of a nonlinear PD control structure and a directly learning structure. Consequently, this new control method possesses both adaptive and on-line learning properties. One of the unique features of the NPD-LC algorithm is that the learning is based on the previous torque profile of the repetitive task. It is proved that the NPD-LC enjoys the asymptotic convergence for both tracking positions and tracking velocities. Simulation studies were conducted by comparing the proposed method with many other existing methods. As a result, it was demonstrated that the NPD-LC method can achieve a faster convergence speed. The proposed NPD-LC is robust and can be implemented for the control of robot manipulators.

Topics: Manipulators
Commentary by Dr. Valentin Fuster
2004;():367-376. doi:10.1115/DETC2004-57135.

The determination of the 6-D singularity locus of the general Gough-Stewart platform is discussed in this article. The derivation of the velocity equation and the corresponding Jacobian matrices is first presented. Then a new procedure is introduced to obtain the analytical expression of the singularity locus, which is a function of six variables (x, y, z, φ, θ, ψ), using the velocity equation. Examples are also given to illustrate the results obtained. Gough-Stewart platforms can be used in several robotic applications as well as in flight simulators. The determination of the singularity locus is a very important design and application issue.

Commentary by Dr. Valentin Fuster
2004;():377-386. doi:10.1115/DETC2004-57137.

Performing complex tasks in Minimally Invasive Surgery (MIS) is demanding due to a disturbed hand-eye co-ordination, the use of non-ergonomic instruments with limited degrees of freedom (DOFs) and a lack of force feedback. Robotic telemanipulatory systems enhance surgical dexterity by providing up to 7 DOFs. They allow the surgeon to operate in an ergonomically favorable position with more intuitive manipulation of the instruments. Commercially available robotic systems, however, are very bulky, expensive and do not provide any force feedback. The aim of our study was to develop a simple mechanical manipulator for MIS. When manipulating the handle of the device, the surgeon’s wrist and grasping movements are directly transmitted to the deflectable instrument tip in 7 DOFs. The manipulator consists of a parallelogram mechanism with steel wires. First phantom experience indicated that the system functions properly. The MIM provides some force feedback improving safety. A set of MIMs seems to be an economical and compact alternative for robotic systems.

Commentary by Dr. Valentin Fuster
2004;():387-395. doi:10.1115/DETC2004-57140.

Because wear, clearance and backlash can be reduced or eliminated, one of the important advantages of compliant mechanisms is their potential to increase the mechanism precision dramatically. In this paper, a spatial compliant mechanism based on the 3-PRRR parallel mechanism driven by linear motors is presented. The design methodology is described and the kinematic and static models are developed. Using the “transmission ratio”, an effective and simple compensation method is designed. The performances of this mechanism are studied with the help of the structure simulation module in Pro-Mechanica. Using an open-loop control with compensation, measurements over a 20×20×20 mm3 workspace show that the position accuracy is smaller than 1 μm, i.e., 0.05% of the full range of motion and that the orientation error is less than 1 μradian.

Commentary by Dr. Valentin Fuster
2004;():397-405. doi:10.1115/DETC2004-57148.

Due to its complex shape and its tightness, it is difficult to prepare the subtalar joint for fusion by means of a minimally invasive approach and conventional straight instruments. The preparation implies the establishment of bleeding contact surfaces. It is desirable to preserve the joint’s shape to keep congruent and smooth contact surfaces for optimal fusion. A compliant instrument was designed to facilitate the preparation by starting the design process from a clinically driven approach. The concept consists of a drill/mill unit at the tip that is steered through the joint by means of a passive automatic control. Since the joint will be fused, the cartilage and the subchondral bone layers should intentionally be damaged. This allows the use of the subtalar joint surfaces themselves as a guiding tunnel to preserve the joint’s shape. Thereto, a guidance frame was placed at the tip. The instrument is equipped with a special shaft that is compliant in one direction (perpendicular to the joint surfaces), and stiff in the two other directions to resist and transmit machining forces. The dimensioning of the instrument was performed by taking into account the specific shape of the subtalar joint. A prototype of the instrument was manufactured. The prototype is powered by a commonly used shaver system. The compliant instrument was tested in cadaver material, and gives promising results. In practice, the difference of the stiffness in y- and z-direction is at least a factor 100.

Commentary by Dr. Valentin Fuster
2004;():407-413. doi:10.1115/DETC2004-57161.

This paper shows how the instantaneous invariants for time-independent motions can be obtained from time-dependent motions. Relationships are derived between those parameters that define a time-dependent motion and the parameters that define its geometrically equivalent time-independent motion. The time-independent formulations have the advantage of being simpler than the time dependent ones, and thereby lead to more elegant and parsimonious descriptions of motions properties. The paper starts with a review of the choice of canonical coordinate systems and instantaneous invariants for time-based spherical and spatial motions. It then shows how to convert these descriptions to time-independent motions with the same geometric trajectories. New equations are given that allow the computation of the geometric invariants from time-based invariants. The paper concludes with a detailed example of the third-order motion analysis of the trajectories of an open, spatial R-R chain.

Topics: Motion
Commentary by Dr. Valentin Fuster
2004;():415-424. doi:10.1115/DETC2004-57164.

Very often, spring-to-gravity-balancing mechanisms are conceived with ideal (zero-free-length l0 =0) springs. However, the use of ideal springs in the conception phase tends to lead to more complex mechanisms because the ideal spring functionality has to be approximated with normal springs. To facilitate construction of (gravity) balancers, employing normal springs (l00) directly mounted between the link attachment points of the mechanism in the conception phase therefore seems beneficiary. This paper discusses spring mechanisms that enable perfect balancing of gravity acting on an inverted pendulum while employing normal springs between the spring-attachment points: The design synthesis of such mechanisms will be explained and balancing conditions will be derived, using a potential energy consideration.

Commentary by Dr. Valentin Fuster
2004;():425-432. doi:10.1115/DETC2004-57166.

This paper discusses mechanisms that allow for perfect static balancing of rotations about a fixed spherical joint by means of ideal springs. Using a potential energy consideration, balancing conditions of a spatial three-spring balancer will be derived. It will be shown that not satisfying these conditions causes non-constant terms in the potential energy expression of the spring-mechanism, which can be eliminated by coupling the spring-mechanism to an inverted pendulum.

Commentary by Dr. Valentin Fuster
2004;():433-442. doi:10.1115/DETC2004-57168.

The design of a novel prismatic drive is reported in this paper. This transmission is based on Slide-O-Cam, a cam mechanism with multiple rollers mounted on a common translating follower. The design of Slide-O-Cam was reported elsewhere. This drive thus provides pure-rolling motion, thereby reducing the friction of rack-and-pinions and linear drives. Such properties can be used to design new transmissions for parallel-kinematics machines. In this paper, this transmission is optimized to replace ball-screws in Orthoglide, a three-DOF parallel robot optimized for machining applications.

Commentary by Dr. Valentin Fuster
2004;():443-449. doi:10.1115/DETC2004-57170.

The purpose of this project is to develop a mobile robot for hazardous terrain exploration. The exploration of hazardous terrain requires the development of a passive mechanism adaptable to such terrain and a sensing system for obstacle avoidance, as well as a remote control. We designed a new mobile robot, the Ronahz 6-wheel robot, which uses a passive mechanism that can adapt to hazardous terrains and building stairways without any active control. The suggested passive linkage mechanism consists of a simple four-bar linkage mechanism. In addition, we install a stereo vision system for obstacle avoidance, as well as a remote control. Wide dynamic range CCD cameras are used for outdoor navigation. A stereo vision system commonly requires high computational power. Therefore, we use a new high-speed stereo correspondence algorithm, triangulation, and iterative closest point (ICP) registration to reduce computation time. Disparity maps computed by a newly proposed, high-speed method are sent to the operator by a wireless LAN equipment. At the remote control site, a three-dimensional digital map around a mobile robot is built by ICP registration and reconstruction process, and this three-dimensional map is displayed for the operator. This process allows the operator to sense the environment around the robot and to give commands to the mobile robot when the robot is in a remote site.

Commentary by Dr. Valentin Fuster
2004;():451-460. doi:10.1115/DETC2004-57171.

This paper describes the design and implementation of a testbed for facilitating the study of human-robot interactions (HRI). HRI has long been a part of robotics research, where humans were typically required to guide the robot task in progress and to ensure safe operation. The current state of human interaction with robots, versus simple “machines” (e.g. in manufacturing automation) is quite different. This called for the need to look into different interaction roles between humans and robots. Robots differ from simple machines in that they are mobile, some may be autonomous and hence not as predictable in their actions. To facilitate the research in this domain, the aim is to develop an easy to use and safe front-end human-robot system for human users to interact with physical mobile robots. This testbed provides different types of system configurations (i.e. one human to one robot, one human to multiple robots, etc.) and interfaces for conducting experiments under different HRI scenarios.

Topics: Robots
Commentary by Dr. Valentin Fuster
2004;():461-468. doi:10.1115/DETC2004-57172.

This study presents an approach, the unit circle (UC), to bridge numerical data from each variable of robotic systems and its corresponding qualitative state so that current qualitative methods (e.g. fuzzy models) can apply to general robotic systems. A manipulator is described as a collection of constraints holding among time-varying, interval-valued parameters. The UC representation is presented, and the continuous motion of the end-effector is evaluated by the change of directions of qualitative angle and qualitative length. Analytical formulas of qualitative velocity and qualitative acceleration are derived. The characteristic mapping is introduced for fault detection and diagnosis in terms of the UC. In the end simulation results demonstrate the feasibility of the UC approach for fault diagnosis. The UC representation of robots concerns a global assessment of the systems behaviour, and it might be used for the purpose of monitoring, diagnosis, and explanation of physical systems. This is the first step to fault diagnosis and remediation for robots (e.g. Beagle 2) using qualitative methods.

Topics: Robotics
Commentary by Dr. Valentin Fuster
2004;():469-487. doi:10.1115/DETC2004-57177.

It is important to conduct error analysis of assemblies in order to ensure that the manufactured parts satisfy the design specifications. Traditionally, mechanical tolerances and clearances are modeled as random variables and the analysis is conducted using probabilistic methods. In this work, a new approach, based on interval analysis, is presented for the modeling and analysis of tolerances and clearances. The basic procedure of interval analysis involving solution of simultaneous nonlinear equations is described. The application of the approach in the fuzzy error analysis of planar mechanisms is outlined. The treatment of the tolerances and clearances of the mechanism as interval numbers leads to a better and a more realistic estimation of the analysis results. Numerical results are presented to illustrate the computational procedure. The results of the interval analysis, although philosophically different, are compared with those given by the probabilistic method for comparable input data; the differences found in the two sets of results are explained in terms of the basic characteristics of the two methods. This work denotes the first application of interval methods for the modeling of tolerances and clearances and the fuzzy error analysis of mechanisms.

Topics: Modeling , Mechanisms
Commentary by Dr. Valentin Fuster
2004;():489-512. doi:10.1115/DETC2004-57181.

This paper reformulates and extends the new, group theoretic, mobility criterion recently developed by the authors, Rico and Ravani [1]. In contrast to the Kutzbach-Grübler criterion, the new mobility criterion, and the approach presented apply to a large class of overconstrained linkages. The criterion is reformulated, in terms of the well known Jacobian matrices, for exceptional linkages; it is extended to linkages with partitioned mobility as well as trivial linkages. In addition, an extension of the criterion is presented that would allow the computation of degrees-of-freedom of several cases of paradoxical linkages. The case of classical paradoxical linkages such as the Bennett and Goldberg linkages still remains unsolved but some insight into the application of the new mobility criterion for these linkages is also presented.

Commentary by Dr. Valentin Fuster
2004;():513-521. doi:10.1115/DETC2004-57182.

This paper describes a method for finding the location of a rigid body such that N specified points of the body lie on N given planes in space. Of special interest is the case N = 6, which is the minimum number to fully constrain the body. This geometric problem arises in two seemingly disparate contexts: metrology, as a generalization of so-called “3-2-1” locating schemes; and robotics, as the forward kinematics problem for 6ES or 6SE parallel-link platform robots. For N = 6, the geometric problem can be formulated algebraically as 3 quadratic equations having, in general, eight possible solutions. We give a method for finding all eight solutions via an 8 × 8 eigenvalue problem. We also show that for N ≥ 7, the solution can be found uniquely as a linear least squares problem.

Commentary by Dr. Valentin Fuster
2004;():523-533. doi:10.1115/DETC2004-57188.

The stability of a rigid body on which two forces are in equilibrium can be assessed intuitively. In more complex cases this is no longer true. This paper presents a general method to assess the stability of complex force systems, based on the notion of dynamic equivalence. A resultant force is considered dynamically equivalent to a given system of forces acting on a rigid body if the contributions to the stability of the body of both force systems are equal. It is shown that the dynamically equivalent resultant force of two given constant forces applies at the intersection of its line of action and the circle put up by the application points of the given forces and the intersection of their lines of action. The determination of the combined center of mass can be considered as a special case of this theorem. Two examples are provided that illustrate the significance of the proposed method. The first example considers the suspension of a body, by springs only, that is statically balanced for rotation about a virtual stationary point. The second example treats the roll stability of a ship, where the metacentric height is determined in a natural way.

Topics: Force , Stability
Commentary by Dr. Valentin Fuster
2004;():535-540. doi:10.1115/DETC2004-57189.

A new hybrid parallel platform device that is based on tensegrity is introduced in this paper. A tensegrity structure is one that is comprised of members that are either in tension (ties) or compression (struts). The device studied in this paper replaces the ‘upper’ and ‘lower’ set of ties of a 3-strut tensesgrity system with rigid bodies. Further, the three struts are replaced by three leg connectors whose lengths can be changed via prismatic actuators. The three remaining ties are replaced by the series combination of a spring and a non-compliant tie where the length of the non-compliant tie can be controlled. An analysis is presented that shows how the connector leg lengths and non-compliant tie lengths can be determined so as to position and orient the upper platform at a desired pose and at a desired total potential energy level. It is the control of the potential energy in the system that makes this new hybrid parallel-platform unique.

Commentary by Dr. Valentin Fuster
2004;():541-548. doi:10.1115/DETC2004-57190.

This article presents an overview of the mechanical design features, fabrication and control of a Rescue Robot (CEDRA) for operation in unstructured environments. As a preliminary step, the essential characteristics of a robot in damaged and unstable situations have been established. According to these features and kinematical equations of the robot, design parameters are optimized by means of Genetic Algorithm. Optimum parameters are then utilized in construction. Upon fabrication, this unit has been tested in clean laboratory environment, as well as, ill-conditioned arenas similar to earthquake zones. The obtained results have been satisfactory in all aspects, and improvements are currently underway to enhance capabilities of the rescue robot unit for various applications.

Commentary by Dr. Valentin Fuster
2004;():549-559. doi:10.1115/DETC2004-57192.

This paper presents a novel conceptional design of a prosthetic finger and its optimization. The finger consists of two phalanges, interconnected by rolling contact joints. A novel conceptual design of the rolling joints is presented that combines the advantages of two previously reported versions, i.e. minor dimensional errors do not result in complete loss of preload and associated backlash, while they guarantee precise positioning as long as the joint load is below the preload. The two phalanges are driven by a single tendon. Consequently, this concept is adaptive, yet underactuated (less degrees of actuation than degrees of freedom). A multibody model of the finger was made, based on a general dynamic model. This paper focusses on the static model that was derived from the general model in order to optimize the dimensional design for uniform force distribution, regardless of size and shape of the object. For the optimization, a genetic algorithm was used. A dimensional design was found that features a standard deviation of the ratio of pinch force to operating force of only 1.2 percent over a range of motion of 60 degrees for each joint.

Commentary by Dr. Valentin Fuster
2004;():561-570. doi:10.1115/DETC2004-57193.

From the standpoint of kinematics, we present a type of three-dof pure spatial translational parallel mechanism with 3-PRPaR topology as an alternative to design automation devices and a regional structure of a six-dof hybrid parallel platform. First, we describe the structural properties of mechanism and analyze its kinematic mobility. It is verified that a pure translational motion does exist through the coordinate transformation technique and the well-known D-H parametric notations. Then, we proceed with the forward and inverse kinematic analysis and derive their analytical closed-form solutions by the matrix algebra method. For the confirmation of the derived equations, some numerical examples are also taken. Furthermore, with the help of the forward kinematics, we derive the workspace in the analytical form. Finally, taking account of the overall Jacobian matrix provides the condition number and the identification of singular configuration is explored based on direct and inverse kinematics Jacobian matrix.

Commentary by Dr. Valentin Fuster
2004;():571-575. doi:10.1115/DETC2004-57196.

This paper introduces a link-type tracked mobile robot developed for demining operations. The robot consists of three parts: a front frame, a rear frame, and a body. The front frame is connected to the rear frame by a rotational passive adaptation mechanism, which is a driving mechanism of the robot. This passive adaptation mechanism enables the proposed robot good adaptability to uneven terrain including stairs. The link structure gives rise to a small, simple, and energy efficient vehicle. In addition, the new demining system adaptable to the proposed tracked mobile robot is discussed on how to clear small anti-personnel mines with a non-explosive method. Both the motion of the demining rake and the design parameters of the demining unit are analyzed. Finally, a presentation follows on how the new demining system can unearth small anti-personnel mines with a non-explosive method.

Topics: Mobile robots
Commentary by Dr. Valentin Fuster
2004;():577-587. doi:10.1115/DETC2004-57206.

This work has established a model for the behavior of the Floating-Opposing-Arm FOA centrifugal clutch and used it to determine the most sensitive design variables, to minimize performance variations, and to predict the behavior of stacked layers. By manufacturing the compliant clutch in multiple layers, not only is it feasible to produce these clutches in high volumes, but the variations in engagement and torque performance decreased, therefore allowing the clutch to perform more consistently. In addition, the FOA design accounts for performance sensitivity to variations in both design parameters and manufacturing processes and minimizes such variations. FOA clutch testing showed that the compliant clutch is comparable to the benchmark clutch in both engagement speed and torque capacity.

Commentary by Dr. Valentin Fuster
2004;():589-594. doi:10.1115/DETC2004-57215.

The PumaPaint site has been allowing users to create paintings remotely over the Internet since 1998. Although this site allows for some artistic creativity, the task is inherently two-dimensional and lacks any real manipulation capability by the remote user. On a similar theme but with much higher complexity, we are creating a robotic device that will allow true, three-dimensional manipulation by the remote user with an added capacity for force control and kinesthetic feedback. The robotic device will be a roughly anthropomorphic pair of limited degree-of-freedom mechanical hands arranged in mutual opposition. We are designing these hands by building parametric models using AutoDesk Inventor® and then fabricating the components in our sparsely equipped machine shop. This paper will preset the design concept and details of the modeling process, plus how this process is informed by our fabrication capacity. The paper will also present the partially completed robotic device and discuss remaining obstacles to completion.

Topics: Robotics , Internet
Commentary by Dr. Valentin Fuster
2004;():595-601. doi:10.1115/DETC2004-57220.

Contact dynamics modeling remains an intensive area of research with new applications emerging in robotics, biomechanics and multibody dynamics areas. Many formulations for contact dynamics problem have been proposed. The two most prominent categories include the discrete approach, which employs the impulse-momentum relations, and the continuous approach, which requires integration of dynamics equations through the contact phase. A number of methods in the latter category are based on an explicit compliant model for the contact force. One such model was developed by Hunt and Crossley three decades ago who introduced a nonlinear damping term of the form λxn ẋ into the contact force model. In addition to proposing the general form of this damping component of the contact force, Hunt and Crossley derived a simple expression for relating the damping coefficient λ to the coefficient of restitution e. This model gained considerable popularity due to its simplicity and realistic physics. It also spurred new research in the area, specifically on how to evaluate the damping coefficient λ. Subsequently, several authors put forward different approximations for λ, however, without clearly revealing the range of validity of their simplifying assumptions or the accuracy limitations of the resulting contact force models. The authors of this paper analyze the various approaches employed to derive the damping coefficient. We also evaluate and compare performance of the corresponding models by using a meaningful measure for their accuracy. A new derivation is proposed to calculate more precisely the damping coefficient for the nonlinear complaint contact model. Numerical results comparing all models are presented for a sphere dropping on a stationary surface.

Topics: Force , Modeling
Commentary by Dr. Valentin Fuster
2004;():603-610. doi:10.1115/DETC2004-57230.

In this paper, a novel six degrees-of-freedom (6-DOF) parallel manipulator actuated by three base-mounted partial spherical actuators is proposed. The parallel manipulator consists of a base, a moving platform and three connecting legs. Each leg has spherical (S), prismatic (P) and universal (U) joints (SPU) in serial manner. The spherical joints are partially actuated due to the fact that the actuators of each leg are used only to specify its leg’s direction. The inverse and forward pose kinematics as well as the singularity points of the aforementioned mechanism is described in the article. In the inverse pose kinematics, active joint variables could be calculated with no need for the evaluation of passive joint variables. It will be shown that the inverse pose kinematics has sixty-four (64) solutions (64 different configurations exists for the inverse pose problem). In the forward pose kinematics, instead of twelve nonlinear equations derived by equaling the transformation matrices of each leg through Denavit-Hartenberg notation, only three nonlinear equations with less nonlinearity could be solved via numerical method, and therefore the numerical method converges more rapidly to the answer. Finally two different sets of singularity points with different natures are obtained.

Commentary by Dr. Valentin Fuster
2004;():611-620. doi:10.1115/DETC2004-57232.

Mandibular distraction osteogenesis is a medical procedure for lengthening the mandible bone by stimulating natural bone-healing mechanisms via a mechanical device that exerts a force on the mandible in one or more directions. Many mandibular distraction devices must be placed externally and most rely on the patient to manually actuate the device each day. This project focuses on the design of an automatically actuated, single degree-of-freedom, implantable distraction device that would be minimally visible after installation. Such a device could reduce errors from patient compliance and would be an important first step toward increasing the capability of future devices. A simple motor and leadscrew system was used with a custom designed impact transmission and controller. A test was conducted on a prototype to determine the feasibility of the design and measure the overall system efficiency. The device was able to move the required 70 N load at a rate of about 1 mm per minute. Compared to an equivalent device utilizing a planetary gearhead to amplify the torque, the impact coupling is significantly less efficient. However, the necessary increase in battery size has only a small impact on the total device length. For a system with the same motor and force output, the impact coupling system is shorter than the gearhead-based system due to a 50% reduction in transmission length.

Topics: Design , Testing
Commentary by Dr. Valentin Fuster
2004;():621-626. doi:10.1115/DETC2004-57234.

Most fixed automations in traditional manufacturing systems are not equipped to manage product variations efficiently. This paper presents a design and configuration for a machine-vision-equipped robotic packing cell that is capable of managing a wide range of product sizes. Product size information is gathered at an earlier stage in the manufacturing process and then transferred electronically to the robot cell. Different controllers are needed to manage robot cell functions related to incoming product, machine vision, robot control, robot manipulator, and multiple layers of safety control information. Each controller, due to the nature of its function, has different advantages in processing different data types. In order to achieve the highest possible robot manipulator utilization rate, the assignment of information processing among controllers needs to be thoughtfully planned, especially for the critical mathematical routines. Coordination and calibration between charged-coupled device (CCD) cameras and robots in existing manufacturing facilities are configured with considerations for building vibrations, lighting conditions, and signal processing assignments among the available devices. System efficiency is improved when the vision signal, robot logical signal, and robot manipulator signal processing units are running cohesively in parallel. The capability of the machine-vision-assisted robot end effector automatic path adjustment, to pick up and pack different sizes of products dynamically, allows a higher level of flexibility and efficiency. This paper describes a feasible design and configuration for an integrated machine vision robotic cell in a manufacturing system.

Commentary by Dr. Valentin Fuster
2004;():627-634. doi:10.1115/DETC2004-57236.

The mobility is analyzed for the 4R1P-type planar closed-loop five-bar linkages using characteristics charts and four-bar Grashof criterion by introducing the virtual link. Based on dimensional relations of the three RR-type links, the five-bars are divided into three categories: Class I, Class II and Class III. Class I satisfies rmax < rmid + rmin , Class II satisfies rmax > rmid + rmin and Class III holds the equality. In the characteristics charts, Class I has a circular slide reachable range with two Grashof regions and two non-Grashof ones separated by three Grashof boundaries C1 , C2 and C3 , as well as one structure boundary B3 ; Class II has a ringed range with one Grashof region and two non-Grashof ones separated by the void boundary A1 instead of C1 , two Grashof boundaries C2 and C3 , as well as B3 ; and Class III has a circular range with three regions separated by C2 , C3 and B3 . From the four-bar Grashof criterion, the five-bars have two fully revolute joints at the Grashof regions while no such joints at the non-Grashof regions. Specially, the five-bars have Grashof change points at the Grashof boundaries and keep in line at A1 and B3 . Each class includes further sub-classes. Several cases are illustrated for the applications.

Topics: Linkages
Commentary by Dr. Valentin Fuster
2004;():635-643. doi:10.1115/DETC2004-57242.

Compliant mechanisms have many advantages over their rigid-body counterparts. One disadvantage however is the fact that motion of the mechanism is associated with elastic energy storage in the compliant parts. This is a problem especially in cases where accurate force transmission is of primary concern, such as in medical graspers. A solution to this problem is to statically balance the elastic forces by the addition of a spring force compensation mechanism, such that the effect of the compliance is neutralized. The complete resulting mechanisms resulting from this concept are called statically balanced compliant mechanisms (SBCMs). This paper presents a feasibility study into the design of a grasper for medical purposes and demonstrates that the concept is possible and practically viable. It is shown that the compliant gripper of a laparoscopic forceps can be statically balanced with a single-piece compliant compensation mechanism, with a balancing error of only 0.03N while dimensions are such that the compensation part of the mechanism can be stored inside the hand grip of the instrument.

Commentary by Dr. Valentin Fuster
2004;():645-658. doi:10.1115/DETC2004-57243.

Proteins are evolution’s mechanisms of choice. Study of nano-mechanical systems must encompass an understanding of the geometry and conformation of protein molecules. Proteins are open or closed loop kinematic chains of miniature rigid bodies connected by revolute joints. The Kinematics community is in a unique position to extend the boundaries of knowledge in nano biomechanical systems. ProtoFold is a software package that implements novel and comprehensive methodologies for ab initio prediction of the final three-dimensional conformation of a protein, given only its linear structure. In this paper, we present the methods utilized in the kinematics notion and kinematics analysis of protein molecules. The kinematics portion of ProtoFold incorporates the Zero-Position Analysis Method and draws upon other recent advances in robot manipulation theories. We claim that the methodology presented is a computationally superior and more stable alternative to traditional molecular dynamics simulation techniques.

Topics: Proteins
Commentary by Dr. Valentin Fuster
2004;():659-667. doi:10.1115/DETC2004-57246.

Shaking forces and moments are often undesired. Complete balancing of these effects usually is associated with considerable additional mass and inertia. This paper proposes a light-weight solution concept to the complete force and moment balancing of linkages based on the combination of a counterweight and a separate counter-rotation into a single element. This element will be called a counter-rotary counterweight (CRCW). It will be shown that for a 1dof rotatable link a reduction of added mass by about 40% and a reduction by about 20% of added inertia have been achieved, as compared to a standard solution, after optimization of both mechanisms for minimal inertia. Generalization of the proposed principle is also discussed.

Topics: Weight (Mass)
Commentary by Dr. Valentin Fuster
2004;():669-677. doi:10.1115/DETC2004-57247.

This paper presents an efficient and novel computational protein prediction methodology called Kineto-Static Compliance Method. Successive Kineto-Static Fold Compliance is a methodology for predicting a protein molecule’s motion under the effect of an inter-atomic force field without the need for molecular dynamic simulation. Instead, the chain complies under the Kineto-Static effect of the force field in such a manner that each rotatable joint changes by an amount proportional to the effective torque on that joint. This process successively iterates until all of the joint torques have converged to zero. This configuration is equivalent to a stable, globally optimized potential energy state of the system or, in other words, the final conformation of the protein. This methodology is implemented in a computer software package named ProtoFold. In this paper, we have used Protofold to predict the final conformation of a small peptide chain segment, an alpha helix, and the Triponin protein chains from a denatured configuration. The results show that torques in each joint are minimized to values very close to zero, which demonstrates the method’s effectiveness for protein conformation prediction.

Topics: Proteins
Commentary by Dr. Valentin Fuster
2004;():679-688. doi:10.1115/DETC2004-57251.

The human ankle anatomical complex comprises the ankle and subtalar joints. The ankle represents the tibiotalar joint, connecting the tibia above, the fibula laterally and the talus below. Modelling of human joint passive motion is of great clinical relevance both for ligament reconstruction and for prosthesis design. The use of (equivalent) planar and spatial mechanisms for the kinematic modelling of joint passive motion proved to be a promising approach as it has been successfully utilized for the knee joint recently. Clinical evidence and experimental measurements led to infer that the relative passive motion between the tibia and the talus is a complex but single repeatable path constrained by articular surfaces and ligaments. The relative motion has the feature of a one degree of freedom rigid body guidance motion. Based on these considerations this paper presents two different equivalent spatial parallel mechanisms for this modelling. The mechanism links are taken from bones, ligaments and tendon structures, while kinematic pairs are appropriately chosen according to the corresponding type of anatomical connections. Simulation results are compared with corresponding experiments ones confirming the potential of the proposed approach.

Commentary by Dr. Valentin Fuster
2004;():689-698. doi:10.1115/DETC2004-57264.

This paper first presents the concept and the fabrication of a new type of compliant rolling joint which combines the advantages of compliant mechanisms with those of rolling link mechanisms. In this joint, flexible bands create the necessary constraints to enforce a rolling movement between two links. Then, the rapid prototyping techniques used for the compliant rolling joint fabrication are described. The kinematics of one application of this joint in a 3-DOF planar parallel mechanism are then presented. A semigraphical method was used to find the solutions to the inverse kinematic problem. Finally, the workspace analysis and the velocity equations are presented.

Commentary by Dr. Valentin Fuster
2004;():699-708. doi:10.1115/DETC2004-57268.

A suitable formulation and the implementing algorithms for involute and octoidal bevel-gear generation are proposed in this paper. In particular, the exact spherical involute tooth profile of bevel gears and their crown-rack is obtained through the pure-rolling motion of a great circle of the fundamental sphere on the base cone. Moreover, the tooth flank surface of octoidal bevel gears is obtained as the envelope of the tooth flat flank of the octoidal crown-rack during the pure-rolling motion of its flat pitch curve on the pitch cone. The proposed algorithms have been implemented in Matlab; several examples are included to illustrate their applicability.

Topics: Algorithms , Gears
Commentary by Dr. Valentin Fuster
2004;():709-716. doi:10.1115/DETC2004-57272.

A rigorous mechanism design approach was used to optimize a six-bar linkage fulfilling the task of lowering a 5-ton roll of paper from one elevation to another while changing its orientation by 90 degrees. This problem was motivated by a need in paper mills to improve upon the poor space and efficiency attributes of the current mechanism. The design approach used the problem statement to develop explicit design specifications. A formal type synthesis process was used to determine the optimal class of mechanism to meet the defined design specifications. This process yielded two classes of advantageous mechanisms, the Watt I and Stephenson III, which were further developed by dimensional synthesis. Using different approaches to the dimensional synthesis technique allowed three different mechanisms to be created that all moved the paper roll through the design positions. Distinguishing the best suited mechanism required a performance evaluation of static and dynamic states in addition to geometric evaluation. This work resulted in the optimization of a Stephenson III six-bar mechanism, generated by creating a path with prescribed timing base four-bar and an auxiliary motion control dyad, that met all of the design specifications.

Topics: Design
Commentary by Dr. Valentin Fuster
2004;():717-726. doi:10.1115/DETC2004-57273.

Most quadruped robots capable of running have employed bounding gaits at speeds far below those at which an animal of equal mass would gallop, which is a similar gait. This paper extends the spring-loaded inverted pendulum (SLIP) model to capture the key characteristics of trotting and galloping in biological systems. The objective is to establish a tool that will aid in determining the speed at which bounding or galloping is efficient for robotic systems. The SLIP model includes a linear damper in the legs to model all energy losses in a stride, and in the case of bounding, the body is taken to have an asymmetrical mass distribution. Results indicate that the model exhibits biological characteristics for both trotting and galloping, although duty factors are unrealistically low. Including leg mass in the models to account for additional energy loss does not offer improvement over the use of a linear damper alone.

Commentary by Dr. Valentin Fuster
2004;():727-731. doi:10.1115/DETC2004-57276.

This paper presents a method for the Jacobian derivation of 5-DOF 3R2T PMs (parallel mechanisms), where 3R denotes three rotational DOFs (degrees of freedom) and 2T denotes two translational DOFs. First the mobility analysis of such kind of parallel mechanisms is reviewed briefly. The Jacobian matrix of the single limb kinematic chain is obtained via screw theory, which is a 6 × 5 matrix. Then it is shown that the mobility analysis of such kind of PM is important when simplifying the 6 × 5 matrix into a 5 × 5 Jacobian matrix. After obtaining the 5 × 5 Jacobian matrix for each limb, a 5 × 5 Jacobian matrix for the whole mechanism can be established.

Commentary by Dr. Valentin Fuster
2004;():733-742. doi:10.1115/DETC2004-57281.

In this paper, a systematic method based on the screw theory is proposed for the geometric synthesis of a family of 3-DoF translational parallel manipulators (TPMs). The theory of screws and reciprocal screws is employed for the analysis of the geometric conditions undergoing the different types of constraints for the TPMs. In terms of these established geometric conditions, limb structures that can be used for constructing TPMs are enumerated, and a number of novel TPMs including both symmetrical structure and asymmetrical structure are synthesized accordingly. On the other hand, some composite kinematic pairs are proposed. The involvement of these composite kinematic pairs into the limbs of a TPM greatly enlarges the family of the TPMs.

Topics: Screws , Manipulators
Commentary by Dr. Valentin Fuster
2004;():743-748. doi:10.1115/DETC2004-57283.

A compliant parallel manipulator (CPM), is a kind of compliant mechanism characterizes a complicate topological structure and multiple degrees of freedom. As one of the kinematic characteristics of a CPM, the mobility of a CPM become complicate compared to its rigid-counterpart. In order to describe such a complicate kinematic characteristic of a CPM, “primary mobility of a compliant parallel manipulator” concept is proposed. By means of the screw theory, a method of quantifying the primary mobility of the CPM is investigated under the ground that the compliance matrix of the manipulator should be calculated primarily. By using this method, the primary mobility of two typical compliant parallel manipulators, one is a planar 3-RRR CPM and the other a spatial 3-RRPR CPM, is addressed respectively. This proposed method is also instructive for analyzing the instantaneous mobility of a general degenerate-DOF parallel manipulator or a Parallel Kinematic Machine (PKM).

Topics: Manipulators
Commentary by Dr. Valentin Fuster
2004;():749-754. doi:10.1115/DETC2004-57286.

This paper presents a prototype of a 4-dof robot wrist based on a redundant kinematic parallel structure. It shows the built prototype, and all subsystems are described in detail. The analysis of mechanism topology is performed, so its kinematic structure is characterized in terms of moving links, active and passive joints. By means of developed kinematic model, occurrence of singular configurations is predicted and workspace is evaluated. Finally, actuator driving torques are estimated for some motions of end-effector.

Commentary by Dr. Valentin Fuster
2004;():755-762. doi:10.1115/DETC2004-57293.

A reactionless machine will not apply reaction forces and moments on the mounting base during motion. This article presents the theory for design of a reactionless planar parallel mechanism. The salient features of this design are: (i) it uses auxiliary parallelograms to locate the center of mass of each leg at the base, (ii) choice of geometric and inertial parameters to satisfy the recationless conditions, (iii) selection of trajectory to achieve a reactionless design. We illustrate the reactionless feature of such a mechanism through computer simulations.

Commentary by Dr. Valentin Fuster
2004;():763-772. doi:10.1115/DETC2004-57294.

This article discusses systematically the characterization of instantaneous point-line motions, and the higher-order relationship between a point-line motion and the associated rigid body motions. The transformation of a point-line between two positions is depicted as a pure translation along the point-line followed by a screw displacement about their common normal and expressed with a unit dual quaternion referred to as the point-line displacement operator. The derivatives of the point-line displacement operator characterize the point-line motion to various orders with a set of characteristic numbers. Such a treatment leads to a consistent expression or unified treatment for the transformation of lines, point-lines, and rigid bodies. The relationships between point-line motions and rigid body motions are addressed in detail up to the third order.

Topics: Motion
Commentary by Dr. Valentin Fuster
2004;():773-779. doi:10.1115/DETC2004-57295.

An algorithm is presented, based on graph theory, for enumerating all feasible serial and/or parallel combined mechanisms from the given rotary or translational power source and specific kinematic building blocks. Through the labeled out-tree representations for the configurations of combined mechanisms, the enumeration procedure is developed by adapting the algorithm for the enumeration of trees. A rotary power source and four kinematic building blocks: a crank-rocker linkage, a rack-pinion, a double-slider mechanism, and a cam-follower mechanism, are chosen as the combination to illustrate the algorithm. And, two examples are provided to validate the algorithm.

Commentary by Dr. Valentin Fuster
2004;():781-788. doi:10.1115/DETC2004-57296.

A point-line refers to the combination of a directed line with an endpoint along the line. To trace a point-line trajectory, one must control not only the trajectory of the endpoint but also the direction of the point-line. This paper examines point-line trajectories by regarding a point-line trajectory as the combination of the endpoint trajectory referred to as the directrix and the trajectory of a unit vector in the point-line direction referred to as the indicatrix. A point-line trajectory is therefore characterized by the characteristics of the directrix, indicatrix, and the relationship between them. The paper presents (1) the characteristic numbers to characterize a point-line trajectory, especially the relationship between the directrix and indicatrix; (2) concepts of velocity at the striction point, the shifting velocity along the striction curve, and the shifting speed along the point-line; (3) The visualization of a point-line motion up to the second order; (4) the coordination of the indicatrix and directrix. An example is given to demonstrate the orientation control of the point-line with respect to the point trajectory.

Commentary by Dr. Valentin Fuster
2004;():789-797. doi:10.1115/DETC2004-57315.

Circuit defect is a fundamental and difficult problem in the process of linkage synthesis. In this paper, a Stephenson six-bar chain can be regarded as combination of a four-bar chain and a dyad. Base on the distance of two common joints shared by the four-bar chain and the dyad, the principle is elucidated and the model is established for circuit identification of Stephenson six-bar chains. The criteria and procedures to automatically identify the circuit and circuit defect of Stephenson six-bar chains are developed. The results provide theoretical basis and technical method for computer aided dimensional scheme selection of Stephenson six-bar mechanisms synthesis. In addition, the model for circuit identification can be directly used to solve the motion order defect that is often encountered in Stephenson six-bar mechanisms synthesis.

Topics: Chain , Circuits
Commentary by Dr. Valentin Fuster
2004;():799-808. doi:10.1115/DETC2004-57317.

This investigation concerns with the mechanical efficiency of Cardan joints. The model includes also the effects due to manufacturing and mounting errors and the influence of rotation speed on the efficiency. The joint has been modeled as an RCCC spatial linkage and the full dynamic analysis performed by means of dual vectors algebra.

Commentary by Dr. Valentin Fuster
2004;():809-817. doi:10.1115/DETC2004-57318.

This paper explores the development and performance of new constant-force compliant mechanisms that involve the addition of a translational spring element to slider-crank constant force mechanisms. The translational spring element has the additional requirement that, similar to a slider, it resists off-axis loads sufficiently to permit translation along only one axis. Geometric and energy storage parameters have been determined by optimization for five classes of mechanisms. The results of the optimization are values for geometric and energy storage parameters for each mechanism class for various levels of the translational spring parameter and various levels of constant-force behavior. The new configurations experience decreasing performance with increasing translational spring stiffness. The potential to implement a translational spring that also acts as a slider link provides the motivation for the new configurations. Such a spring would have the potential to completely remove friction from the mechanism and provide a constant-force solution that could replace current solutions such as hydraulic or pneumatic devices. The new configurations also have the potential to be manufactured as one piece or in layers, opening up new arenas for compressive constant-force mechanisms. Prototyping and testing of one of the new configurations are included as an example to demonstrate the use of the behavioral model.

Commentary by Dr. Valentin Fuster
2004;():819-830. doi:10.1115/DETC2004-57322.

In this paper a simulation of the cam actuator of a robotized gearbox is presented. The actuator is a barrel cam which moves a pin according to a prescribed motion law. The model of contact, between cam and follower, is based on a finite elements approach of the theory of beam on continuous elastic foundation, modified by the authors to include the effect of shear on the deformation and dynamic effects. The foundation modulus is computed using Hertz theory taking also into account the change of principal curvature values during the relative motion between cam and follower. This model has been embodied in a complete gear shifting simulation in order to compare the effects on contact forces, wear actions and shifting time of different cam profiles.

Commentary by Dr. Valentin Fuster
2004;():831-840. doi:10.1115/DETC2004-57323.

Cable robots have been extensively used for loading and unloading of cargo in shipping industries. A novel six-degree of freedom two-stage cable robot has been proposed by NIST for skin-to-skin transfer of cargo. In this paper, we look at a planar version of this two-stage cable robot. The disturbance motion from the sea is considered while modeling the dynamics of robot. The problem of robust control of the end-effector in the presence of unknown disturbances, along with maintaining positive tensions in the cables, is tackled by using the idea of redundancy. Simulation results show the effectiveness of the control strategy.

Commentary by Dr. Valentin Fuster
2004;():841-849. doi:10.1115/DETC2004-57324.

Design robustness is somewhat connected to tolerances. In fact, the lower is the sensitivity of the kinematic function to the deviations of manufacturing process, the higher is the robustness of the design. In this investigation is described a tolerance analysis method based on dual vectors kinematic modeling of spatial linkages and on Monte Carlo simulation of the random variables. In the present analysis the hypothesis of rigid bodies is valid and only kinematic variables are considered in output. The method is applied to a Cardan joint modelled as an RCCC linkage with main dimensions considered as stochastic variables with Gaussian distribution. Dual vectors are well known in kinematic analysis and synthesis of spatial mechanisms. When compared with traditional vectorial methods, dual vectors show an enhanced capability to model misalignments among kinematic pairs axes. Although this is not the first time that dual vectors are used for the kinematic and dynamic analysis of spatial mechanisms with manufacturing errors, the present use of dual vectors to model joint clearances seems somewhat novel.

Topics: Linkages
Commentary by Dr. Valentin Fuster
2004;():851-858. doi:10.1115/DETC2004-57325.

An innovative flexure design concept is developed that involves the use of composite materials with bonded piezoelectric actuators to create a “smart actuator” for nanometer scale positioning both in and out of plane. The utilization of smart actuators/sensors will have many benefits over current positioning stages including the utilization of advantageous composite properties, the increase of the resonant frequency of the stage system, the simplification of the manufacturing procedure, and possible increased range of motion. Enhanced mechanical response is made possible by using composite materials and proper layer orientation. This work is a “proof of concept” design for the ultra-precision smart actuator. Analytical and numerical models were developed to determine the response of the composite material and piezoelectric actuator system. Experimentation was performed on the system to verify the results of the mathematical and numerical models. Good correlation between mathematical, numerical, and experimental data was seen in axial and bending modes of operation. Results from the parametric and experimental verification studies are presented to illustrate the efficacy of the design concept.

Commentary by Dr. Valentin Fuster
2004;():859-865. doi:10.1115/DETC2004-57326.

The spatial RPRP linkage is a closed-loop linkage composed of two revolute and two prismatic joints. It is an overconstrained linkage, which does not obey the Grübler criteria. This paper investigates the finite displacement of the RPRP linkage and shows that all possible finite displacement screws of the coupler link form a screw system of the second order. This paper explores two configurations of the RPRP linkages, folded and unfolded. The screw systems for both types of RPRP linkage are obtained by intersecting two 3-systems corresponding to the RP and PR dyads. A numerical example is provided to verify the result.

Topics: Screws , Linkages
Commentary by Dr. Valentin Fuster
2004;():867-874. doi:10.1115/DETC2004-57330.

In this paper, the energetics of a flapping wing micro air vehicle is analyzed with the objective of design of flapping wing air vehicles. The salient features of this study are: (i) design of an energy storage mechanism in the air vehicle similar to an insect thorax which stores part of the kinetic energy of the wing as elastic potential energy in the thorax during a flapping cycle; (ii) inclusion of aerodynamic wing models using blade element theory and inertia of the mechanism using rigid body modeling techniques; (iii) optimization of parameters of the energy storage mechanism using the dynamic models so that energy input from the external actuators during a flapping cycle is minimized. A series of engineering prototypes based on these studies have been fabricated which justify the use of these mathematical techniques.

Commentary by Dr. Valentin Fuster
2004;():875-883. doi:10.1115/DETC2004-57340.

The paper presents an algorithm for the real-time evaluation of the actual end-effector orientation (pose) of general parallel spherical wrists. Conceptually, the method relies on the evidence that the pose of a rigid body is defined once the location of at least two linearly independent vectors attached to the body is known. The location of these vectors of the wrist end-effector is determined by the solution of the direct position analysis of some properly chosen kinematic chains (legs) of the manipulator. In order to accomplish this analysis, extra-sensors, which measure suitable non-actuated variables of the chosen legs, need to be placed in addition to the ones normally embedded in the servo motors, i.e. the sensors which measure the actuated variables. From a mathematical point of view, the algorithm is built on the Polar Decomposition of a matrix and has inherent least square features. Thus, together with measurement redundancy, i.e. more sensors (extra-sensors) than the mechanism degrees of freedom, the method also allows minimizing the influence of both round-off and measurement errors on the estimation of the location of the wrist end-effector. The method is general but, in order to prove its effectiveness, without loss of generality it has been customized to the solution of the (3-UPS)S fully parallel wrist architecture. Comparison of the proposed method, in both its general and specialized form, with others from the literature is provided.

Topics: Sensors
Commentary by Dr. Valentin Fuster
2004;():885-891. doi:10.1115/DETC2004-57341.

In a recent study, Rastegar, et al. (2003), presented a special class of planar and spatial linkage mechanisms in which for a continuous full rotation or continuous rocking motion of the input link, the output link undergoes two continuous rocking motions. In a special case of such mechanisms, for periodic motions of the input link, the output motion is periodic with a doubled fundamental frequency. The above class of linkage mechanisms were referred to as “speed-doubling” linkage mechanisms. Such mechanisms can be cascaded to further double the fundamental frequency (rocking motion) of the output motion. They can also be cascaded with other linkage mechanisms to obtain crank-rocker or crank-crank type of mechanisms. The conditions for the existence of “speed-doubling” linkage mechanisms were also provided. In this paper, a study of the dynamics of a “speed-doubling” linkage mechanism is presented. It is shown that such mechanisms have dynamic advantage over regular mechanisms designed to achieve similar output motions. The main advantage of such mechanisms is shown to be their lower peak input torque requirement, and that the required torque generally has lower amplitude high-frequency components. The speed-doubling mechanisms have practical applications, particularly when higher output speeds are desired, since higher output motions can be achieved with lower input speeds and smaller motors.

Commentary by Dr. Valentin Fuster
2004;():893-900. doi:10.1115/DETC2004-57343.

We anticipate that the motion of biped machines should be similar to human motion to achieve dynamic stability. In this paper, a novel approach to compute dynamically stable gait of a planar six degree-of-freedom biped is presented. This approach is analytical and is based on periodic property of a gait cycle. The resulting gait satisfies all dynamic stability criteria.

Commentary by Dr. Valentin Fuster
2004;():901-908. doi:10.1115/DETC2004-57352.

A new technique for modeling rigid bodies undergoing spatial motion using an explicit time-integration finite element code is presented. The key elements of the technique are: (a) use of the total rotation matrix relative to the inertial frame to measure the rotation of the rigid bodies; (b) time-integration of the rotational equations of motion in a body fixed (material) frame, with the resulting incremental rotations added to the total rotation matrix; (c) penalty formulation for creating connection points (virtual nodes which do not add extra degrees of freedom) on the rigid-body where joints can be placed. The use of the rotation matrix along with incremental rotation updates circumvents the problem of singularities associated with other types of three and four parameter rotation measures. Benchmark rigid multibody dynamics problems are solved to demonstrate the accuracy of the present technique.

Commentary by Dr. Valentin Fuster
2004;():909-914. doi:10.1115/DETC2004-57354.

Based on the model for circuit analysis of a Stephenson six-bar chain, this paper presents a feature point loop method to identify the motion order defect of a Stephenson six-bar chain. The proposed method can determine whether a linkage synthesized by precision point synthesis can pass through all positions in the prescribed order and yet not have to carry through the detailed kinematic analysis of the linkage. In addition, this method is applicable to any type Stephenson six-bar linkage and regardless of whichever link is used as the input. The result resolves one fundamental and troublesome problem encountered in a Stephenson six-bar linkage synthesis.

Topics: Motion , Chain
Commentary by Dr. Valentin Fuster
2004;():915-922. doi:10.1115/DETC2004-57362.

One method for generating a Watt II six-bar with a sliding output is to identify the location of its six revolute (R) joints. The location of the output prismatic (P) joint does not need to be identified as it can be assumed to allow vertical motion through the final R joint. If classified, like the 4R, based on the range of motion of each link as cranking or rocking, a number of possibilities exist classifying the four moving links connecting the six R joints. In this paper, we introduce a methodology for ensuring that the link connecting the first two R joints is a crank. As there are four more locations to specify, each location needs to be selected in light of the locations of the previous R joints. To accomplish this goal, we identify the Watt II six-bar transition linkages on the boundary between classifications.

Topics: Motion , Linkages
Commentary by Dr. Valentin Fuster
2004;():923-933. doi:10.1115/DETC2004-57373.

Static balancing can be used as a means to compensate certain (disturbance) forces in a mechanism. This paper describes a two-step design strategy for statically balanced mechanisms, balanced with normal extension springs. During the first step, ideal springs (with zero free length) are used to develop a perfectly balanced design. This can be done with the aid of a design framework, which is based on a set of elementary rules for the modification of a standard configuration of idealized springs under constant system potential energy. In the second step, the ideal springs are replaced with normal springs and the resulting unbalanced system is numerically optimized, using a genetic algorithm. The emphasis of this paper will be on the second step. A laparoscopic grasper is used as a design example and is optimized for the specified range of motion. It is optimized with respect to a constant total potential energy function and a constant force transfer function between the operating force and the resulting force that is applied by the grasper.

Commentary by Dr. Valentin Fuster
2004;():935-942. doi:10.1115/DETC2004-57375.

Kinematic calibration is a process that estimates the actual values of geometric parameters to minimize the error in absolute positioning. Measuring all the components of Cartesian posture assure identification of all parameters. However, measuring all components, particularly the orientation, can be difficult and expensive. On the other hand, with partial pose measurements, experimental procedure is simpler. However, all parameters may not be identifiable. This paper proposes a new device that can be used to identify all kinematic parameters with partial pose measurements. Study is performed for a 6 DOF (degree-of-freedom) fully parallel Hexa Slide manipulator. The device, however, is general and can be used for other parallel manipulators. The proposed device consists of a link with U joints on both sides and is equipped with a rotary sensor and a biaxial inclinometer. When attached between the base and the mobile platform, the device restricts the end-effector’s motion to 5 DOF and measures two position components and one rotation component of the end-effector. Numerical analyses of the identification Jacobian reveal that all parameters are identifiable. Computer simulations show that the identification is robust for the errors in the initial guess and the measurement noise. Intrinsic inaccuracies of the device can significantly deteriorate the calibration results. A measurement procedure is proposed and cost functions are discussed to prevent propagation of the inaccuracies to the calibration results.

Commentary by Dr. Valentin Fuster
2004;():943-951. doi:10.1115/DETC2004-57378.

A general formulation for the synthesis of the pitch cones of N-lobed elliptical bevel gears and the pitch surface of their conjugate crown-rack is proposed. In particular, both pitch cones and the pitch surface of their crown-rack can be obtained for any number and combination of lobes and in any configuration during their pure-rolling motion. This formulation has been implemented in a suitable Matlab program and several significant examples are shown, where circular bevel gears become a particular case. Computer animations are also available at URL: http://webuser.unicas.it/weblarm/larmindex.htm.

Topics: Gears
Commentary by Dr. Valentin Fuster
2004;():953-962. doi:10.1115/DETC2004-57379.

This paper presents a sensitivity analysis of the Orthoglide, a 3-DOF translational Parallel Kinematic Machine. Two complementary methods are used to analyze its sensitivity to its dimensional and angular variations. First, a linkage kinematic analysis method is used to have a rough idea of the influence of the dimensional variations on the location of the end-effector, and shows that the variations in design parameters of the same type from one leg to another one have the same influence on the end-effector. However, this method does not allow the designer to know the influence of the variations in the parallelograms. Thus, a differential vector method is used to study the influence of the dimensional and angular variations in the parts of the manipulator, and particularly the variations in the parallelograms, on the position and orientation of the end-effector. It turns out that the isotropic kinematic configuration of the manipulator is the least sensitive one to its geometrical variations, contrary to the closest configurations to its kinematic singular configurations, which are the most sensitive to geometrical variations.

Commentary by Dr. Valentin Fuster
2004;():963-969. doi:10.1115/DETC2004-57380.

Colonoscopy is a standard medical procedure in which a long and flexible endoscope is inserted into the rectum for inspection of the large intestine and for simple interventions. Pushing the endoscope tip from behind via a long and flexible tube leads easily to buckling when the tip comes in contact with sharp curves in the intestinal wall. Buckling is accompanied by painful cramps and makes it difficult to complete the procedure. A way to avoid buckling is not to push the tip from behind, but to use the friction with the intestinal wall to pull the tip forward. This paper describes the state-of-the-art in research on intestinal locomotion methods and presents a new locomotion method based on a rolling donut that is positioned around the endoscope tip. The donut functions like a circular caterpillar and is constructed from three stents that generate high friction with the intestinal wall. The diameter of the donut can be changed and the stents can be driven independently to reduce slip in intestinal curves. The resulting Rolling-Stent Endoscope contains a new steerable mechanism by which the tip can be bent in all directions over a very large angle. The Rolling-Stent Endoscope was applied for a patent and a prototype is under development for evaluation in the intestine of a pig.

Topics: stents
Commentary by Dr. Valentin Fuster
2004;():971-976. doi:10.1115/DETC2004-57381.

This paper investigates the equivalent contour representation of the soft contact, proposes torsional stiffness coefficients relating the contact contour to a soft-robotic-finger contact and develops the finite elastic elements of the soft finger. The contact is modeled between a grasped object and a soft fingertip as a number of contours with a set of equivalent elastic contact points arranged on each contour. The stiffness matrix model is presented and related to the torsional stiffness coefficients. The paper further reveals the effect of the robotic finger stiffness on the stiffness coefficients by modeling the finger with a finite number of elastic elements arranged serially. The couples which arise from the elastic elements give translational and torsional stiffness characteristics of the soft contact. This presents a way of measuring soft contact stiffness. In particular, the torsional stiffness is investigated with its variance with the contours. A case study is presented at the end.

Topics: Robotics
Commentary by Dr. Valentin Fuster
2004;():977-985. doi:10.1115/DETC2004-57382.

The paper presents the position analysis of a spatial structure composed of two platforms mutually connected by one RRP and three SS serial kinematic chains, where R, P, and S stand for revolute, prismatic, and spherical kinematic pair respectively. A set of three compatibility equations is laid down that, following algebraic elimination, results in a 28th -order univariate algebraic equation, which in turn provides the addressed problem with 28 solutions in the complex domain. Among the applications of the results presented in this paper is the solution to the forward kinematics of the Tricept, a well-known in-parallel-actuated spatial manipulator. Numerical examples show adoption of the proposed method in dealing with two case studies.

Commentary by Dr. Valentin Fuster
2004;():987-992. doi:10.1115/DETC2004-57387.

In this paper, the problem of identification of cam profiles is addressed by proposing a suitable kinematic model. A suitable test-bed has been designed and built at LARM: Laboratory of Robotics and Mechatronics in Cassino for experimental tests. The study proposed in this paper has been carried out specifically for circular-arc cams and validation tests have given satisfactory results for identification of the profile curvature.

Topics: Cams , Mechatronics , Robotics
Commentary by Dr. Valentin Fuster
2004;():993-1005. doi:10.1115/DETC2004-57388.

The conceptual design of compliant mechanisms is generally performed using one of two methods: topology optimization or the Pseudo-Rigid-Body Model. In this paper, we present a conceptual design methodology which utilizes a building block approach. The concept of the instant center is developed for compliant mechanisms and is used to characterize the building blocks. The building block characterization is used in guiding the problem decomposition. The compliant four-bar building block is presented as a base mechanism for the conceptual design. The geometric advantage is used as a quantitative measure to guide the designer in determining the shape of the building block. An example problem demonstrates the methodology’s capacity to obtain viable conceptual designs in a straightforward manner. Resulting mechanisms satisfy initial kinematic requirements and are ready for further refinement using size and geometry optimization.

Commentary by Dr. Valentin Fuster
2004;():1007-1014. doi:10.1115/DETC2004-57399.

In this paper a macro parallel robot is presented in which conventional bearings are replaced by pseudo-elastic flexure hinges. The robot consists of a spatial parallel structure with three translational degrees of freedom and is driven by three linear direct drives. The structure has been optimized with respect to workspace and transmission ratio. Additionally, in simulations with the FEM tool ANSYS different geometrical arrangements and combinations of flexure hinges have been investigated with respect to the dynamic behavior of the compliant mechanism. Due to the symmetrical character of the structure and the optimized design of the combined flexure hinges the structure is very stiff. The experimental measured repeatability of the compliant robot is below 0.3 μm.

Topics: Robots
Commentary by Dr. Valentin Fuster
2004;():1015-1021. doi:10.1115/DETC2004-57401.

An ultra-precision diaphragm flexure stage was designed and optimized for pure out-of-plane Z-motion guidance with an effort to minimize parasitic lateral and tilt motions. The diaphragm flexure stage will be used to guide an objective lens along the optical axis in a high performance microscope. A concept with cross-linked radial spoke geometry akin to a bicycle wheel was designed to maximize its in-plane to out-of-plane stiffness ratio. The geometry was then optimized using finite element analysis. Several prototypes were manufactured using an abrasive water jet machining center and tested with a reconfigurable test fixture. Sets of two and four diaphragms were tested. Negligible improvement in parasitic motion was observed between the two and four-diaphragm configurations. The experimental results for out-of-plane stiffness matched the finite element result within 0.3%. Negligible hysteresis in vertical motion was observed. Maximum lateral parasitic motion on the order 2 μm was measured over a Z-motion range of 140 μm. The lateral displacement was linear (R2 =0.998) and repeatable, and could be an artifact of misalignment of the diaphragms in the test fixture.

Commentary by Dr. Valentin Fuster
2004;():1023-1032. doi:10.1115/DETC2004-57403.

The paper introduces tape springs as elements of fully compliant mechanisms. The localized folds of tape springs serve as compact revolute joints, with a very small radius and large range of motion, and the unfolded straight segments serve as links. By exploiting a tape spring’s ability to function as both links and joints, we present a new method of realizing fully compliant mechanisms with further simplification in their construction. Tape springs, typically found in carpenter tape rules, are thin-walled strips having constant thickness, zero longitudinal curvature, and a constant transverse curvature. The paper presents a closed-loop tape spring mechanism. By representing its folds as idealized revolute joints and its variable length links as sliding joints connecting rigid links, we present a modified Gruebler’s equation to determine its kinematic and idle degrees of freedom. To realize practical utility of tape spring mechanisms, we propose a simple actuation scheme incorporating shape memory alloy (SMA) wire actuators and successfully demonstrate its performance with a proof-of-concept prototype. The paper also presents potential applications for actuated tape spring mechanisms including a large displacement translational mechanism, planar positioning mechanisms, bi-stable, multi-stable, and variable stiffness mechanisms.

Commentary by Dr. Valentin Fuster
2004;():1033-1042. doi:10.1115/DETC2004-57406.

In this paper, a new five-degree-of-freedom parallel manipulator is described and modeled. This structure has been specially designed for medical applications that require in the same time mobility, compactness and accuracy around a functional point. The purpose of this robotic device is to help practitioners to perform accurate needle insertions while preserving them from harmful intra-operative X-ray imaging devices. The system is built from revolute joints, among which only five joints are actuated to convey the required five degrees of freedom to its moving platform. A numerical simulation of the workspace and a physical prototype are presented.

Commentary by Dr. Valentin Fuster
2004;():1043-1054. doi:10.1115/DETC2004-57411.

The use of redundant actuation in the design and control of active automotive suspension systems is described. Redundantly actuated systems consist of more active force / torque inputs than degrees-of-freedom and allow for active control of the effective stiffness of the system to the environment without a change in the equilibrium position. A frontal plane half-car, double A-arm, independent suspension model is investigated. Results show that five actuators, with one connecting the two suspensions, is required for full stiffness and motion control. Due to the dependence of this approach on correct stiffness modeling a previously developed stiffness model is reviewed. The validity of this model is illustrated through some simple yet sufficient examples.

Commentary by Dr. Valentin Fuster
2004;():1055-1061. doi:10.1115/DETC2004-57412.

A mechanism is presented that can generate insect wing like motion. This motion includes both flapping out of the stroke plane in addition to twist of the wing. The mechanism has a single degree-of-freedom and employs a five bar mechanism in addition to an auxiliary four bar mechanism. The parameters in the mechanism were optimized to generate a prescribed motion of the wing taken from a hawk moth kinematic flight data. A scaled model of the mechanism was fabricated to verify practical feasibility of the design. In future we will miniaturize this mechanism on our flying bird prototypes.

Commentary by Dr. Valentin Fuster
2004;():1063-1072. doi:10.1115/DETC2004-57413.

Screws reciprocal to five lines appear in the inverse Jacobian and statics equations of many serial and parallel robots. By closely examining the geometry of systems of five lines, a set of geometric procedures for computing reciprocal screws is derived. The methodology is based on the fact that one screw in a cylindroid reciprocal to four lines is reciprocal to a fifth independent line. The constructions are easy to implement and provide geometric insight into reciprocal screws and singularities.

Topics: Screws
Commentary by Dr. Valentin Fuster
2004;():1073-1079. doi:10.1115/DETC2004-57414.

This paper describes the design and utilization of an advanced dynamic dynamometer test bed for electromechanical robot actuators. Typically actuators are tested at steady state under continuous load, but this is insufficient to characterize the nonlinear properties of actuators. The test bed was specifically designed with the capability to emulate high bandwidth complex duty cycle loads. Furthermore, the test bed is modular which enables rapid reconfiguration of test components. The test bed is comprised of a servo motor which serves as a dynamic load emulator, a brake, a clutch, full sensor array for comprehensive monitoring of test variables, and supporting motion control hardware. Two preliminary tests of a permanent magnet synchronous motor (PMSM) were conducted to demonstrate some of the test bed’s capabilities. These included a torque ripple test and a test to determine torque variation for a sinusoidal load at constant speed. The results show increased torque ripple for increasing speed. The sinusoidal loading test demonstrates the nonlinearities associated with actuator dynamics. Future testing will focus on the full characterization of an actuator.

Topics: Robots , Actuators
Commentary by Dr. Valentin Fuster
2004;():1081-1102. doi:10.1115/DETC2004-57415.

Prior investigations have presented the use of optimal control theory in the design of high-speed cam follower systems. These investigations were constrained by the difficulty of numerical solutions of optimal control problems, and this limited the types of criteria investigated, the state inequality constraints considered, and the realism of the models used. In recent years numerical solution techniques based on direct multiple shooting and using specially structured sequential quadratic programming have become available. These are capable of handling complex optimization criteria and imposing state and control in-equality constraints. This paper investigates and illustrates the potential of such methods in revolutionizing high speed automotive cam design. Cam design is complicated by the number of partially competing criteria one is interested in. This work synthesizes an optimal cam design approach, considering a range of speeds, the area under the lift curve, Hertzian contact stress, vibrations and residual vibrations, energy loss, cam curvature, follower force, and contact stress. The paper illustrates how all of these criteria can be integrated into the optimization in the design stage. Since the polydyne method is not an optimization procedure, the resulting design is superior to polydyne design in all aspects considered.

Commentary by Dr. Valentin Fuster
2004;():1103-1111. doi:10.1115/DETC2004-57416.

Compliant mechanisms present several design advantages that may be exploited in prosthetic joint design — low friction and wear, low part count, light weight, high reliability, efficient manufacturing and assembly, etc. However, they also come with their share of challenges. For example, in the past compliant mechanisms have not been viable alternatives to rigid-body joints in high compression applications. Two principles, isolation and inversion, can be successfully applied to many compliant joints to increase their ability to withstand high compressive loads and make them viable design alternatives to some current prosthetic joint designs. This paper presents preliminary design concepts for a compliant prosthetic knee. The inverted cross-axis flexural pivot design is selected and a proof of concept prototype is constructed and tested. The prototype exceeds the functional specification of 550 lbs, and satisfies a maximum deflection of 140°, an infinite life for a deflection of 110°, and a weight under 2 lbs.

Commentary by Dr. Valentin Fuster
2004;():1113-1118. doi:10.1115/DETC2004-57420.

Mechanism synthesis requires the use of optimization methods to obtain approximate solution whenever the desired number of positions the mechanism is required to traverse exceeds a few (five in a 4R linkage). Deterministic gradient-based methods are usually impractical when used alone because they move in the direction of local minima. Random search methods on the other hand have a better chance of converging to a global minimum. This paper presents a tabu-gradient search based method for optimum synthesis of planar mechanisms. Using recency-based short-term memory strategy, tabu-search is initially used to find a solution near global minimum, followed by a gradient search to move the solution ever closer to the global minimum. A brief review of tabu search method is presented. Then, tabu-gradient search algorithm is applied to synthesize a four-bar mechanism for a 10-point path generation with prescribed timing task. As expected, Tabu-gradient base search resulted in a better solution with less number of iterations and shorter run-time.

Commentary by Dr. Valentin Fuster
2004;():1119-1127. doi:10.1115/DETC2004-57422.

The synthesis of spherical motion generators in the presence of an incomplete set of finitely-separated attitudes is discussed in this paper. Given that five attitudes of the coupler link define a discrete set of dyads, any number of attitudes smaller than five is considered incomplete in this paper. The attitudes completing the set are determined so as to produce a robust linkage against variations in these attitudes. To this end, a theoretical framework as well as a general methodology for robust synthesis are laid down. Robustness is needed in this context to overcome the presence of uncertainty due to the selection of the intermediate attitudes, which many a time are left up to the mechanism designer’s judgment. To validate the concepts and illustrate the application of the methodology proposed here, we include a numerical example.

Topics: Motion , Generators
Commentary by Dr. Valentin Fuster
2004;():1129-1138. doi:10.1115/DETC2004-57424.

Although bevel-gear wrists are widely used in industrial manipulators due to their simple kinematics and low manufacturing cost, their gear trains function under rolling and sliding, the latter bringing about noise and vibration. Sliding is inherent to the straight teeth of the bevel gears of these trains. Moreover, un-avoidable backlash introduces unmodeled dynamics, which mars robot performance. To alleviate these drawbacks, a gearless pitch-roll wrist is currently under design for low backlash and high stiffness. The wrist consists of spherical cam-rollers and spherical Stephenson linkages; two roller-carrying disks drive a combination of cams and Stephenson mechanisms rotating as a differential mechanism. In this paper, the design of the chain of spherical Stephenson mechanisms (SSMs) is introduced. The problem of the dimensional synthesis is addressed and interference avoidance is discussed. An embodiment of the concept is also included.

Topics: Chain , Design , Mechanisms
Commentary by Dr. Valentin Fuster
2004;():1139-1146. doi:10.1115/DETC2004-57427.

This paper presents a practical scheme for path and trajectory generation with applications in real-time simulation of robotic systems. The path planning is solved as a series of local obstacle avoidance problems which produce via points in the 3D Cartesian space as well as the normal vector to the obstacle surface at each via point. This vector is later utilized for trajectory generation. An intermediate orientation representation is used to guarantee continuity at the angular acceleration level for real-time simulation and control purposes. The proposed method is implemented in a real-time simulation environment for testing and final application. The real-time data generation allows hardware-in-the loop simulation, which is crucial in many aerospace applications.

Commentary by Dr. Valentin Fuster
2004;():1147-1153. doi:10.1115/DETC2004-57434.

The joint torques in hip, knee and ankle are computed using inverse dynamic model during standing up for a paraplegic patient. The joint torque comprises the dynamical torque due to the inertia forces, and a passive torque due to the muscles and gravitational torque. It has been observed that the contribution to the joint torques by the gravitational torque is dominant. On the basis of this result, a gravity balanced assistive device is proposed for the elderly and impaired people such as spinal cord injury and paraplegic patients. This passive device uses a hybrid method to identify the center of mass of the system using auxiliary parallelograms first. Next appropriate springs are connected to the device to vanish the total potential energy of the system due to the gravity during standing up. A prototype with the underlying principles is currently being fabricated at the University of Delaware.

Commentary by Dr. Valentin Fuster
2004;():1155-1164. doi:10.1115/DETC2004-57435.

In a recent study, the authors presented a systematic method for the modification of the output motion of linkage mechanisms with closed-loop chains using cams positioned at one or more of the mechanism joints. In this paper, the method is applied to the design of a linkage mechanism with an integrated cam mechanism for the purpose of eliminating the high harmonic component of the output link motion. The mechanism may be synthesized using well-developed linkage mechanism synthesis techniques for the intended application. Based on this method, a cam mechanism is synthesized for a prescribed output link motion while limiting the output motion to a simple harmonic motion with the frequency of its constant input velocity. The mechanism is constructed and tested. In mechanisms with relatively rigid links, the primary source of high harmonics in the output motion is the nonlinearity of the kinematics of their closed-loop chains. With the present method, a selected range or ranges of high harmonic motions generated due to such nonlinearities may be eliminated by integrating appropriately designed cams. By eliminating the high harmonic component of the output motion of a mechanism, the potential vibrational excitation that the mechanism can impart on the overall system, including its own structure, is greatly reduced. The resulting system should therefore be capable of operating at higher speeds and with increased precision.

Topics: Motion , Linkages , Mechanisms
Commentary by Dr. Valentin Fuster
2004;():1165-1172. doi:10.1115/DETC2004-57439.

Compliant mechanisms have been designed for various types of applications to transmit desired forces and motions. In this paper, we explore an application of compliant mechanisms for active vibration isolation systems. For this type of application, an actuator and a compliant mechanism are used to cancel undesired disturbance, resulting in attenuated output amplitude. An actuator provides external energy to the system while a compliant mechanism functions as a transmission controlling the amount of displacement transmitted from the actuator to the payload to be isolated. This paper illustrates, based on preliminary results of finite element analyses (FEA), that a compliant mechanism equipped with an actuator can be used as an active vibration isolator to effectively cancel a known sinusoidal displacement disturbance at low frequencies by using a feedforward disturbance compensation control. The nonlinear FEA shows that a sinusoidal displacement disturbance of 6.0 mm amplitude is reduced by 95% at 3.9 Hz and 91% at 35.1 Hz with a sinusoidal displacement controlled input of 0.73 mm amplitude.

Commentary by Dr. Valentin Fuster
2004;():1173-1181. doi:10.1115/DETC2004-57441.

A control algorithm to manipulate the input actuation forces on hinge-free compliant mechanism models synthesized utilizing continuum topology optimization formulations is implemented and successfully tested in this work, such that the mechanisms follow their specified trajectories and nearby trajectories with high degree of accuracy even when confronted with different resistance forces. The validity of the proposed formulations is tested and demonstrated on number of practical problems involving finite deformation.

Commentary by Dr. Valentin Fuster
2004;():1183-1191. doi:10.1115/DETC2004-57444.

There is a need for HMMWV dynamic models for autonomous control of these vehicles. Currently, such four-wheel-drive vehicle models are unavailable. This paper derives the kinematics, dynamics and computation aspects of the problem. The subsystem models, e.g. tire, brake and powertrain, are presented as well. Our results show agreement with published simulations of the HMMWV. The current simulations in Mat-lab execute 30 times slower than real-time. Future work will implement this model in a planning algorithm for autonomous vehicle control.

Commentary by Dr. Valentin Fuster
2004;():1193-1201. doi:10.1115/DETC2004-57447.

This research uses new developments in redundancy resolution and real-time capability analysis to improve the ability of an articulated arm to satisfy task constraints. Task constraints are specified using numerical values of position, velocity, force, and accuracy. Inherent in the definition of task constraints is the number of output constraints that the system needs to satisfy. The relationship of this with the input space (degrees of freedom) defines the ability to optimize manipulator performance. This is done through a Task-Based Redundancy Resolution (TBRR) scheme that uses the extra resources to find a solution that avoids system constraints (joint limits, singularities, etc.) and satisfies task constraints. To avoid system constraints, we use well-understood criteria associated with the constraints. For task requirements, the robot capabilities are estimated based on kinematic and dynamic manipulability analyses. We then compare the robot capabilities with the user-specified requirement values. This eliminates a confusing chore of selecting a proper set of performance criteria for a task at hand. The breakthrough of this approach lies in the fact that it continuously evaluates the relationship between task constraints and system resources, and when possible, improves system performance. This makes it equally applicable to redundant and non-redundant systems. The scheme is implemented using an object-oriented operational software framework and its effectiveness is demonstrated in computer simulations of a 10-DOF manipulator.

Commentary by Dr. Valentin Fuster
2004;():1203-1210. doi:10.1115/DETC2004-57450.

A point-line is the combination of a directed line with a reference point on it. In this article, the spatial distance between two point-lines in space is defined based on a point-line displacement model. The displacement of a point-line from one position to the other is uniquely described as the composition of a pure translation along the point-line and a screw displacement about the common normal of the two positions. It is shown that such a displacement model leads to a simpler configuration of the underlying screw triangle and defines the shortest distance between two point-lines. The dual quaternion algebra is used to describe the idea with mathematic expressions.

Topics: Screws , Displacement
Commentary by Dr. Valentin Fuster
2004;():1211-1220. doi:10.1115/DETC2004-57453.

A formulation for design of continuous, hinge-free compliant mechanisms is developed and examined within a continuum structural topology optimization framework. The proposed formulation involves solving two nested optimization problems. In the inner problem the arrangement of a constrained amount of structural material is optimized to maximize the mechanism’s mutual potential energy in response to a force loading at the input port while working against artificial springs on the input and output ports. As the relative stiffness of the artificial springs increases, the material continuity of the mechanism also increases to the point where de facto “hinge” regions are eliminated. In the outer problem, one solves for an appropriate amount of structural material that yields the desired compliance characteristics of the mechanism when working against the workpiece resistance. Different aspects of the proposed formulation are demonstrated on a number of examples and discussed.

Commentary by Dr. Valentin Fuster
2004;():1221-1230. doi:10.1115/DETC2004-57455.

This paper presents an analysis tool and design method for MEMS parallel mechanisms. Due to processing constraints in MEMS fabrication, flexure joints are frequently used in MEMS mechanisms. Flexure joints offer advantages over other joint designs due to their monolithic characteristics. They can be used to reduce the size of manipulators or to increase the precision of motion. Their inherent flexibility, however, also results in task space compliance which needs to be carefully designed to match the task specification. This paper presents an analysis and design tool for such mechanisms by using the differential kinematics. Performance metrics are chosen based on manipulability and task stiffness matrices, which in turn are used in a multi-objective optimization. As an illustrative example, a 1-DOF MEMS parallel mechanism based on the macro- and meso-scale models designed by NIST is considered with several choices of performance metrics and design variables. The resulting designs are successfully fabricated using DRIE process.

Commentary by Dr. Valentin Fuster
2004;():1231-1239. doi:10.1115/DETC2004-57466.

A vibratory platform is a crucial part of automated assembly and has a similar structure to many parallel mechanisms. To investigate the platform, it is necessary to measure the helicoidal motion produced. This paper develops a new way of measuring the helicoidal motion by using reflective projection and by developing an analytical relationship between the measured vectors. A further relationship is developed between the cross sectional area of the laser light, the displacement of the platform and the output voltage from a photodiode. The study leads to a novel development of a non-invasive laser based measuring device. The paper further develops a dynamic calibration method, leading to calibration methods for both angular and linear displacements. The method is then confirmed by measurement to an accuracy of 20μm.

Topics: Lasers , Motion , Springs
Commentary by Dr. Valentin Fuster
2004;():1241-1249. doi:10.1115/DETC2004-57468.

Laparoscopy is abdominal surgery performed with long tools inserted through small incisions. The use of small incisions reduces patient trauma, but also eliminates the surgeon’s ability to directly view and touch the surgical environment. These limitations generally restrict the application of laparoscopy to less complex procedures. Large robots external to the patient have been used to aid in the manipulation of the tools and improve dexterity. This paper presents a theoretical and experimental analysis of miniature in vivo robots. The objective is to develop a wireless mobile imaging robot that can be placed inside the abdominal cavity during surgery. Such robots will allow the surgeon to view the surgical environment from multi-angles. The motion of these in vivo robots will not be constrained by the insertion incisions.

Commentary by Dr. Valentin Fuster
2004;():1251-1260. doi:10.1115/DETC2004-57472.

PPR-PMs (parallel manipulators) are the parallel counterparts of the 3-DOF PPR serial robots, which are composed of two P (prismatic) joints and one R (revolute) joint. For a PPR-PM, the moving platform can rotate arbitrarily about an axis undergoing a planar translation. This paper deals with the type synthesis of 3-DOF PPR-PMs. At first, virtual chains are introduced to represent the motion patterns of 3-DOF motions and relevant results from screw theory are recalled. A method is then proposed for the type synthesis of 3-DOF PPR-PMs. Using the proposed approach, the type synthesis of 3-DOF PPR-PMs is performed in three steps, namely, the type synthesis of legs for PPR-PKCs (parallel kinematic chains), the type synthesis of PPR-PKCs, and the selection of actuated joints of PPR-PMs. The three steps are dealt with in detail consequently. The characteristics of the proposed approach is that the type synthesis of legs for PPR-PKCs is reduced to the type synthesis of 3-DOF overconstrained single-loop kinematic chains and thus easy to perform. In addition to all the 3-DOF PPR-PKCs and 3-DOF PPR-PMs proposed in the literature, a number of new 3-DOF PPR-PKCs and 3-DOF PPR-PMs are identified. It is also found that there are no PPR-PMs with identical types of legs.

Topics: Screws , Chain , Manipulators
Commentary by Dr. Valentin Fuster
2004;():1261-1271. doi:10.1115/DETC2004-57473.

This paper presents a novel methodology for detecting collisions of cylindrically shaped rigid bodies moving in three dimensions. This algorithm uses line geometry and dual number algebra to exploit the geometry of cylindrical objects to facilitate the detection of collisions. First, the rigid bodies are modelled with infinite cylinders and a necessary condition for collision is evaluated. If the necessary condition is not satisfied then the two bodies do not collide. If the necessary condition is satisfied then a collision between the bodies may occur and we proceed to the next stage of the algorithm. In the second stage the bodies are modelled with finite cylinders and a definitive necessary and sufficient collision detection algorithm is employed. The result is a straight-forward and efficient means of detecting collisions of cylindrically shaped bodies moving in three dimensions. This methodology has applications in spatial mechanism design, robot motion planning, and workspace analyses of parallel kinematic machines such as Stewart-Gough platforms. A case study examining a spatial 4C mechanism for self collisions is included.

Commentary by Dr. Valentin Fuster
2004;():1273-1282. doi:10.1115/DETC2004-57476.

Interest in cooperative systems typically arises when certain tasks are either too complex to be performed by a single agent or when there are distinct benefits that accrue by cooperation of many simple agents. A quantitative examination of performance enhancement, due to the implementation of cooperation, is critical. In this paper, we focus on the development of a quantitative performance-analysis framework for a cooperative system with multiple wheeled mobile manipulators physically transporting a common payload. Each mobile manipulator module consists of a differentially-driven wheeled mobile robot with a mounted planar three-degree-of-freedom (d.o.f.) manipulator. A composite cooperative system is formed when a payload is placed at the end-effectors of multiple such modules. Such a system possesses the ability to change its relative configuration as well as accommodate relative positioning errors of the mobile bases. However, the combination of nonholonomic constraints due to the mobile bases, holonomic constraints due to the closed kinematic loops formed and the varying actuation of the joints within the cooperative system requires careful treatment for realizing the payload transport task. In this paper, we will analyze the cooperative composite system within a constrained mechanical system framework, by extending methods developed for treatment of articulated-closed-chain systems. Specifically, we will focus on the velocity-level kinematic modeling, while taking into account the nonholonomic/holonomic constraints and different joint-actuation schemes within the system. We then examine the applicability of a manipulability measure (isotropy index), to quantitatively analyze the system-level performance of the cooperative system, with these different joint-actuation schemes, with representative case-studies.

Topics: Mobile robots
Commentary by Dr. Valentin Fuster
2004;():1283-1291. doi:10.1115/DETC2004-57488.

The problem of self-reconfiguration of modular robots is discussed, and an algorithm for efficient parallel self-reconfiguration is presented. While much of the previous work has been focused on the lattice-type modular robots, this paper addresses the self-reconfiguration of chain-type robots. Relatively little attention has heretofore been given to this sub-problem, and of the existing work, none incorporates the kinematic limitations of real-life robots into the reconfiguration algorithm itself. The method presented here is based on understanding a robot’s physical “composition” using a graph-theoretic robot representation, and it sheds new light on self-reconfiguration of chain-type modular robots by incorporating elements of the robot kinematics as part of the criteria in choosing reconfiguration steps.

Topics: Robots , Algorithms , Chain
Commentary by Dr. Valentin Fuster
2004;():1293-1301. doi:10.1115/DETC2004-57494.

This paper systematically presents an effective algorithm for the dynamic balancing of multi-degree-of-freedom parallel mechanisms with multiple legs and the dynamic equivalence between point masses and arbitrary moving platforms. The mass and inertia of the moving platform are replaced by point masses located at the points of attachment of the legs to the platform and the mechanisms are balanced by considering each of the legs independently. The validity and feasibility of this algorithm is first verified both theoretically and using numerical simulations in ADAMS. Two, three and four point masses are respectively discussed for different cases. Finally, some reactionless planar and spatial multi-degree-of-freedom parallel mechanisms synthesized based on this algorithm are given.

Commentary by Dr. Valentin Fuster
2004;():1303-1310. doi:10.1115/DETC2004-57495.

While there is extensive literature available on parallel manipulators in general, there has been much less attention given to cable-driven parallel manipulators. In this paper, we address the problem of analyzing the reachable workspace using the tools of semi-definite programming. We build on earlier work [1, 2] done using similar techniques by deriving limiting conditions that allow us to compute analytic expressions for the boundary of the reachable workspace. We illustrate this computation for a planar parallel manipulator with four actuators.

Topics: Cables , Manipulators
Commentary by Dr. Valentin Fuster
2004;():1311-1322. doi:10.1115/DETC2004-57497.

The field of distributed-compliance mechanisms has seen significant work in developing suitable topology optimization tools for their design. These optimal design tools have grown out of the techniques of structural optimization. This paper will build on the previous work in topology optimization and compliant mechanism design by proposing an alternative design space parameterization through control points and adding another step to the process, that of subdivision. The control points assist a specific design to be represented as a solid model during the optimization process. The process of subdivision creates an additional number of control points that help smooth the surface (for example a C2 continuous surface depending on the method of subdivision chosen) creating a manufacturable design free of traditional numerical instabilities. Note that these additional control points do not add to the number of design parameters. This alternative parameterization and description as a solid model effectively and completely separates the design variables from the analysis variables during the optimization procedure. The motivation behind this work is to avoid several of the numerical instabilities that occur in topology optimization and to create an automated design tool from task definition to functional prototype created on a CNC or rapid-prototype machine. This paper will describe the complaint mechanism design process including subdivision and will demonstrate the procedure on several common examples.

Commentary by Dr. Valentin Fuster
2004;():1325-1338. doi:10.1115/DETC2004-57503.

Change-point mechanisms are shown to be significant in the design of surface micromachined MEMS. The design space of change-point mechanisms is derived for an arbitrary single loop change-point mechanism using a global and local approach. A function on the design space, the mechanism’s length, is constructed for fourbars. An inversion operator, a mapping from the design space to the design space, is also constructed for fourbars. The method for constructing the function and the operator is shown to be capable of extension to single loop change-point mechanisms with five or more links. The results give insight into design possibilities and limitations of change-point mechanisms.

Commentary by Dr. Valentin Fuster
2004;():1339-1348. doi:10.1115/DETC2004-57506.

Screw quantities provide geometric insight into three-dimensional mechanics modeled by rigid bodies and lumped parameters. Four distinct cases of time differentiation are examined by combining fixed and moving body derivatives (fundamental to rigid body mechanics) with material and local derivatives (fundamental to continuum mechanics). Three combinations always yield another screw quantity while the most common, the material derivative with respect to the fixed body, does not. Two fundamental formulations are examined with this last derivative, Euler’s Laws and the gravitational loading of an elastic system. By coincidence, the formulations appear screw-like when they are expressed at the center-of-mass but, in contrast to actual screw formulations, they do not retain invariant forms when expressed at arbitrary points.

Commentary by Dr. Valentin Fuster
2004;():1349-1355. doi:10.1115/DETC2004-57520.

This paper addresses the problem of obtaining models of nonlinear mechanisms for use in virtual prototyping systems. The objective of the described modeling procedure is to more accurately represent the feel of complex mechanisms in force feedback applications. The model structure selected for this study is a 2nd -order impedance model, with mass, damping, stiffness, and coulomb friction that vary with position. An automated testbed system is described, as well as a segmented least-squares algorithm for estimating position-varying model parameters. Baseline results are obtained for a system comprised of a mass, dashpot, and nonlinear (hardening) spring.

Topics: Modeling , Mechanisms
Commentary by Dr. Valentin Fuster
2004;():1357-1362. doi:10.1115/DETC2004-57523.

This paper studies the inverse static analysis of a planar parallel mechanism with compliant limbs. A known force and moment is applied to the moving platform, and it is required to determine the assembly configurations, or equilibrium points. Partial derivatives of the potential energy function yields the equilibrium conditions. The geometric and static constraints lead to a system of ten polynomials with ten unknowns. We use polynomial homotopy method to find that there are as many as 70 equilibrium configurations. Two examples with equilateral geometry are provided. We also examine the system behavior during a movement between selected equilibrium positions.

Commentary by Dr. Valentin Fuster
2004;():1363-1369. doi:10.1115/DETC2004-57524.

A spatial linkage is defined by a workpiece supported by one or more serial chains. We say that the spatial linkage can perform a given task when this task belongs to its workspace. A task can be specified as a discrete set of positions or as a set of curves or surfaces. Spatial serial chains can be synthesized to perform specified tasks, and combined to form more complicated linkages. Recently we developed Synthetica 1.0, a Java-based architecture for the computer-aided-design of spatial linkages that contains task definition and visualization, and linkage visualization, analysis and synthesis modules. In this paper, we present the new version, Synthetica 2.0. Among other additions, the new version includes task interpolation, a selection matrix, a generic synthesizer and several specialized synthesis procedures for constrained spatial linkages.

Commentary by Dr. Valentin Fuster
2004;():1371-1380. doi:10.1115/DETC2004-57531.

A foot path planning algorithm is presented for a robot with six limbs symmetrically located on the faces of its hexagonal body, enabling it to walk at a constant height with an alternating tripod gait. The symmetry results in near omni-directional locomotion capability, so the algorithm is formulated for walking in any direction and at any height. The approach is to determine the maximum length foot path through each limb’s workspace and then modify those foot paths based upon static stability analysis. The stability analysis is conducted in two phases to ensure stability without excessively reducing step length. Compared to an optimization approach, the algorithm yields foot paths within 9.1% of the maximal foot paths for all directions and heights. Unlike the optimization approach, the developed algorithm is computationally efficient enough to be implemented in realtime.

Topics: Robots
Commentary by Dr. Valentin Fuster
2004;():1381-1388. doi:10.1115/DETC2004-57541.

The performance of a contact-aided compliant mechanism that functions as a cycle doubler is studied in this paper via nonlinear finite element simulations. The topology of this mechanism was obtained from a systematic synthesis procedure and is reported elsewhere. Although the design was obtained for a quasi-static specification, the kinematic characteristics of the design suggest its ability to function adequately at low to moderate frequencies. The scalability of the design and its single-piece construction enable fabrication using different materials at various length scales. Therefore, it is possible to choose a scale and material combination that yields the frequency doubling action for various input frequencies. Explicit dynamic nonlinear finite element simulations are used to verify the functionality of the design at two different length scales: macro (device footprint of 289 sq. cm) corresponding to an input frequency of 20 Hz and meso (device footprint of a square of 14.3 sq. cm) corresponding to an input frequency of 1 kHz. Experiments on a macro scale prototype are used to validate the FE simulations for low frequencies.

Topics: Testing , Cycles
Commentary by Dr. Valentin Fuster
2004;():1389-1394. doi:10.1115/DETC2004-57549.

This paper uses line geometry to find an elegant solution to the kinematic registration problem involving reconstruction of a spatial displacement from data on three homologous points at two finitely separated positions of a rigid body. The bisecting linear line complex of two position theory in kinematics is used in combination with recent results from computational line geometry to present an elegant computational geometric method for the solution of this old problem. The results have applications in robotics, manufacturing, and biomedical imaging. The paper considers when minimal, over-determined, and perturbed sets of point data are given.

Topics: Geometry
Commentary by Dr. Valentin Fuster
2004;():1395-1406. doi:10.1115/DETC2004-57553.

An intelligent mechanism is a traditional mechanism that incorporates within it, elements of intelligence both at the design stage. Thepurpose is to increase the operational envelop of the original meachanism so that it can now exhibit some from of learning, adapting, or responding to changes in the environment or new operating conditions. Since mechanisms are usually designed for repetitive motions, the application of intelligence that makes specific use of repetitive characteristics will markedly improve the functionality of the mechanisms. This is particularly true if the mechanisms have been incorporated with such intelligence at the early stages of design. This article develops the concept of an intelligent mechanism through the application of Iterative Learning Control and Repetitive Control methodologies to several traditional mechanisms. These range from the simple slider-crank and four-bar linkage to timing belts.

Topics: Mechanisms
Commentary by Dr. Valentin Fuster
2004;():1407-1418. doi:10.1115/DETC2004-57554.

We present a new kinematic calibration algorithm for redundantly actuated parallel mechanisms. The calibration algorithm for a non-redundant case does not apply for a redundantly actuated parallel mechanism, because the angle error of the actuating joint varies with position and the geometrical constraint fails to be consistent. Such change of joint angle error comes from constraint torque variation with each kinematic pose. To calibrate a redundant parallel mechanism, one therefore has to consider constraint torque equilibrium and the relationship of constraint torque to torsional deflection, in addition to geometric constraint. In this paper, we develop the calibration algorithm for a redundantly actuated parallel mechanism using these three relationships, and formulate cost functions for an optimization algorithm. As a case study, we executed the calibration of a 2-degree of freedom (DOF) parallel mechanism with three actuators using the developed algorithm. Coordinate values of tool plate were measured using a laser ball bar and the actual kinematic parameters were identified with a new cost function of the optimization algorithm. Experimental results showed that the accuracy of the tool plate improved by 82% after kinematic calibration in a redundant actuation case.

Commentary by Dr. Valentin Fuster
2004;():1419-1426. doi:10.1115/DETC2004-57557.

The California Department of Transportation, like many such departments in the United States, is working on developing a new family of electronic sensing devices for the purpose of monitoring certain characteristics of road vehicles as they move along the highway. The devices currently under development are to be located overhead individual highway traffic lanes, from where they can have a clear “view” of ground vehicles. In order to deploy these devices, there is a need to develop the capability to safely and efficiently mount them above highway traffic lanes, using existing overhead bridges and sign structures as support structures. This paper presents a technical study of a universal support platform for these devices. The study discusses such issues as mobility, reliability, and resistance to environmental and other hazards. Results of tests conducted on a prototype are also presented.

Commentary by Dr. Valentin Fuster