0

33rd Mechanisms and Robotics Conference

2009;():3-10. doi:10.1115/DETC2009-86516.

This paper presents the design of a wearable upper arm exoskeleton that can be used to assist and train arm movements of stroke survivors or subjects with weak musculature. In the last ten years, a number of upper-arm training devices have emerged. However, due to their size and weight, their use is restricted to clinics and research laboratories. Our proposed wearable exoskeleton builds upon our extensive research experience in wire driven manipulators and design of rehabilitative systems. The exoskeleton consists of three main parts: (i) an inverted U-shaped cuff that rests on the shoulder, (ii) a cuff on the upper arm, and (iii) a cuff on the forearm. Six motors, mounted on the shoulder cuff, drive the cuffs on the upper arm and forearm, using cables. In order to assess the performance of this exoskeleton, prior to use on humans, a laboratory test-bed has been developed where this exoskeleton is mounted on a model skeleton, instrumented with sensors to measure joint angles and transmitted forces to the shoulder. This paper describes design details of the exoskeleton and addresses the key issue of parameter optimization to achieve useful workspace based on kinematic and kinetic models.

Topics: Cables , Design , Optimization
Commentary by Dr. Valentin Fuster
2009;():11-16. doi:10.1115/DETC2009-86713.

In this paper, modeling and control of a new cable-driven robot is presented. In this mechanism, the cable arrangement eliminates the rotational motions leaving the moving platform with three translational motion. The mechanism has potentials for large scale manipulation and robotics in harsh environments. In this article kinematics and dynamics models for the proposed cable-driven architecture are derived. Additionally, Feedback linearization under input constraints is used for the control of the robot. The control algorithm ensures the cable tensions are positive while minimizing the sum of all the torques exerted by the actuators. Finally, the implementation of the proposed method is demonstrated through simulation.

Commentary by Dr. Valentin Fuster
2009;():17-24. doi:10.1115/DETC2009-86718.

Wind tunnels are a standard tool to evaluate the air flow properties of aerodynamical vehicles in model scale. This is widely used to optimize the design of aircrafts and aircraft components. Additionally, the hydrodynamic properties of marine components like ship hulls or propulsion systems can be predicted. It is desirable to guide the models along defined trajectories during the tests to vary the angle of attack. Parallel wire robots were successfully used to perform airplane maneuvers in wind tunnels due to their good aerodynamical and mechanical properties. Compared to aircraft design, marine models are very heavy (up to 500kg). Thus, the positioning system must be very stiff to avoid vibrations. Additionally, fast maneuvers require powerful drives. Nevertheless, the positioning system should not influence the air flow. In this contribution, a novel design is presented. Additionally, a new realtime capable force distribution calculation method for parallel tensed systems is presented.

Commentary by Dr. Valentin Fuster
2009;():25-34. doi:10.1115/DETC2009-86720.

Wire robots consist of a movable end-effector which is connected to the machine frame by motor driven wires. Since wires can transmit only tension, positive wire forces have to be ensured. During workspace analysis, the wires forces need to be calculated. Discrete methods do not produce satisfying results, since intermediate points on the discrete calculation grids are neglected. Using intervals instead of points leads to reliable results. Formulating the analysis problem as a Constraint-Satisfaction-Problem (CSP) allows convenient transition to the synthesis problem, i.e. to find suitable designs for practical applications. In this paper, two synthesis approaches are employed: Design-to-Workspace (i.e. calculation of an optimal robot layout for a given workspace) and an extension called Design-to-Task (i.e. calculation of the optimal robot for a specific task). To solve these optimization problems, the paper presents approaches to combine the reliability and robustness of interval-based computations with the effectiveness of available optimizer implementations.

Topics: Robots , Wire , Design
Commentary by Dr. Valentin Fuster
2009;():35-43. doi:10.1115/DETC2009-87407.

Prestressable pin jointed structures are statically indeterminate but kinematically they can be either overdeterminate or indeterminate. Tensegrities are special cases of the latter class. Although, all prestressable structures have variable stiffness that can be controlled by prestress, it is only a subgroup of them such as tensegrities that make effective variable stiffness springs. An N-gon tensegrity prism is investigated as a basis for making a variable stiffness spring called Tensegrity Prism spring. The appropriate deflection direction is found and force-displacement and stiffness equations of a translational tensegrity prism spring with n bars are found. In addition, the main characteristics of a variable tensegrity prism spring and the methods that improve them are discussed.

Topics: Springs
Commentary by Dr. Valentin Fuster
2009;():45-53. doi:10.1115/DETC2009-87424.

This paper discusses the kinetostatic model of a symmetric cable-driven parallel manipulator that consists of eight cables driving a moving platform in planar translational motions. This analysis is the initial phase of the development of a large-workspace manipulator for warehousing applications. The paper discusses the kinematics, workspace, and stiffness of the manipulator. An analytical expression for the workspace is presented. To enhance the manipulator stiffness, the cable layout is optimized based on maximizing the lowest natural frequency of the moving platform in the workspace.

Topics: Cables , Manipulators
Commentary by Dr. Valentin Fuster
2009;():55-63. doi:10.1115/DETC2009-87457.

Cable-actuated parallel manipulators combine benefits of large workspaces, significant payload capacities and high stiffness by virtue of the cable actuation. However, redundant/surplus cables are required to overcome the unidirectional nature of forces exertable by cables. This leads to actuation redundancy which needs to be resolved in order to realize some of the benefits. We study the implication of using actuation redundancy to tailor the workspace (task space) stiffness of the cable robot system. Suitable trajectory tracking control schemes are developed that additionally achieve secondary goal of active stiffness control to improve disturbance rejection, under positive control input constraint We demonstrate the performance of these control schemes using a point-mass cable robot system modeled within a virtual prototyping (VP) implementation framework.

Commentary by Dr. Valentin Fuster
2009;():65-74. doi:10.1115/DETC2009-87542.

Flexible endoscopes are mainly used for diagnostics and performing simple therapeutic tasks inside human cavities but are now becoming the key instrument for the incisionless surgery known as natural orifice transluminal endoscopic surgery (NOTES). Since the current endoscope technology gives limited maneuverability, dexterity, and functionality, a number of new endoscope designs have been proposed. Due to miniaturization, conduit, and actuation simplicity, many of the new designs rely on cable-actuating mechanisms similar to the current technology. Basic kinematical and static analyses for this device have not appeared in the literature. In this paper the articulated section of a planar cable-driven endoscope is modeled as a serial robot. The kinematic and static analyses for single-jointed and multi-jointed endoscope structures are performed to relate tip motion to the controlling inputs. Pre-tensioning cables increases the endoscope stiffness and extends its range of operation.

Topics: Cables , Endoscopes
Commentary by Dr. Valentin Fuster
2009;():75-81. doi:10.1115/DETC2009-87597.

The main problem in cable-driven mechanisms is their tensionability, i.e. maintaining positive tension in all cables against any external load. Since the advent of these mechanisms, it was known that with one redundant cable, one can guarantee the tensionability of a rigid body driven by cables. However, the problem of tensionability in multibody systems driven by cables has remained almost intact. In a previous work by the authors, some necessary conditions on the number of cables and their possible distributions for tensionability of a general multibody were found. However, unlike the rigid body case, the sufficiency of these conditions for the tensionability of a multibody cannot be easily guaranteed. In this paper, we find the necessary and sufficient conditions for the tensionability of an arbitrary two-link multibody. Our approach investigates the equilibrium of such systems first by assuming that the cables can apply force in both directions (pull and push). Then, unidirectional force characteristic of the cables is incorporated and the associated conditions on the cables and their distributions are found. The heart of the method is based on a matrix called dependency matrix which carries the equilibrium properties of the system in a more compact form. Also, it has relatively clear connections to the geometry of the system which makes it easy to use.

Commentary by Dr. Valentin Fuster
2009;():83-91. doi:10.1115/DETC2009-87677.

This paper addresses the forward and inverse kinematics of payloads carried by aerial robots. We address the cases with one, two, and three aerial robots and derive the kinematics and conditions for stable static equilibrium. For the case with one or two robots, we can establish the maximum number of equilibrium positions. The three-robot case is seen to be much harder primarily because of the non-negative tension constraints. We restrict the set of possible solutions to the forward and inverse problems by considering the equations of static equilibrium and kinematic constraints. Analytic and numeric methods to determine equilibrium configurations and stability are presented.

Commentary by Dr. Valentin Fuster
2009;():93-100. doi:10.1115/DETC2009-87705.

This paper presents a kinetostatic analysis of a class of hybrid cable-driven parallel manipulators consisting of cables and a rigid-link serial leg for lower-degree-of-freedom tasks. The serial leg provides the task constraints and the cables provide the required actuation to move the platform in the task space. Having a rigid-link serial leg providing the required constraints for the moving platform in lower-degree-of-freedom applications reduces the number of cables required to support the moving platform in the task space. As a result, the manipulator can be designed with fewer cables and actuators. In order for the manipulator to balance arbitrary applied wrenches, redundant actuation is required. This redundant actuation can be obtained from a redundant actuator mounted on one (or more) serial-leg joint(s). Conditions for analyzing the capability of the redundant actuation to generate cable tension are explained in the paper. A discussion is also given on the optimization of the cable layout and the redundant actuator forces to minimize the tension generated in the cables.

Topics: Cables , Manipulators
Commentary by Dr. Valentin Fuster
2009;():101-110. doi:10.1115/DETC2009-86075.

This paper presents a novel class of 3-DOF translational compliant parallel manipulators (CPMs) based on flexure motion. The analytic mathematic modeling of CPMs is first developed. The analysis of CPMs is then implemented. It is shown that the proposed CPMs have many characteristics such as large range of motion, negligible cross-axis coupling, actuator complete isolation, and no loss motion and no rotational yaw. The inverse relationships of force-displacement of the 3-DOF CPM are further derived to calculate the input forces required for generating a specified path. In addition, the 3-DOF CPM can also be turned into a 2-DOF CPM. This work lays the foundation for the development of new spatial CPMs based on flexure motions for applications such as ultra precision manipulation.

Commentary by Dr. Valentin Fuster
2009;():111-121. doi:10.1115/DETC2009-86476.

Dielectric Elastomer Actuators (DEA) has great potential for low cost, high performance robotic and mechatronic devices. However, the reliability of these actuators remains an important issue when used in continuous strain applications. To improve actuators reliability, DEAs can be used in a binary or bistable manner where actuators flip between two stable positions, thus maintaining one of two equilibrium states without any electrical energy input. This paper presents an antagonistic bistable DEA concept using a single, planar polymer film that can lead to compact high force multilayered actuators. The system is made bistable by the addition of carbon fiber leaf springs designed to maximize actuator strain output. The strong viscoelastic nature of the chosen polymer film significantly affects the system’s output force and is accounted for in the Bergstrom-Boyce material model. The model shows good agreement with experimental stress relaxation curves and is used to set the leaf springs’ force curve. Experimental results have shown that the acrylic polymer film’s (VHB 4905) strong viscoelastic nature limits the actuator speed at ∼ 0.9 mm/s; at higher speeds, the leaf springs cannot be matched with the proposed concept. The study also demonstrates that the proposed antagonistic actuator configuration is an interesting solution to provide reliable bistable actuation for compact structures and that developing polymer films with low viscoelasticity is key for optimal performance.

Commentary by Dr. Valentin Fuster
2009;():123-132. doi:10.1115/DETC2009-86480.

Affecting 1 out of 8 subjects in the U.S., prostate cancer is the most common form of cancer in men. Current medical procedures could be improved by the development of an MRI compatible (Magnetic Resonance Imaging) needle manipulator system, to precisely reach small tumors (<5 mm) inside the prostate. This paper presents and analyzes the potential of such a needle manipulator concept, based on hyper-redundant binary air muscles, all controlled by MRI compatible valves (e.g. piezoelectric or dielectric elastomer actuators). The proposed manipulator uses 12 polymer air muscles, each driven by 2 different actuation pressures, offering a total of 4096 (212 ) discrete needle positions. Based on a hyperelastic continuum mechanics air muscle model, a theoretical manipulator design is used to evaluate clinically-relevant design metrics, such as size, stiffness, workspace, accuracy and sensitivity. In this model, the manipulator’s equilibrium configuration (for a given set of input pressures and applied forces) is found by minimizing the system’s potential energy. The model capability is verified experimentally by a one degree of freedom (DOF) prototype. Simulation results show that the proposed elastically averaged air muscle concept can meet all design requirements. In particular, the needle workspace of about 70 mm by 80 mm entirely covers the prostate area, where targets are accurately reachable within 0.7 mm. Also, the pneumatic actuators can generate high forces leading to a system stiffness of ∼4.6 N/mm at the needle tip. Such stiffness can adequately sustain the needle during insertion with minimal deflection to guaranty accurate positioning.

Commentary by Dr. Valentin Fuster
2009;():133-139. doi:10.1115/DETC2009-86509.

This paper presents a closed-form expression of the mass matrix of the right circular hinges which are widely used in compliant mechanism systems. The flexure hinge is treated as a two nodes element, in which there are three degrees of freedom at each node. The cosine functions with high-order terms are introduced, and the closed-form expression for each element of the mass matrix of the single-axis right circular flexure hinge is obtained. Based on the closed-form expression, the dynamic equations of motion of a piezodriven 3-DOF compliant precision micro-positioning stage are developed. The lower order modal frequencies are experimentally identified and the dynamic analysis model is proved to be feasible. Dynamic and static output comparisons show that the dynamic analysis is highly necessary for a precision micro-positioning stage. The closed-form expression of the mass matrix of the right circular hinge made it more convenient for dynamic analysis of the compliant mechanisms by using finite element approach.

Commentary by Dr. Valentin Fuster
2009;():141-149. doi:10.1115/DETC2009-86637.

Three-dimensional multilayer wide curves are spatial curves with variable cross sections and multiple materials. This paper introduces a geometric optimization method for spatial multimaterial compliant mechanisms and structures by using three-dimensional multilayer wide curves. In this paper, every multimaterial connection is represented by a three-dimensional multilayer wide curve and the whole spatial multimaterial compliant mechanism or structure is modeled as a set of connected three-dimensional multilayer wide curves. The geometric optimization of a spatial multimaterial compliant mechanism or structure is considered as the optimal selection of control parameters of the corresponding three-dimensional multilayer wide curves. The deformation and performance of spatial multimaterial compliant mechanisms and structures are evaluated by the isoparametric degenerate-continuum nonlinear finite element procedure. The problem-dependent objectives are optimized and the practical constraints are imposed during the optimization process. The optimization problem is solved by the MATLAB constrained nonlinear programming algorithm. The effectiveness of the proposed geometric optimization procedure is verified by the demonstrated examples.

Commentary by Dr. Valentin Fuster
2009;():151-161. doi:10.1115/DETC2009-86684.

This paper presents a screw theory based approach for the type synthesis of compliant mechanisms with flexures. We provide a systematic formulation of the constraint-based approach which has been mainly developed by precision engineering experts in designing precision machines. The two fundamental concepts in the constraint-based approach, constraint and freedom, can be represented mathematically by a wrench and a twist in screw theory. For example, an ideal wire flexure applies a translational constraint which can be described a wrench of pure force. As a result, the design rules of the constraint-based approach can be systematically formulated in the format of screws and screw systems. Two major problems in compliant mechanism design, constraint pattern analysis and constraint pattern design are discussed with examples in details. This innovative method paves the way for introducing computational techniques into the constraint-based approach for the synthesis and analysis of compliant mechanisms.

Commentary by Dr. Valentin Fuster
2009;():163-170. doi:10.1115/DETC2009-86694.

The objective of this investigation is to present a concept as well as mathematical modeling and finite element modeling of a statically balanced compliant laparoscopic grasper. To obtain force feedback, the positive stiffness of the compliant grasper was statically balanced by a negative-stiffness compensation mechanism. The negative stiffness has been produced by pairs of pre-stressed initially-curved pinned-pinned beams out of linear elastic material, arranged perpendicular to the link driving the grasper. First, the conceptual design is explained. Subsequently, its behavior is mathematically formulated and then finite element modeling is implemented using a commercially available finite element modeling package. Finally, a stress-optimized design of a negative-stiffness compensation mechanism and the effect of parameter changes on the accuracy are obtained. The results illustrate the efficiency of the applied analysis methods for the case of statically balancing the laparoscopic grasper. It also demonstrates the efficiency of the balancer concept. The proposed procedure is found to be convenient for this set of problems, and can probably be applied to other similar practical problems.

Topics: Modeling
Commentary by Dr. Valentin Fuster
2009;():171-181. doi:10.1115/DETC2009-86712.

The honeycomb based discretization shows promise in yielding checkerboard and point flexure free optimal solutions to various topology design problems. The mesh-free material mask overlay method further promises unadulterated “black and white” optimal solutions as opposed to schemes where material is interpolated between the “void” and “filled” states in a cell [26]. Here, we propose improvements in the material mask overlay method by judiciously choosing the number of masks during a sequence of sub-searches for the optimal solution. We use an alternative mutation based zero order search which allows the use of a small population of solutions and also maintains diversity between them. Thus, multiple solutions can be simultaneously obtained for non-convex topology optimization formulations. We solve two classical problems each on optimal stiff structures and compliant mechanisms to illustrate pathology free, “black and white” topology synthesis.

Commentary by Dr. Valentin Fuster
2009;():183-191. doi:10.1115/DETC2009-86754.

The objective of this paper is to present a comparative analysis for large deflections of a cantilever beam under free end point load. pseudo rigid body model (PRBM), non-linear beam theory numerically solved with integration (NLBT-NUM), linear beam theory (LBT), finite element modeling (FEM) using an available commercially FEM package, non-linear beam theory solved with direct nonlinear solution (NLBT-DNS) and experimental evaluation (EXP), have been implemented. For the purpose of comparison, the relation between the displacements, rotating angle of the tip and applied force were calculated and shown graphically. The accuracy of the path of the tip as a function of the force is compared with the NLBT-NUM, which is taken as a reference. In addition, computation times and implementation convenience were recorded. In the case of a perpendicular load, the PRBM is accurate and has little computation time. The NLBT-NUM, NLBT-DNS and FEM analysis are accurate, but the computation time is longer. The NLBT-DNS has been introduced for the first time and provides semi-exact closed form solutions for both horizontal and vertical position. In case of a non-perpendicular load, the NLBT-NUM and FEM analysis are the only accurate methods while computation time is less for the numerical solution. In conclusion, the PRBM and the FEM are recommended for the cases of perpendicular load and non-perpendicular load respectively. Finally, it can be concluded that the more accurate methods take more computation time, and that the accuracy is affected by load cases.

Commentary by Dr. Valentin Fuster
2009;():193-214. doi:10.1115/DETC2009-86845.

Compliant mechanisms are rapidly gaining importance, yet their design remains challenging. A great variety of methods are being developed as it is reported in a growing stream of publications. However, so far no review of this body of literature is available. This paper provides a comprehensive and conceptual overview of the main notions behind the most relevant design methods for compliant mechanisms. Rigid-Body-Replacement methods including the Pseudo-Rigid-Body model and the FACT approach are covered, as well as Building Block approaches. In addition an introduction and explanation on Topology Optimization and Shape Optimization is provided, including their most common parameterizations and formulations. This work aims to serve as an introduction into compliant mechanism design methods and as a reference work for more experienced scholars and professionals. It is intended to be a starting point for the exploration of the literature, as well as a guide to specific papers about a particular design problem one may have. For this reason, the paper presents the methods in a wide perspective, emphasizing the conceptual ideas behind every method and refers to literature for details and advanced features.

Commentary by Dr. Valentin Fuster
2009;():215-222. doi:10.1115/DETC2009-86847.

Passively compliant legs have been instrumental in the development of dynamically running legged robots. Having properly tuned leg springs is essential for stable, robust and energetically efficient running at high speeds. Recent simulation studies indicate that having variable stiffness legs, as animals do, can significantly improve the speed and stability of these robots in changing environmental conditions. However, to date, the mechanical complexities of designing usefully robust tunable passive compliance into legs has precluded their implementation on practical running robots. This paper describes a new design of a “structurally controlled variable stiffness” leg for a hexapedal running robot. This new leg improves on previous designs’ performance and enables runtime modification of leg stiffness in a small, lightweight, and rugged package. Modeling and leg test experiments are presented that characterize the improvement in stiffness range, energy storage, and dynamic coupling properties of these legs. We conclude that this variable stiffness leg design is now ready for implementation and testing on a dynamical running robot.

Commentary by Dr. Valentin Fuster
2009;():223-232. doi:10.1115/DETC2009-86850.

A unified procedure for the synthesis of planar linkages that may take the form of rigid body, fully compliant or partially compliant mechanisms is presented. The procedure automates the selection of mechanism topology as characterized by the number and connectivity of the links as well as the nature of the connections between the links, the mechanism shape as characterized by the shapes of the individual links, and the mechanism dimensions which include the locations of the joints and the cross-sectional dimensions of the links. The synthesis task is posed as an optimization problem and is solved by a hybrid, elite-preserving genetic algorithm. Three examples of compact mechanisms that trace different non-smooth paths in response to a single, monotonic and bounded force input are used to illustrate the synthesis capability of the procedure. Prototypes of the designs are built and tested to verify their performance.

Topics: Linkages
Commentary by Dr. Valentin Fuster
2009;():233-240. doi:10.1115/DETC2009-86901.

A compliant suspension linkage based on the Peaucellier mechanism is presented. The suspension uses large-deflection viscoelastic beams to achieve straight-line motion and to provide energy dissipation. Kinematics and force analysis of the linkage are presented. In preparing to simulate the system dynamics, it was noticed that no adaptation of the pseudo-rigid body model for viscoelastic beams had been previously presented. Therefore, a new general approach for modeling viscoelastic, large-deflection beams in compliant mechanisms is described within the context of the pseudo-rigid-body model. This method is applied in simulation of the Peaucellier-based compliant suspension under a variety of input conditions.

Commentary by Dr. Valentin Fuster
2009;():241-248. doi:10.1115/DETC2009-86990.

A compliant add-on mechanism for polycentric prosthetic knees was designed to provide two additional features beyond those already provided by the knee, the additional features being a stable equilibrium position when the knee is bent for the sitting posture and a moment-rotation profile that helps prevent excessive heel rise. The mechanism, dubbed the Bistable Compliant Extension Aid (BCEA), was developed and analyzed using finite-element-analysis (FEA) software. The BCEA was shown to satisfy the design requirements, prevention of excessive heel rise and providing a stable sitting position, based on its reaction moments for knee flexions ranging between 0 and 90 degrees.

Commentary by Dr. Valentin Fuster
2009;():249-257. doi:10.1115/DETC2009-87067.

This paper discusses the possibility of designing self-adaptive fingers with compliant joints capable of pinch preshaping, i.e. seizing an object with only the distal phalanges. Following a discussion on how this can be achieved with non-compliant fingers, the authors presents a new technique that can be applied to compliant fingers. However, this method cannot be followed exactly but, if carefully analyzed, sufficiently approximated providing that certain compliant joints of the linkage have a significantly larger stiffness than the others. The examples of two designs, one two-phalanx finger and a similar design with three phalanges, are presented and discussed. Finally, the behavior of a gripper built using the method presented in the paper is verified using both a dynamic software package and a finite element analysis (FEA) software.

Commentary by Dr. Valentin Fuster
2009;():259-268. doi:10.1115/DETC2009-87248.

A significant amount of research has been conducted in developing optimal synthesis techniques for compliant mechanisms with the expectation that distributed devices would result from the continuum design domain. However, it is commonly noted that much of this work has resulted in mechanisms that show localized rather than distributed compliance. This behavior has been attributed to a variety of sources including numerical discrepancies in the model, objective function formulation, and design parameterizations. In this paper, the nature of compliance distribution over particular objective function formulations and design parameterization are further considered in the absence of numerical or resolution issues. The intent is to better understand the behavior of the objective function over multidimensional subsets of the design space that include a direct measure for distribution of compliance. The approach is based on a simple, representative compliant mechanism formed as a segmented beam model. This mechanism is considered to be representative of compliant mechanism behavior in systems where elastic deformation is dominated by bending. Closed-form solutions for the elastic response of this representative mechanism are presented and parametric studies of the response of traditional objectives over subsets of the design space are conducted. The results show that in the absence of numeric artifacts, mechanism efficiencies are improved as mechanisms tend toward lumped compliance when single objectives are considered on mechanisms dominated by bending. However, when more than one objective is deemed important in the design, there exist preferred regions of the workspace, not necessarily in a lumped region, that depend largely on the interaction of the multiple objectives. Of these preferred regions, one lies in a moderately lumped region (h2 /h1 ≈ 0.2) and one in a distributed region (h2 /h1 ≈ 0.7). The designs in these regions reveal a higher viability in simultaneously satisfying the multiple objectives. This result is based on a visualization of the design space based on measuring the correlation of a multiple objectives over the design space. The results demonstrate several of the factors which contribute to this behavior, and provide an initial measure of the importance of each. Finally, suggestions are provided based on these results that can be used to improve the optimization process if the desire is to achieve distributed compliance.

Commentary by Dr. Valentin Fuster
2009;():269-276. doi:10.1115/DETC2009-87290.

Limited resources are currently available to assist engineers in implementing compliant members into mechanical designs. As a result, engineers often have little to no direction incorporating compliance, though compliant mechanisms have characteristics that prove to be more advantageous in some circumstances. This paper proposes a classification scheme for compliant elements and mechanisms that can be used to compose a reference source intended to help engineers find existing compliant designs they can incorporate in their own designs. The classification scheme decomposes compliant mechanisms and elements of compliant mechanisms according to their respective functions.

Commentary by Dr. Valentin Fuster
2009;():277-287. doi:10.1115/DETC2009-87331.

This paper explores the concept of reconfigurable compliant mechanisms. We define these to be fully or partially compliant mechanisms whose performance can be modified after they have been fabricated. Specifically, we are interested in the nature and extent of in situ reconfigurability in compliant mechanisms. In other words, we seek to understand the range of performance that can be achieved by these mechanisms without requiring significant reassembly. The material properties such as the storage modulus of a newly studied class of materials — shape memory polymers — vary by over an order of magnitude over a temperature range of 20 – 50 C. These polymers also allow the fixing of moderate to large strains (20 – 75%) experienced at high temperatures for extended periods of time, while retaining the ability to remember their original shape when reheated to the same high temperatures. These two properties make shape memory polymers a natural candidate for the fabrication of reconfigurable compliant mechanisms. We explore various means for introducing reconfigurability in compliant mechanisms, and from these, select a subset that is suitable for in situ reconfiguration. Quasi-static nonlinear finite element simulations are used to study the change in performance due to reconfiguration of four fully compliant mechanisms made of a shape memory polymer. Preliminary results indicate that noticeable qualitative and quantitative changes in performance can be achieved by these mechanisms.

Commentary by Dr. Valentin Fuster
2009;():289-297. doi:10.1115/DETC2009-87334.

In the last 20 years the interest in compliant mechanisms has experienced a rebirth with a continuous growth in research with the consequently increase of literature in this area. This work presents a classification scheme to organize literature on compliant mechanisms. The literature classification is based on the design methodology for conventional mechanisms; which is proposed as a compatible design process also for compliant mechanisms by assuming some limitations and extensions in the employed terms. As an example, 120 relevant research documents in compliant mechanisms have been classified. The work done on topology optimization was excluded due to its extension.

Commentary by Dr. Valentin Fuster
2009;():299-312. doi:10.1115/DETC2009-87445.

In this section we implement a characterization based on eigen-twists and eigen-wrenches for the deformation of a compliant mechanism at a given point of interest. For 2-D mechanisms, this involves characterizing the compliance matrix at a unique point called the center of elasticity. At the center of elasticity, the translation and rotational compliances are decoupled. We give an intuitive graphical understanding of compliance at this point by representing the translational compliance as an ellipse and the coupling between the translational and rotational parameters as vectors (Coupling vectors). This representation gives us an intuitive understanding of series and parallel combination of building blocks. We obtain a parametric variation of these quantities for a compliant dyad building block, and show with examples how a mechanism can be synthesized by a combination of building blocks to obtain desired deformation requirements. We also propose a combination of series and parallel concatenation to achieve more than one specification simultaneously. Such a characterization can be extended to synthesize involving multiple ports.

Commentary by Dr. Valentin Fuster
2009;():313-323. doi:10.1115/DETC2009-87451.

Particularly when high-fidelity force feedback is required, such as in surgical forceps, the energy loss between input and output in compliant mechanisms is undesired. To restore the force feedback, the principle of static balancing can be applied, where a balancing segment with a negative stiffness is added to a compliant mechanism. Currently there are no mature methods for the design of statically balanced compliant mechanisms (SBCM). The goal of this paper is to investigate the possibility of extending the Building Block Approach for the design of statically balanced compliant mechanisms. To this end, the Building Block Approach is extended with negative stiffness balancing building blocks that can be added to a designed compliant mechanism. To demonstrate the feasibility of the method, a statically balanced compliant gripper was designed by this Extended Building Block Approach. The maximum operating force of the unbalanced gripper of 3.5 N was reduced to −1 N for the balanced gripper. Thus, the gripper is slightly overbalanced. The gripper example demonstrates the functionality of the proposed method; the input-output stiffness of a compliant mechanism can be severely reduced by a balancing segment.

Commentary by Dr. Valentin Fuster
2009;():325-333. doi:10.1115/DETC2009-87557.

A novel fabrication process and design optimization method for a mesoscale forceps is presented. This work is part of a larger research effort to design and fabricate nanoparticulate enabled surgical instruments using an iterative fabrication-design technique. The current paper focuses on the fabrication of thick (∼hundreds of microns) two dimensional parts with large aspect ratios (length/width > 20). The paper also describes an optimization method that accounts for manufacturing requirements and material strength. The process begins with the fabrication of an array of molds on refractory substrates using a modified UV lithography technique. In parallel, engineered ceramic nanocolloidal slurries are prepared for gel-casting into the molds. Mold infiltration takes place via a squeegee technique adapted from screen printing with excess slurry removed using an ethanol bath. Finally, the photoresist molds are removed via pyrolysis, and ceramic parts sintered to full density. Employing this manufacturing technique for the compliant micro forceps design is advantageous because a large number of parts can be produced with large aspect ratios, sharp edges (∼ 1 μm), and a resolution of 2 μm. An optimization algorithm, using ANSYS optimization module, is formulated to determine the effect of dimensional parameters and material strength on the optimal design and predicted performance of the compliant meso forceps. Three ultimate strength values are separately implemented as a stress constraint in our optimization problem. Results conclude that our manufacturing process is capable of producing meso scale forceps considering the anticipated ultimate strength at this scale.

Topics: Manufacturing , Design
Commentary by Dr. Valentin Fuster
2009;():335-343. doi:10.1115/DETC2009-87598.

Soft robotic manipulators are continuum robots made of soft materials that undergo continuous elastic deformation and produce motion with a smooth backbone curve. These manipulators offer significant advantages over traditional manipulators due to their ability to conform to their surroundings, move with dexterity and manipulate objects of widely varying size using whole arm manipulation. Theoretically, soft robots have infinite degrees of freedom (dof), but the number of sensors and actuators are limited. Many dofs of soft robots are not directly observable and/or controllable, complicating shape sensing and controlling. In this paper, we present two methods of shape sensing for soft robotic manipulators based on a geometrically exact mechanical model. The first method use s load cells mounted at the base of the manipulator and the second method makes use of cable encoders running through the length of the manipulator. Simulation results show an endpoint localization error of less than 3% of manipulator length.

Topics: Manipulators , Shapes
Commentary by Dr. Valentin Fuster
2009;():345-358. doi:10.1115/DETC2009-87808.

To utilize beam flexures in constraint-based flexure mechanism design, it is important to develop a qualitative and quantitative understanding of their constraint characteristics in terms of stiffness and error motions. This paper provides a highly generalized yet accurate closed-form load-displacement model for two-dimensional beam flexures, taking into account the nonlinearities arising from load equilibrium applied in the deformed configuration. In particular, stiffness and error motions are parametrically quantified in terms of elastic, load-stiffening, kinematic, and elastokinematic effects. The proposed beam constraint model incorporates any general loading conditions, boundary conditions, initial curvature, and beam shape. The accuracy and effectiveness of the proposed beam constraint model is verified extensively by non-linear Finite Elements Analysis.

Commentary by Dr. Valentin Fuster
2009;():359-365. doi:10.1115/DETC2009-86097.

Using the Implicit Constraint Approach (ICA), a general procedure is presented for analyzing mechanism or robotic devices of arbitrary topology and complexity. The ICA allows for a compact formulation with the unique feature of automatically including real joint stiffness and damping characteristics. This formulation is the core of a computer program, ICAP, that directly provides all the body kinematics and joint reaction forces of any planar mechanism. An interactive graphic user interface to the program’s physics engine provides a simple method for describing the mechanical system and supporting the presentation of the analysis results. ICAP was developed to be used in the MATLAB environment, but could be modified for stand-alone use. Numerical examples are presented for a truck full (left and right side) front suspension and a Watt six-bar linkage.

Commentary by Dr. Valentin Fuster
2009;():367-376. doi:10.1115/DETC2009-86230.

This paper discusses the utility of using simple stiffness and vibrations models, based on the Jacobian matrix of a manipulator and only the rigidity of the actuators, whenever its geometry is optimised. In many works, these simplified models are used to propose optimal design of robots. However, the elasticity of the drive system is often negligible in comparison with the elasticity of the elements, especially in applications where high dynamic performances are needed. Therefore, the use of such a simplified model may lead to the creation of robots with long legs, which will be submitted to large bending and twisting deformations. This paper presents an example of manipulator for which it is preferable to use a complete stiffness or vibration model to obtain the most suitable design and shows that the use of simplified models can lead to mechanisms with poorer rigidity.

Topics: Deformation , Robots , Design
Commentary by Dr. Valentin Fuster
2009;():377-383. doi:10.1115/DETC2009-86359.

To achieve high positioning accuracy, short transportation time and small swing angle of an overhead crane system, both motion and stabilization control strategies are required. Under different loading condition, an over-head crane system may experience a range of parameter variation. Consequently, to enhance both efficiency and safety and to extend the system application to other engineering fields, a robust controller is required. In this paper, an over-head crane is modeled as a nonlinear two degrees of freedom system with parametric uncertainties. The force exerted on trolley is considered as the control input of the system. Applying an inversed-base controller based on μ-analysis, robust stability and nominal performance are achieved for the perturbed plants. Using μ-synthesis with DK-iteration algorithm, an optimal robust controller is designed which guarantees the robust performance too.

Topics: Cranes
Commentary by Dr. Valentin Fuster
2009;():385-389. doi:10.1115/DETC2009-87054.

This paper presents a partial decoupling method for the well-known matrix method commonly in use for the linear inverse dynamics problem of planar mechanisms. Any mechanism with a dyad of binary links can benefit from the proposed method, which involves extracting submatrices by looking for sufficient columns of zeros such that a reduced problem may first be solved with an equal number of scalar equations and unknowns. Then some equal-and-opposite internal joint forces are transferred to the remaining FBDs and the solution proceeds until the input link is solved in a decoupled manner. The method leads to significant reductions in computational cost for common planar mechanisms. The kinematics & dynamics textbooks have overlooked this partial decoupling in their presentations of the matrix method for inverse dynamics.

Commentary by Dr. Valentin Fuster
2009;():391-398. doi:10.1115/DETC2009-87131.

Electromagnetic valve actuation systems for automotive combustion engines must provide extremely fast valve motion when the engine speed is high, but they also need to ensure low valve seating velocities during engine idle. These two constraints are difficult to combine in conventional spring assisted electromagnetic valve actuation devices that operate at a fixed resonance frequency. This paper focuses on a mechanism with two distinct configurations for low and high speed engine operations respectively. The mechanism is based on two pivoting cams. The synthesis of the cam profile ultimately determines the performance of the actuation system. An algorithm is presented that provides a time optimum cam profile for the high speed cam. The low speed cam is designed to allow for servo control of the valve system. A control scheme that aims to minimize electric losses in the drive system is also introduced. Both the cam synthesis algorithms and the control algorithm are applied to a typical automotive valve train and a digital simulation is used to validate the effectiveness of the mechanical cam design and control scheme.

Topics: Valves
Commentary by Dr. Valentin Fuster
2009;():399-409. doi:10.1115/DETC2009-87529.

This paper presents a methodology for determining the static joint torques of a digital human model considering balance for both standing and seating tasks. An alternative and efficient formulation of the Zero-Moment Point (ZMP) for static balance and the approximated (ground/seat) support reaction forces/moments are derived from the resultant reaction loads, which includes the gravity and externally applied loads. The proposed method can be used for both standing and seating tasks for assessing the stability/balance of the posture. The proposed formulation can be beneficial to physics-based simulation of humanoids and human models. Also, the calculated joint torques can be considered as an indicator to assess the risks of injuries when human models perform various tasks.

Commentary by Dr. Valentin Fuster
2009;():411-420. doi:10.1115/DETC2009-87575.

This work presents a comparison between two actuation methods for planar and spherical four-bar mechanisms. The first is actuation by a torque applied at either of the fixed pivots. The second actuation method uses a linear actuator connected between ground and the coupler. For any four position task, planar or spherical, a one parameter set of dyads is found that may be used to guide the body through the four positions. Any two of these dyads, when coupled, define a potential four-bar solution to the task. A sampling across the set of all possible mechanisms that solve the four position task may be compared by analyzing the internal static loads of the four-bar mechanisms. The comparison was conducted to determine if coupler-driven four-bars have reduced internal loading when compared to torque-driven mechanisms. Four position tasks were used for comparison of mechanisms designed for the same task, under the assumption that the optimal torque-driven mechanism would have a different set of kinematic parameters than the optimal coupler-driven mechanism.

Topics: Torque
Commentary by Dr. Valentin Fuster
2009;():421-432. doi:10.1115/DETC2009-86039.

This paper deals with the kinematic analysis and design optimization of disc cam mechanisms with eccentric translating roller followers. The objective function considered in the present work takes into account the three major parameters that influence the final cam size, namely the base circle radius of the cam, the radius of the roller and the offset of the follower. Furthermore, geometric constraints related to the maximum pressure angle and minimum radius of curvature are included to ensure good working conditions of the system. Finally, an application example is offered.

Commentary by Dr. Valentin Fuster
2009;():433-439. doi:10.1115/DETC2009-86073.

Based on the edge-based array representation of loops in the topological graphs of kinematic chains, this paper first proposes three arithmetic operations of loops. Then the concept of the independent loop set as well as it determination rules is introduced, and a new structure decomposition algorithm of kinematic chains is presented. Based on the algorithm, an automatic and efficient method for rigid subchain detection and driving pair selection of kinematic chains is proposed. Finally, an index is proposed to assess computation complexity of kinematic analysis with respect to different driving pair selections.

Commentary by Dr. Valentin Fuster
2009;():441-451. doi:10.1115/DETC2009-86105.

Based on the Position and orientation characteristic (POC) equation of serial mechanisms proposed by the author, this paper presents a novel systematic approach for structure synthesis of rank-degenerated serial mechanisms and over-constrained single-loop kinematic chains (KCs) (excluding the Bennett mechanism etc). This approach is totally different from the approaches based on the screw theory and based on the displacement subgroup, and only simple mathematical tools (vector algebra, etc.) are used. Using this approach, the structure types of serial mechanisms with the specified ranks and the specified degree of freedom (DOF) are synthesized firstly. After that, using the structure types of the obtained serial mechanisms, structure types of over-constrained single-loop KCs with the specified ranks and the specified DOF can be generated in a straightforward way. The structure types of the obtained serial mechanisms can be used as branches of parallel mechanisms. The structure types and the ranks of the obtained over-constrained single-loop KCs can be used to calculate the DOF of multi-loop mechanisms. In fact, the systematic approach proposed in this paper is a key component of the systematic approach for structure synthesis of parallel mechanisms.

Topics: Chain , Mechanisms
Commentary by Dr. Valentin Fuster
2009;():453-462. doi:10.1115/DETC2009-86107.

Basic principles and main characteristics of three approaches for structure synthesis of robot mechanisms (the screw theory-based approach, the displacement subgroup-based approach and the approach based on position and orientation characteristic (in short, POC) ) are studied and compared in this paper. The comparison deals with the mathematical tools, the symbolic representation of mechanism topological structure, the mathematical representation of POC of the output motion link with respect to the frame link, the basic equations for structure synthesis of serial and parallel mechanisms and relevant operation rules, and the characteristics of their synthesized mechanisms, etc. This comparative study shows that the POC-based approach is totally different from the other two approaches: (1) the POC-based approach requires only simple mathematical tools (such as vector algebra, set theory, etc), (2) the POC-based approach is conceptually simpler and therefore easier to understand and to use, and (3) the POC-based approach is more general.

Topics: Robots , Mechanisms
Commentary by Dr. Valentin Fuster
2009;():463-473. doi:10.1115/DETC2009-86125.

The fixed pivots of a planar 4R linkage that can achieve four design positions are constrained to a center-point curve. The curve is a circular cubic function and plots can take one of five different forms. The center-point curve can be generated with a compatibility linkage obtained from an opposite pole quadrilateral of the four design positions. This paper presents a method to identify design positions that generate distinctive shapes of the center-point curves. The form of the center-point curve is dependent on whether the shape of the opposite pole quadrilateral is an open or closed form of a rhombus, kite, parallelogram, or when the sum of two sides equals the other two. Interesting cases of three and five position synthesis are also explored. Four and five position cases are generated that have center points at infinity allowing a PR dyad with line of slide in any direction to achieve the design positions. Further, a center-point curve for five specific design positions is revealed.

Commentary by Dr. Valentin Fuster
2009;():475-483. doi:10.1115/DETC2009-86184.

Theoretically, parallel manipulators perform higher precision than their serial counterparts. However, the output accuracy is sensitive to their configurations and dimensions. This paper presents a kind of parallel manipulator with kinematically redundant structure, which can improve the output accuracy by optimizing the error transmission from the active joints to the end-effector. With the kinematic redundancy, free redundant variables can be defined as second task variables, which provide the possibility to select a proper configuration for least error transmission at any pose (the position and orientation) of the end-effector for a given task. Contrast to non-redundant manipulators, the output errors of the proposed manipulator, caused by the active joints input errors, can be optimized rather than determined. By this goal, new limbs with redundant parallel structures are introduced to non-redundant planar parallel manipulators. Numerical example shows that the new architecture has the potential to enhance the output accuracy for a given pose or prescribed trajectory of the end-effector.

Topics: Manipulators
Commentary by Dr. Valentin Fuster
2009;():485-491. doi:10.1115/DETC2009-86243.

Among the 3D single-loop overconstrained linkages, quite a number of them are combinations of Bennett linkages. Mobility on the overconstrained linkages including the Bennett-based linkages is known to be one of the difficult topics in kinematics. In the paper, a new approach based on the linear superposition principle for determining the orders of Bennett-based linkages is proposed, and the mobility of some typical Bennett-based linkages is calculated with the Modified Grübler-Kutzbach criterion. In addition, geometric properties of some of the screw systems are employed to identify whether the mobility is global.

Topics: Linkages
Commentary by Dr. Valentin Fuster
2009;():493-502. doi:10.1115/DETC2009-86654.

An important difficulty in the design of parallel manipulators is their reduced practical workspace, due mainly to the existence of a complex singularity locus within the workspace. The workspace is divided into singularity-free regions according to assembly modes and working modes, and the dimensioning of parallel manipulators aims at the maximization of those regions. It is a common practice to restrict the manipulator’s motion to a specific singularity-free region. However, a suitable motion planning can enlarge the operational workspace by means of transitions of working mode and/or assembly mode. In this paper, the authors present an analytical procedure for obtaining the loci of cusp points of a parallel manipulator as algebraic expressions of its dimensional parameters. The purpose is to find an optimal design for non-singular transitions to be possible.

Topics: Manufacturing
Commentary by Dr. Valentin Fuster
2009;():503-512. doi:10.1115/DETC2009-86722.

This paper introduces a novel approach to automated mechanism synthesis called “convertible agents”. The evolutionary computing technique has been developed specifically for the unique design challenges encountered when synthesizing a mechanism for both type and dimensionality. Several case studies are presented which demonstrate the approach’s effectiveness over earlier solution strategies. In these studies, six different planar single-degree-of-freedom mechanism types are considered: a four-bar mechanism, Stephenson’s six-bar-mechanisms (types I, II, and III), and Watt’s six-bar-mechanisms (types I and II). The method is readily scalable to account for any number of different mechanism types and complexities.

Topics: Mechanisms
Commentary by Dr. Valentin Fuster
2009;():513-522. doi:10.1115/DETC2009-86723.

The search of Pareto-optimal solutions for the optimal design of Low-Mobility Parallel Manipulators with Schönflies motion is the subject of this paper. As a working example, a four-degree-of-freedom symmetric parallel manipulator for Schönflies-motion generation is taken. In previous work, analytically found objective functions for the optimal design were used. As a consequence, some limitations were detected and new functions are required. First, a manipulator description is made, and kinematic and dynamic problems are solved. Next, an operational and dexterous workspace along with its volume is found making use of a discretization. Further, the variation of this volume with dimensional parameters is shown for purpose of optimal design. Similarly, the manipulator’s dexterity based on the Frobenius norm is found and weighted with the measure of dispersion. Then, upon a type of testing trajectory over this workspace, kinematic and dynamic results in the actuators are proposed as objective functions in multiobjective optimization.

Commentary by Dr. Valentin Fuster
2009;():523-530. doi:10.1115/DETC2009-86769.

Today’s machine tool industry mainly consists of small and medium-sized enterprises. Thus, the simulation of new products often does not seem to be cost effective due to the small number of items produced and the high cost of simulation tools. Nevertheless, the use of simulation tools is essential in order to tap the full potential of new challenging concepts like parallel kinematic machines. This paper presents a simulation method supporting the development process of parallel kinematic machine tools from the first concept to the prototype. In order to render the method applicable for the machine tool industry, a special focus is placed on tool efficiency. A modular modeling concept will ensure that the structure of the first kinematic model of the concept phase can be enhanced during the development process and developed into more detailed models, e.g. for dimensioning calculations or to study the dynamic behavior of machine tools. Thus, the method efficiently supports the whole development process with a simulation model gradually increasing in detail according to the requirements of the machine tool designer.

Commentary by Dr. Valentin Fuster
2009;():531-538. doi:10.1115/DETC2009-87140.

This paper deals with the development of an efficient method for synthesizing crank-rocker mechanisms that are capable of generating perceptually simple and smooth paths that can be approximated by the first and second harmonics of Fourier series. Through the harmonic analysis of the loop closure equations of the crank-rocker mechanism, analytical relations among the nine design variables are identified. This reduces the dimensions of the search space to two and thereby greatly speed up the synthesis process.

Topics: Mechanisms
Commentary by Dr. Valentin Fuster
2009;():539-545. doi:10.1115/DETC2009-87170.

This paper offers an exact solution for the perfect gravity-balance of a class of spatial manipulators with no translational joints. The methodology used is based on the concept of conservation of the gravitational and elastic potential energy of the system. The overall gravitational potential energy of a serial-connected, n-link manipulator is identified to be contributed by n subsystems, where each subsystem is kinematically equivalent to one of the primary links of the manipulator, and possesses the accumulated mass of its post-connected links with a fixed mass center located on the subsystem. The gravitational potential energy of such a subsystem can be fully balanced by the elastic potential energy of the spring fitted between the link and its adjacent pseudo-base. Since the rotation axis of the pseudo-base is required to be in the direction of gravity, n serial-connected RSSR modules are constituted along the primary chain of the manipulator to provide a pseudo-base for each of the primary links. With one linear, zero-free-length spring fitted between each of the primary links of the manipulator and its associated pseudo-base, a static equilibrium of the considered mechanism in any configuration can be reached. A numerical example of the model of a six-DOF industrial robot has demonstrated the success of the proposed methodology.

Commentary by Dr. Valentin Fuster
2009;():547-557. doi:10.1115/DETC2009-87288.

The Chebychev-Kutzbach-Grübler criterion (CKG Criterion) is useful in evaluating the mobility of robotic mechanisms. However, it cannot correctly evaluate the mobility of some mechanisms (e.g., overconstrained mechanisms, mechanisms with Passive DOF, etc.). In this paper, we propose a more general method for automatically evaluating mobility, based on computer algebra. In this method, the constraints caused by links and joints are expressed as simultaneous algebraic equations (SAE), which is converted into a type of canonical form, constructed from special polynomials called the Gröbner Bases. The mobility can be then determined by checking the appearance sequence of the variables in the Gröbner Bases. The results are reliable because the method consists only of symbolic calculations and are free of numerical error problems. Moreover, the method is automatically applicable to spatial mechanisms containing various types of joints as no heuristics are required. We show the mobility analysis of three robotic mechanisms for which the CKG Criterion is not applicable, and show that the proposed method can correctly evaluate their mobility.

Commentary by Dr. Valentin Fuster
2009;():559-575. doi:10.1115/DETC2009-87345.

This paper proposes a geometric way to generate metamorphic configurations and investigates metamorphic principles based on geometrized displacement group. Metamorphic reconfiguration techniques are revealed as the variations of kinematic joints, kinematic links and geometric orientation constraints particularly by examining the invariant configuration properties of a mechanism. The nature of all these configuration changes belongs to geometric constraint category. Metamorphic configuration units are proposed as the irreducible reconfiguration modules to envelop these reconfiguration techniques. It can self-reconfigure or be combined to generate metamorphosis. Moreover, the geometrized displacement group is lent to achieve a geometric representation for configuration modelling and further reconfiguration operations. Based on seting up kinematic group extended qualitatively according to its group structure, geometrized displacement group modelling is proposed for these identified metamorphic configuration units. The investigated group motion-matrix is an integration of its displacement group properties and kinematic extensions. Then defined geometric constraint relations and the proposed dependence rules lead to metamorphic principles. In this way, metamorphic process is mapped to matrix operations under group extensions and their compositions. Design examples and a metamorphic joint with six configurations are given to illustrate the feasibility of these metamorphic principles.

Topics: Mechanisms
Commentary by Dr. Valentin Fuster
2009;():577-584. doi:10.1115/DETC2009-87366.

In this paper, we use seven-position synthesis to add four TS constraints to a TRS serial chain robot and obtain a two degree-of-freedom spatial eight-bar linkage. The TRS chain is an elbow manipulator, similar to a PUMA robot. We synthesize a TS dyad to connect the base of the robot to its forearm, and then we synthesize three TS dyads that connect the upper arm of the robot to its end-effector. The result is a two degree-of-freedom spatial eight-bar linkage that moves through seven prescribed positions. It consists of a TRST loop supporting a 3TS-RS platform, which we denote as a TS-TRS-3TS spatial linkage. We formulate and solve the design equations for the TS dyads, and analyze the resulting eight-bar linkage. An example demonstrates our results.

Topics: Linkages , Chain
Commentary by Dr. Valentin Fuster
2009;():585-594. doi:10.1115/DETC2009-87447.

This paper presents elliptical rolling contact joints as an alternative to circular rolling contact and conventional revolute joints where high quality force transmission—low friction and backlash—with variable output are desired. Parameters specific to the joint and its position are defined in terms of relative link angles and elliptical surface geometry. These relationships allow elliptical rolling contact joints to be incorporated in vector loop summations used in kinematic analysis. Notably, elliptical rolling contact is developed as the more general case of which circular rolling contact is a subset. Elliptical rolling contact joints are shown to offer several benefits over circular rolling contact, including: reduced Hertz contact stresses, variable output velocity, maximum use of contact interface by distributing small rotations across surfaces of small curvature, reduced forces on constraining members, and no-slip pure rolling provided exclusively by connecting links (or flexures) without the need for gear teeth or friction.

Commentary by Dr. Valentin Fuster
2009;():595-604. doi:10.1115/DETC2009-87516.

This paper offers a unified method for a complete and unified treatment on the mobility identification and rectification of any planar and spherical six-bar linkages regardless the linkage type and the choice of the input, output, or fixed links. The method is based on how the joint rotation spaces of the four-bar loop and a five-bar loop in a Stephenson six-bar linkage interact each other. A Watt six-bar linkage is regarded as a special form of Stephenson six-bar linkage via the stretch and rotation of a four-bar loop. The paper offers simple explanation and geometric insights for the formation of branch (circuit), sub-branch, and order of motion of six-bar linkages. All typical mobility issues, including branch, sub-branch, and type of motion under any input condition can be identified and rectified with the proposed method. The method is suitable for automated computer-aided mobility identification. The applicability of the results to the mobility analysis of serially connected multiloop linkages is also discussed.

Topics: Linkages
Commentary by Dr. Valentin Fuster
2009;():605-614. doi:10.1115/DETC2009-87747.

Metamorphic mechanisms are a class of mechanisms that change their mobility during motions. The deployable and retractable characteristics of these unique mechanisms generate much interest in further investigating their behaviors and potential applications. This paper investigates the resultant configuration variation based on adjacency matrix operation and an improved approach to mechanism synthesis is proposed by adopting elementary matrix operations on the configuration states so as to avoid some omissions caused by existing methods. The synthesis procedure begins with a final configuration of the mechanism, then enumerates possible combinations of different links and associates added links with the binary inverse operations in the matrix transformations of the configuration states, and finally obtains a synthesis result. An algorithm is presented and the topological symmetry of links is used to reduce the number of mechanisms in the synthesis. Some mechanisms of this kind are illustrated as examples.

Topics: Mechanisms
Commentary by Dr. Valentin Fuster
2009;():615-625. doi:10.1115/DETC2009-87818.

In this paper, we present an interactive, visual design approach for the dimensional synthesis of planar 4R, 5R, and 6R closed chains for a given rational motion using constraint manifold modification. This approach is implemented in an interactive software tool that provides mechanism designers with an intuitive way to determine the dimensional parameters of planar mechanisms, and in the process equips them with an understanding of the design process. The theoretical foundation of this work is based on representing planar displacements with planar quaternions which can be seen as points in a special higher dimensional projective space (called the image space), and on formulating the kinematic constraints of closed chains as algebraic surfaces in the image space. Kinematic constraints under consideration limit the positions and orientation of the coupler in its workspace. In this way, a given motion of a mechanism in the Cartesian space maps to a curve in the image space that has to stay within the bounds of the algebraic surfaces. Thus, the problem of dimensional synthesis is reduced to determining parameters of equations that describe algebraic surfaces. This paper is an extension of our previous work in which we had presented some preliminary results based on an optimization method. We show that the interactive approach presented here is general in nature, and can be easily used for the dimensional synthesis of any mechanism for which kinematic constraints can be expressed algebraically. The process of designing is fast, intuitive, and especially useful when an optimization based approach would be computationally demanding, and mathematically difficult to formulate. This simple approach also provides a basis for students and early designers to learn and understand the process of mechanism design by simple geometric manipulations.

Topics: Motion , Chain , Design , Manifolds
Commentary by Dr. Valentin Fuster
2009;():627-632. doi:10.1115/DETC2009-86074.

This paper attempts to establish the unified topological models and corresponding mathematical representations for planar simple joint, multiple joint and geared (cam) kinematic chains. First, the conventional topological representation models of kinematic chains are introduced. Then new topological models of multiple joint and geared (cam) kinematic chains, which are derived from the topological graph of simple joint kinematic chains, are presented. The characteristics of the new topological graphs and their associations with the topological graph of simple joint kinematic chains are also addressed. The most important merit of the new topological graphs is that it makes it much easier to do unified structure synthesis and further establish conceptual design platform for various planar mechanisms of these kinds.

Topics: Mechanisms
Commentary by Dr. Valentin Fuster
2009;():633-638. doi:10.1115/DETC2009-86213.

In this paper, the self motions of a novel 3-DOF fully decoupled translational parallel robot, called the Pantopteron, are presented. The Pantopteron is similar to the well-known Cartesian parallel manipulator (Tripteron), but due to the use of pantograph linkages, an amplification effect is achieved. Therefore, equipped with the same actuators, the mobile platform of the Pantopteron moves faster than that of the Tripteron. This amplification is defined by the magnification factor of the pantograph linkages. The self motions are probably the most critical types of singularities a manipulator can meet. Therefore it is of utmost importance to have a good knowledge of them. Design considerations are also discussed in order to create Pantopterons without self motions.

Topics: Motion
Commentary by Dr. Valentin Fuster
2009;():639-645. doi:10.1115/DETC2009-86281.

The manipulator workspace denotes the work area of a manipulator and it is an important foundation of designing a robot. The workspace of a novel symmetrical 4-DOF 3-RRUR parallel manipulator is researched here based on the inverse solution. Five essential conditions of a point within the workspace of the manipulator are presented and the cause of the existence of each condition is given. According to the characteristics of this parallel mechanism and the geometrics, and based on the inverse kinematical solution of the mechanism, the reachable workspace under the five conditions is ascertained. The three-dimensional graph and its section graphs of the workspace are obtained. The thoroughness of the workspace of the manipulator is proved. The mechanism with the advantages of simple symmetric structure and large stiffness can be applied to the developments of four-dimensional force sensors, NC positioning platforms, parallel machine tools and micro-positional parallel manipulators. This work is with great significance to the practical application of the manipulator.

Topics: Manipulators
Commentary by Dr. Valentin Fuster
2009;():647-653. doi:10.1115/DETC2009-86294.

This paper presents the development of computational simulation based on the dynamics of a robotic bird. The study analyze the wing angle of attack and the velocity of the bird, the tail influence, the gliding flight and the flapping flight with different strategies and algorithms of control. The results are positive for the construction of flying robots. Some highlights are given about the fist implemented architecture of the structure of a robotic bird. This platform consists on a body, wings and tail with actuators independently controlled though a microcontroller; a radio transmission system and batteries are used in order to avoid wired connections between the computer and the robot.

Topics: Robots
Commentary by Dr. Valentin Fuster
2009;():655-662. doi:10.1115/DETC2009-86328.

The aim of this project is to design, study and build an “eel-like robot” prototype able to swim in three dimensions. The study is based on the analysis of eel swimming and results in the realization of a prototype with 12 vertebrae, a skin and a head with two fins. To reach these objectives, a multidisciplinary group of teams and laboratories has been formed in the framework of two French projects.

Topics: Robots
Commentary by Dr. Valentin Fuster
2009;():663-673. doi:10.1115/DETC2009-86338.

A self-balanced quadruped walking machine with leg mechanisms of 10-bar linkage is designed by a systematic approach. At first, an existing leg mechanism of 10-bar linkage is selected as the tentative design, and the dimensional synthesis is performed to obtain the desired foot trajectory by the optimization technique of ALM. Next, the speed ratio between the crank speed during the support phase and that during the transfer phase is decided by a two-speed control method to achieve a sufficient time period of support phase. Then, in order to make sure that there are always at least three legs on the ground for the wave gait to enhance the stability of locomotion, the foot point of each leg at a specific time is placed on the specific position upon the foot trajectory. The force analysis and computer simulation are carried out to evaluate the requirements for driving and to recognize the characteristics of the designed walking machine. And, the force transmission during the full cycle is realized and the specification of the driving motor is decided. Finally, a prototype of the designed quadruped walking machine is constructed and it is proven that this design is practical and feasible.

Commentary by Dr. Valentin Fuster
2009;():675-680. doi:10.1115/DETC2009-86351.

This paper presents a new generalized control hardware architecture based on embedded on-board wireless communication network between robot’s links and modules such as the actuators and sensors. This approach results in modular control hardware architecture since no cable connections are used between the actuators and sensors in each of a given mobile robot subsystems (links). The effectiveness of this approach is experimentally demonstrated and validated by implementing it with a hybrid mobile robot mechanism as a case study. The hybrid mobile robot mechanism integrates the locomotion mechanism and manipulator arm mechanism as one entity to support both locomotion and manipulation simultaneously and interchangeably.

Commentary by Dr. Valentin Fuster
2009;():681-688. doi:10.1115/DETC2009-86551.

The paper reports work in progress on the development of an innovative gearless pitch-roll wrist (PRW) for robotic applications. The PRW bears the morphology of a bevel-gear differential, its novelty lying in the absence of gears. Indeed, the PRW motivating this study is based on cams and rollers, intended to overcome the drawbacks of their bevel-gear counterparts—backlash, Coulomb friction and low stiffness. A testbed designed for parameter identification is introduced here. The paper discusses the mathematical modeling of the testbed, starting from its iconic model. The mathematical model is used to obtain the frequency response of the whole testbed, regarded as a multiple-input-multiple-output system, under the assumption that the parts of the spherical epicyclic train are rigid. The numerical values for the inertia parameters used in the model were taken from CAD models, those for stiffness and damping, as yet unknown, were estimated from a similar testbed reported elsewhere. The work ahead targets the experimental derivation of the Bode plots of the testbed, from which the numerical values of its inertia, stiffness and damping parameters are to be estimated. Moreover, having computed the stiffness and damping parameters of the testbed, the next step will be to drive the PRW at high frequencies, of the order of 1 kHz, to enable the identification of the stiffness and damping parameters of the PRW proper.

Topics: Modeling , Robotics
Commentary by Dr. Valentin Fuster
2009;():689-700. doi:10.1115/DETC2009-86770.

Generally, adjustment of gravity equilibrator to a new payload requires energy, e.g. to increase the pre-load of the balancing spring. A novel way of energy-free adjustment of gravity equilibrators is possible by introducing the concept of a storage spring. The storage spring supplies or stores the energy necessary to adjust the balancer spring of the gravity equilibrator. In essence the storage spring mechanism maintains a constant potential energy within the spring mechanism; energy is exchanged between the storage and balancer spring when needed. Various conceptual designs using both zero-free-length springs and regular extension springs are proposed. Two models were manufactured demonstrating the practical embodiments and functionality.

Commentary by Dr. Valentin Fuster
2009;():701-710. doi:10.1115/DETC2009-86777.

Nuclear material processing operations present numerous challenges for effective automation. Confined spaces, hazardous materials and processes, particulate contamination, radiation sources, and corrosive chemical operations are but a few of the significant hazards. However, automated systems represent a significant safety advance when deployed in place of manual tasks performed by human workers. The replacement of manual operations with automated systems has been desirable for nearly 40 years, yet only recently are automated systems becoming increasingly common for nuclear materials handling applications. This paper reviews several automation systems which are deployed or about to be deployed at Los Alamos National Laboratory for nuclear material handling operations. The needs that resulted in the development of these systems can be found throughout the nuclear industry. Highlighted are the current social and technological challenges faced in deploying automated systems into hazardous material handling environments and the opportunities for future innovations.

Commentary by Dr. Valentin Fuster
2009;():711-720. doi:10.1115/DETC2009-86790.

Retinopathy of prematurity is caused by abnormal blood vessel development in the retina of a premature infant. Current options for surgery tables used in laser treatment of this condition are limited. Full-size operating tables or table attachments are used but provide restricted patient manipulation and cause the surgeon to assume ergonomically undesirable positions. A stand-alone four-degree-of-freedom (4-DOF) infant surgical table was designed and is presented in this paper. The new table enables the surgeon to manipulate the patient while sitting in a comfortable position. The table platform can pitch left/right and fore/aft. The table platform can rotate 360° and translate vertically. Two linear actuators and a motor with ball screw provide the three degrees of freedom for table pitch and rotation through a spherical wrist-mechanism. A ball screw and motor achieve vertical movement of the table platform. The rigors of surgery and associated space constraints were accounted for in this design. The design consists of three subassemblies which can be disassembled for transport between operating theaters. A wide base is used to prevent the table from tipping. Biocompatible materials have been selected for all parts. Lastly, foot controls are used to keep the surgeon’s hands free.

Topics: Design , Surgery
Commentary by Dr. Valentin Fuster
2009;():721-726. doi:10.1115/DETC2009-86808.

Razor clams (Ensis directus) are one of nature’s most adept burrowing organisms, able to dig to 70cm at nearly 1cm/s using only 0.21J/cm. Ensis reduces burrowing drag by using motions of its shell to fluidize a thin layer of substrate around its body. Although these shell motions have an energetic cost, moving through fluidized rather than packed soil results in exponentially lower overall energy consumption. This paper describes the design and testing of RoboClam, a device that mimics Ensis digging methods to understand the limits of razor clam-inspired burrowing, how they scale for different environments and conditions, and how they can be transferred into engineering applications. Using a genetic optimization solver, we found that RoboClam’s most efficient digging motion mimicked Ensis shell kinematics and yielded a power law relationship between digging energy and depth of n = 1.17, very close to the ideal value of n = 1. Pushing through static soil has a theoretical energy-depth power law of n = 2, which means that Ensis-inspired burrowing motions can provide exponentially higher energy efficiency and nearly depth-independent drag resistance.

Commentary by Dr. Valentin Fuster
2009;():727-738. doi:10.1115/DETC2009-87355.

This paper presents the design and preliminary evaluation of an elastic lower-body exoskeleton (eExo). Human legs behave in a spring-like fashion while running. We selected a design that relied solely on material elasticity to store and release energy during the stance phase of running. The exoskeleton included a novel knee joint with a cam and a Bowden cable transferring energy to and from a waist-mounted extension spring. We used a friction-lock clutch controlled by hip angle via a pneumatic cylinder to release the cable during swing phase for free movement of the leg. The design also incorporated a composite leaf spring to store and release energy in the distal portion of the exoskeleton about the foot and ankle. Preliminary test data for our target subject showed that his typical leg deflection was 0.11 m with leg stiffness of 16 kN/m while running at 3.0 m/s. We used these values to set the desired stiffness (60±15% of the normal leg stiffness, or 9.6±2.4 kN/m) and deflection (0.11 m) of the exoskeleton. We created simplified multi-body and full finite element quasi-static models to achieve the desired system stiffness and validate our results, respectively. The final design model had an overall stiffness of 7.3 kN/m, which was within the desired range. We fabricated a single-leg prototype of the exo–skeleton that weighed 7.1 kg. We tested the exoskeleton stiffness quasi-statically and found a stiffness of 3.6 kN/m. While running, the exoskeleton provided ∼30% of the total leg stiffness for two subjects. Although the stiffness was lower than desired, the fabricated prototype demonstrated the ability of a quasi-passive exoskeleton to provide a significant portion of an individual’s leg stiffness while running.

Commentary by Dr. Valentin Fuster
2009;():739-747. doi:10.1115/DETC2009-87361.

This paper describes the design and experimental testing of a variable effective compliance transmission for use with revolute joints in robotic systems. The transmission consists of a parallel arrangement of a torsional spring and a rotary magnetorheological (MR) fluid damper and is placed between a brush–less DC gearmotor and an end-effector. While the stiffness of the torsional spring is constant, the damping of the MR damper can be adjusted by changing the applied current, which alters the apparent rigidity of the system. Hence, the transmission exhibits a variable “effective compliance”. Experimental results show that increasing the damping can reduce position errors during acceleration and deceleration phases of motion. Impact tests indicate that for small currents applied to the damper (low damping), the presence of the torsional spring reduces impact loads during constant velocity motion. In combination, these results indicate that the variable effective compliance transmission can improve safety without sacrificing precise motion control.

Topics: Dampers , Design , Testing
Commentary by Dr. Valentin Fuster
2009;():749-758. doi:10.1115/DETC2009-87412.

The Bistable Aerial Platform (BAP) functions as a switch in that its platform can lock in two positions, up or down. The Quadrantal Bistable Mechanism (QBM), the principle component of the BAP, is described in detail. A second component of the device, the Helico-Kinematic Platform (HKP), is still under investigation. It is anticipated that the model of the QBM, described here, will be combined with the HKP model, when complete, to form a full model of the Bistable Aerial Platform.

Commentary by Dr. Valentin Fuster
2009;():759-766. doi:10.1115/DETC2009-87426.

The goal of this work is to investigate the design of a joint suitable for use in a digital robot. Digital robotics, for the purposes of this paper, refers to robots with discrete joint output positions and binary input. For each proposed design, the input is a linear position, and the output is an angular position. Utilizing the kinematics of each design, the mechanical advantage is generated and corresponding design charts are generated for both a two and three position slotted slider as well as a rotating angular slotted block driven by either a rack and gear or a modified Geneva drive.

Topics: Robots , Design
Commentary by Dr. Valentin Fuster
2009;():767-773. doi:10.1115/DETC2009-87479.

Robotic devices have made inroads in various areas of medical practice. This paper offers a design of robot kinematics for ultrasound probe manipulation to obtain reproducible Achilles tendon images for quantifying injury or response to treatment. The design includes a motor-controlled 4-DOF arm with an additional smaller, passive four-bar linkage mount for the ultrasound probe to optimize surface contact with the subject and increase the mobility to 5 DOF.

Commentary by Dr. Valentin Fuster
2009;():775-780. doi:10.1115/DETC2009-87512.

This paper introduces a novel metamorphic gripper proposed as a step towards the solution of the bin picking problem. This gripper makes uses of multiple poses while maintaining contact with the part allowing for in-hand manipulation without the use of finger gaiting. The gripper is analyzed using classical techniques of degrees of freedom and adjacency matrices. This analysis allows for a basic understanding of the motion of the gripper.

Topics: Design , Testing , Grippers
Commentary by Dr. Valentin Fuster
2009;():781-787. doi:10.1115/DETC2009-87515.

In many systems, an acceleration event triggers some downstream operation. In such devices the “acceleration event of interest” must be identified from among spurious accelerations, which are often of appreciable magnitude — sometimes greater than the event of interest — but are of short duration, perhaps one tenth that of the event of interest. Such devices, therefore, are essentially inertia-driven timing mechanisms. The principles presented here allow for designing practical triggering devices actuated by arbitrarily long acceleration events where traditional mechanical inertia switches would have to be impractically large due to the non-linearity of component motion. Here, during acceleration of the mechanism, a plurality of spring-mass elements are at first immobilized, but are released in succession. The final stage of the chain is then used to trigger some downstream event. The key is that the time during which individual components are permitted to displace is only a fraction of the total delay time. Therefore, the delay elements are never given the opportunity to achieve high velocities and hence do not travel great distances during the duration of the delay. In addition to illustrating the general approach, design examples showing novel mechanisms for immobilizing and successively releasing the inertial elements are offered.

Commentary by Dr. Valentin Fuster
2009;():789-796. doi:10.1115/DETC2009-87526.

A novel class of two-stage, vibration-based electrical energy generators is presented for linear or rotary input motions in applications which the input speed is relatively low and varies significantly over time such as wind mills, turbo-machinery used to harvest tidal flows, devices for harnessing coastal wave energy, and the like. Current technologies use magnet-and-coil based electrical generators in such machinery. However, to make the generation cycle efficient, gearing or other similar mechanisms must be used to increase the input speed. Variable speed-control mechanisms are also usually needed to achieve high energy conversion efficiency. Additionally, in many applications, such as those where energy is to be harvested from very low frequency oscillations of a platform such as a buoy or a ship, the use of speed increasing mechanisms such as gearing or the like is impractical. In this paper, a novel class of two-stage electrical energy generators that could operate with very low speed and highly variable input rotations and/or oscillations is described. The first stage consists of simple linkage mechanisms, which are used to excite vibratory elements. These two-stage generators are designed to convert low-speed and highly variable input rotations and oscillations to relatively high and constant frequency vibratory motions, which are then used to generate electrical energy using mechanical to electrical energy conversion devices such as piezoelectric elements. The design of a number of such two-stage generators together with a discussion of their potential applications is presented. Specifically, the computer simulated output of a small wave energy harvester design is presented.

Commentary by Dr. Valentin Fuster
2009;():797-805. doi:10.1115/DETC2009-87543.

Flapping wing motion produces positive lift in the down stroke and negative lift in the upstroke under zero forward velocity. Large birds frequently exhibit flight behavior where their wings are folded during the upstroke, thus lowering the air resistance as the wing is moved upwards. The result is reduced magnitude of negative lift produced during the upstroke, relative to the positive lift produced in the down stroke, where the wings are unfolded and the area is increased. We expect that by incorporating this style of upstroke wing folding into miniature air vehicle (MAV) platforms, beneficial flight properties would arise. Specifically, a portion of the wings’ overall lift will be generated by upstroke folding and downstroke unfolding, even at zero forward velocity. Such a capability will reduce the reliance on aerodynamic lift produced due to the forward motion of the MAV. This in turn would reduce the minimum flight-sustaining forward velocity and thus enhance MAV maneuverability by allowing for a reduced turning radius. Incorporating wing folding into a miniature air vehicle platform presents a unique challenge due to strict weight constraints present at small sizes. Using actuators to accomplish folding actively is not feasible due to the added weight of the actuators and the need for an on-board control system to synchronize the folding with the wing flapping motion. Therefore, the folding motion must be accomplished passively, since this is currently the only viable option in miniature MAVs. We have developed a passive, spatially distributed, one-way folding mechanism. This mechanism has been incorporated into a flying MAV testbed, and has successfully shown that the flapping wing MAV with folding wings is capable of flying at reduced forward velocity, while maintaining the payload carrying capacity.

Topics: Vehicles , Wings
Commentary by Dr. Valentin Fuster
2009;():807-817. doi:10.1115/DETC2009-87657.

Modular robotic systems can form arbitrary shapes that best suit task requirements. Such a system comprised of microscale components could form reconfigurable microstructures or high resolution physical prototypes. This paper presents methods aimed at miniaturization of this programmable matter system towards the millimeter scale or smaller. The Right Angle Tetrahedron Chain Externally-actuated Testbed (RATChET) can be folded into arbitrary 3D shapes. The tetrahedron shaped modules are designed to have limited complexity and employ technologies which can be realized at the microscale. The tools developed to design the module’s compliant mechanism can be used to develop small scale modules in the future. Experiments with centimeter scale modules demonstrate that an external actuator can fold a chain of right angle tetrahedrons into 3D shapes. If given the fold pattern to make a shape, a simulator determines the motion sequence for the 2DOF external actuator to fold that pattern.

Topics: Chain , Shapes
Commentary by Dr. Valentin Fuster
2009;():819-828. doi:10.1115/DETC2009-87678.

Over the past decade, small satellites have gained the interest of the space industry as a new and cost effective approach for servicing space assets. To address the special constraints inherent to the component miniaturization required for these satellites, researchers in the Space, Automation and Manufacturing Mechanisms Laboratory (SAMM) are exploring foldable mechanisms and their effectiveness for providing autonomous rendezvous and docking capabilities for small space vehicles. This paper focuses particularly on the design of autonomous docking mechanisms for space vehicles within the small satellite class known as picosatellite (size and mass requirements: 1 kilogram mass within a 10×10×10 centimeter cube). The docking mechanisms deployment scenario is a dual satellite system comprised of two small satellites (a chaser and a target). The chaser has attitude and translational control capability, while the target is a passive satellite having only attitude stabilization capability. This paper will first present a review of the existing docking mechanism technology utilized in space. This is followed by details of a foldable mechanism approach for providing small satellites autonomous docking capabilities. This includes geometric and dynamic analysis conducted in ADAMS software simulations.

Commentary by Dr. Valentin Fuster
2009;():829-839. doi:10.1115/DETC2009-87824.

This paper presents the design and fabrication of a novel minimally invasive surgical (MIS) tool — FlexDex™ — that provides enhanced dexterity, intuitive actuation, and natural force feedback in a cost-effective compact package. These attributes are accomplished by means of a fundamentally new MIS tool design paradigm that employs a tool reference attached to the surgeon’s arm, and utilizes a virtual center at the tool input that coincides with the surgeon’s wrist. The resulting physical configuration enables a highly intuitive one-to-one mapping of the surgeon’s arm and hand motions at the tool input to the end-effector motions at the tool output inside the patient’s body. Furthermore, a purely mechanical design ensures low-cost, simple construction, and natural force feedback. A functional decomposition of the proposed design paradigm and associated physical configuration is carried out to identify key modules in the system. This allows for the conceptual and detailed design of each module, followed by system-level integration. The key innovative aspects of the tool design include a three-dimensional parallel-kinematic virtual center mechanism, a decoupled 2DoF end-effector design, and the associated transmissions system.

Commentary by Dr. Valentin Fuster
2009;():841-850. doi:10.1115/DETC2009-86201.

A new method combining trajectory planning and coordination or formation control of robotic and autonomous systems is presented. The method generates target trajectories that are either asymptotically stable or result in a stable limit cycle. The former case is used to implement formation control. Coordination is guaranteed in the latter case due to the nature of limit cycles where non-crossing independent paths are automatically generated from different starting positions that smoothly converge to closed orbits. The use of position feedback in the trajectory generation allows for simultaneous determination of a stable tracking control law and consideration of constraints and system limitations. The tracking control law presented in this work is based on sliding mode control which is suitable for real-time implementation. It is also robust to modeling uncertainties and disturbances normally encountered in autonomous operations. A system of robotic manipulators and a group of autonomous vehicles are used as examples to demonstrate the capabilities and advantages of the proposed method.

Commentary by Dr. Valentin Fuster
2009;():851-860. doi:10.1115/DETC2009-86247.

For their inherent stability and simplicity, wheeled robots are very common in robotics applications — but a major drawback of wheeled robots is their inability to navigate over large obstacles or steps without assistance. Active systems that have been designed for use on wheeled robots to lift the robot over a step — such as USU’s T3 and Virginia Tech’s IMPASS — are effective, but are limited due to the size, cost, and power required for the additional actuators. A novel, inertially actuated, passive dynamic system, excited by the motion of the robot, is introduced to allow a wheeled robot to “pop a wheelie” on each axle and hop over a step. The system investigated here is a sliding mass-spring that shifts forward and backward based on the acceleration of the base robot. By coordinating the acceleration and deceleration of the robot, the front wheels can lift over a step and the rear wheels can be pulled up afterward — both actions being a product of inertial actuation. Key advantages of this system are that the design is simple, cost-effective, and can be adjusted and retrofit to a different wheeled robot in the future with little effort. This paper presents the development of a novel inertially actuated, passive dynamic step climbing wheeled robot. Derivations of the dynamic model of the inertially actuated system are given and a computer simulation and experiments of an implementation of this sliding mass system are presented, followed by conclusions with possibilities for future work.

Topics: Robots
Commentary by Dr. Valentin Fuster
2009;():861-870. doi:10.1115/DETC2009-86529.

This paper presents a method to control a manipulator system grasping a rigid-body payload so that the motion of the combined system in consequence of external applied forces to be the same as another free-floating rigid-body (with different inertial properties). This allows zero-g emulation of a scaled spacecraft prototype under the test in a 1-g laboratory environment. The controller consisting of motion feedback and force/moment feedback adjusts the motion of the test spacecraft so as to match that of the flight spacecraft. The stability of the overall system is analytically investigated, and the results show that the system remains stable provided that the inertial properties of two spacecraft are different and that an upperbound on the norm of the inertia ratio of the payload to manipulator is respected. Important practical issues such as calibration and sensitivity analysis to sensor noise and quantization are also presented. Finally, experimental results are presented.

Commentary by Dr. Valentin Fuster
2009;():871-878. doi:10.1115/DETC2009-86541.

Although adding compliant, frictional material on robotic fingers to improve the performance is generally accepted, at least for underactuated hands this effect is hardly quantified. In this study, the phalanges of an underactuated hand in an experimental setup were firstly covered with material of different friction coefficients but equal contact compliance, while the force to pull a grasped object completely out of the hand was measured. Then, the phalanges were covered with material of different compliance while the same measurements were done. In the latter experiment, the effect of contact friction was eliminated by using a specially designed testbed that emulates a frictionless object. The experiments showed an increase of the maximal pull force by 250% when the friction coefficient of the contact material increased from 0.25 to 1.51. The compliance of the contact material had a marginal effect on this maximum force. Finally, the pull force was calculated by a static grasp model, incorporating contact friction and linear contact compliance. Trends similar to the experiments were observed in these simulations.

Topics: Friction
Commentary by Dr. Valentin Fuster
2009;():879-885. doi:10.1115/DETC2009-86677.

In this paper, a nonlinear observer is designed to estimate the angle of leg for an underactuation biped robot with four links and three actuators. First, using a diffeomorphism transformation, the nonlinear system changes to a system with triangular form. Then, a nonlinear observer, based on the second order sliding mode, is designed for the transformed system. The control law uses only measurable variables, that can be estimated by the observer, for tracking the desired stable gait. Finally, simulation results are presented to show the behavior of biped robot and convergency of the observer in finite time during stable walking.

Commentary by Dr. Valentin Fuster
2009;():887-896. doi:10.1115/DETC2009-86700.

A reactionless mechanism is one which does not exert any reaction force or moment on its base at all times, for any arbitrary trajectory of the mechanism. This paper addresses the static and dynamic balancing of a two-degree-of-freedom parallel planar mechanism (five-bar mechanism). A simple and effective adaptive balancing method is presented that allows the mechanism to maintain the reactionless condition for a range of payloads. Important proofs concerning the balancing of five-bar mechanisms are also presented. The design of a real mechanism where parallelogram linkages are used to produce pure translations at the end-effector is also presented. Finally, using dynamic simulation software, it is shown that the mechanism is reactionless for arbitrarily chosen trajectories and for a variety of payloads.

Commentary by Dr. Valentin Fuster
2009;():897-903. doi:10.1115/DETC2009-86789.

Parallel mechanisms including the 2P RR have opened new horizons for the industrial robotics designer. Enhancements in terms of speed, stability, precision and cost reduction are attainable when compared to higher degree of freedom serial designs. This comes at the cost of a reduced workspace and a more challenging dynamic analysis. The nature of the closed kinematic chain dictates that the application of Newton-Euler analysis methods becomes cumbersome, as the systems of equations describing the dynamics must be solved simultaneously. Moreover, the use of these methods within the context of the parallel mechanism fails to provide designer with the insights apparent from an explicit model. The application of Lagrangian analysis techniques to this robotic linkage is sought as a means of both simplifying the solution of the resulting dynamics and gaining the perspective which only the insight offered by an explicit model can provide.

Commentary by Dr. Valentin Fuster
2009;():905-914. doi:10.1115/DETC2009-87069.

This paper shows an overview of the walking model by grouping them into two large groups: the models with the concentrated mass and models with distributed mass. As an example of the models with concentrated mass, a mass-spring inverted pendulum model is shown, accompanied with a short analysis. As an example of a more complex model, a 13 DOF walking robot model is analyzed including the model kinematics, dynamics and controls accompanied with numerical solutions (simulations) for particular desired joint trajectories, recorded from a real human walking cycle. Kinematic and Dynamic analysis is discussed including results for joint torques and ground force necessary to implement the prescribed walking motion. This analysis is accompanied with a limited comparison with available experimental data. Finally, an inverse plant and tracking error linearization based controller design approach is described accompanied with results analysis and conclusions about the controller performances.

Commentary by Dr. Valentin Fuster
2009;():915-921. doi:10.1115/DETC2009-87132.

Maximum virtual stiffness is a critical performance measure for haptic devices. Stable haptic interaction is necessary for realistic feeling of virtual environment. The virtual environment is determined by the application and device. To ensure the stable haptic interaction, the virtual environment must be suitable for the device. Therefore, the virtual stiffness should not be greater than the minimum value of maximum virtual stiffness that a haptic device can stably render in the workspace. This paper proposes a method, utilizing the eigenvalue and eigenvector of stiffness matrix in joint space, to analyze and measure the maximum virtual stiffness distribution in the work space of a haptic device. Therefore, for a given haptic device, the maximum virtual stiffness at each position and orientation can be forecasted by this method. A new sufficient condition for haptic stability is also presented in the view of driven motor in this paper. A series experiments validate the effectiveness of this method.

Topics: Haptics , Stiffness
Commentary by Dr. Valentin Fuster
2009;():923-928. doi:10.1115/DETC2009-87303.

In this paper, the kinematics and dynamics of free-floating space robot system with dual-arms are analyzed. It is shown that the dynamic equations of the system are nonlinearly according to inertial parameters. In order to overcome these problems, the system is modeled as under-actuated robot system, and the idea of augmentation approach is adopted. It is demonstrated that the dynamic equations of the system can be linearly depending on a group of inertial parameters. Based on this result, a robust variable structure control scheme for free-floating space robot system with dual-arms with uncertain inertial parameters to track the desired trajectories in joint space is proposed, and a planar space robot system with dual-arms is simulated to verify the proposed control scheme. The advantage of the control scheme proposed is that it requires neither measuring the position, velocity and acceleration of the floating base with respect to the orbit nor controlling the position and attitude angle of the floating base. In addition to this advantage, it is computationally simple, because of choosing the controller robust for the uncertain inertial parameters rather than explicitly estimating them online.

Topics: Robots
Commentary by Dr. Valentin Fuster
2009;():929-936. doi:10.1115/DETC2009-87513.

An optimal motion planning formulation of throwing for a biped human mechanism is proposed as an extension of a previous study. The unique characteristics of the throwing task—highly redundant, highly nonlinear, and highly dynamic—are addressed in this presentation within the framework of multibody dynamics and optimization. To generate physically feasible throwing motions in a fully predictive method without input reference, rigorous dynamic models are associated with the constraints. Given the target location and the object mass, the algorithm outputs the motion, required actuator torques, release parameters, balance criterion, and ground reaction forces. Overarm and sidearm throwing motions are generated as optimal solutions, which demonstrate valid kinematic and kinetic cause-effect relations.

Commentary by Dr. Valentin Fuster
2009;():937-946. doi:10.1115/DETC2009-87639.

As the first round of baby boomers turn 65 in 2011, we must be prepared for the largest demographic group in history that could need long term care from nursing homes and home health providers. The development of socially assistive robots for health care applications can provide measurable improvements in patient safety, quality of care, and operational efficiencies by playing an increasingly important role in patient care in the fast pace of crowded clinics, hospitals and nursing/veterans homes. However, there are a number of research issues that need to be addressed in order to design such robots. In this paper, we address one of the main limitations to the development of intelligent socially assistive robots for health care applications: Robotic control architecture design and implementation with explicit social and assistive task functionalities. In particular, we present the design of a unique learning-based multi-layer decision making control architecture for utilization in determining the appropriate behavior of the robot. Herein, we explore and compare two different learning-based techniques that can be utilized as the main decision-making module of the controller. Preliminary experiments presented show the potential of the integration of the aforementioned techniques into the overall design of such robots intended for assistive scenarios.

Commentary by Dr. Valentin Fuster
2009;():947-953. doi:10.1115/DETC2009-86162.

A CAD variation geometry approach is proposed for accurately solving the position-orientation, linear velocity/acceleration, and Euler angular velocity/angular acceleration of a symmetrical 3-dof 3-UPU parallel robot. Based on the finite-difference method as the foundations, using the computer aided geometry constraints and dimension driving technique, the simulation mechanism of the 3-UPU parallel robot with Euler angles is created, and the position-orientation can be got. Then the linear velocity/acceleration simulation mechanisms and Euler angular velocity/angular acceleration simulation mechanisms are created. When modifying the driving dimension of three driving limbs, the configurations of the simulation mechanisms are varied correspondingly, and all of the kinematical parameters are solved automatically. The simulation solutions are verified by an analytic approach. The results show that the CAD variation geometry is not only fairly quick and straightforward, but is also advantageous from viewpoint of accuracy and repeatability.

Commentary by Dr. Valentin Fuster
2009;():955-960. doi:10.1115/DETC2009-86175.

The inverse kinematics and the driving forces of a 3RPS-3SPR serial-parallel manipulator (PM) with 6 degree of freedoms (DOFs) are solved in this paper. This 3RPS-3SPR serial-PM includes a lower 3-RPS PM and an upper 3-SPR PM. First, the inverse displacement is solved based on the geometrical constraint and the dimension constraint of this PM. Second, the 9×9 and 6×6 form inverse Jacobian matrices are derived and the driving forces are solved by using principle of virtual work. Finally, the numerical example is given.

Commentary by Dr. Valentin Fuster
2009;():961-967. doi:10.1115/DETC2009-86227.

This paper proposes a new technique to estimate the center of mass (CoM) of mechanical systems defined by an articulated set of rigid bodies. This technique is based on the use of the statically equivalent serial chain, a serial chain representation on any multi-link branched chain. Through the use of this model, and without any knowledge of each individual body’s CoM or CoM location, a simple method to estimate the mechanical system’s CoM is developed. This method is validated with the CoM estimation of the Hoap-3 humanoid robot. A sensitivity calculation for estimating the CoM in this way is also presented.

Commentary by Dr. Valentin Fuster
2009;():969-978. doi:10.1115/DETC2009-86261.

This paper investigates the singular configurations of five-degree-of-freedom parallel mechanisms generating the 3T2R motion and comprising five identical legs of the RP UR type. The general mechanism was recently revealed by performing the type synthesis for symmetrical 5-DOF parallel mechanisms. In this study, some simplified designs are proposed for which the singular configurations can be predicted by means of the so-called Grassmann line geometry. This technique can be regarded as a powerful tool for analyzing the degeneration of the Plücker screw set. The main focus of this contribution is to predict the actuation singularity, for a general and simplified design, without expanding the determinant of the inverse Jacobian matrix (actuated constraints system) which is highly nonlinear and difficult to analyze.

Commentary by Dr. Valentin Fuster
2009;():979-986. doi:10.1115/DETC2009-86408.

It will be shown how to generate under-actuated manipulators by substituting non-holonomic spherical pairs (nS pairs) for (holonomic) spherical pairs (S pairs) in fully-parallel manipulators (FPMs). Through this pair substitution, an under-actuated manipulator, previously proposed by one of the authors, will be demonstrated to be generated from an inversion of the 6-3 FPM. Moreover, the kinetostatic analysis of this manipulator will be reconsidered to obtain a simple and compact formulation. This reformulated analysis can be used both in the design of the under-actuated manipulator, and in its control.

Topics: Robots , Manipulators , Design
Commentary by Dr. Valentin Fuster
2009;():987-992. doi:10.1115/DETC2009-86595.

Biped robots have multiple degrees of freedom for walking and hence they consume a lot of energy. In this paper it is proposed that adding torsion springs at the joints of an 8 DOF biped will lead to reduced energy consumption during walk. First the dynamic equations of motion of the biped robot are obtained incorporating the torsion springs at the joints. Using the dynamic model the total energy consumed during walk was evaluated for a single step. A Genetic Algorithm (GA) based algorithm was developed for finding the energy optimal trajectory during gait by comparing all the possible trajectories. It is first proved that addition of torsion springs at the joints lead to reduction of energy consumption as compared to a biped with no springs. All the gait parameters were then optimized to get the optimum values for the spring constants at each joint, reference angle of springs and length of each step. It is proved that using these optimal parameters the proposed biped robot consumes the least energy.

Commentary by Dr. Valentin Fuster
2009;():993-1002. doi:10.1115/DETC2009-86714.

From the viewpoint of kinematics, a three-dof Prism-type pure translational parallel robot is presented for the development of automatic assembly devices and a regional structure of a six-dof hybrid parallel platform. First, we describe the structural properties of the robot and analyze its kinematic mobility. A pure translational motion is verified to exist through the well-known D-H symbolic notations and the coordinate-transformation-matrix technique. What follows are the forward, inverse kinematics analysis, and their closed-form solutions by the matrix algebra approach. For further confirmation of the correctness of derived equations, some numerical examples are also given. With the help of the analytical displacement kinematics, we identify the volume of workspace. At last, taking account of the 3×3 reduced Jacobian matrix provides the condition number and the identification of singularity of configuration is explored based on the direct and inverse kinematics Jacobian matrix.

Topics: Kinematics , Robots
Commentary by Dr. Valentin Fuster
2009;():1003-1015. doi:10.1115/DETC2009-86912.

This paper presents the work addressing the issue of standing up after falling down for a novel three-legged mobile robot STriDER (Self-excited Tripedal Dynamic Experimental Robot). The robot is inherently stable when all three feet are on the ground due to its tripod stance, but it can still fall down if it trips while taking a step or if unexpected external forces act on it. The unique structure of STriDER makes the simple task of standing up challenging for a number of reasons; the high height of the robot and long limbs require high torque at the actuators due to its large moment arms; the joint configuration and length of the limbs limit the workspace where the feet can be placed on the ground for support; the compact design of the joints allows limited joint actuation motor output torque; three limbs do not allow extra support and stability in the process of standing up. This paper examines four standing up strategies unique to STriDER: three feet, two feet and one foot pushup, and spiral pushup. For all of these standing up strategies, the robot places its feet or foot at desired positions and then pushes the feet against the ground thus, lifting the body upwards. The four pushup methods for standing up were analyzed and evaluated considering the constraints such as, static stability, friction at the feet, kinematic configuration and joint motor torque limits, thus determining the suggested design and operation parameters. The motor torque trends as the robot stands up using pushup methods were investigated and the results from the analysis were validated through experiments.

Topics: Robots
Commentary by Dr. Valentin Fuster
2009;():1017-1025. doi:10.1115/DETC2009-86923.

This paper presents work on the gait and gait transition analysis for a novel mobile robot that uses two actuated spoke wheels. Gait transitions, known as acyclic feed forward patterns, allow the robot to switch from one type of gait to another during walking and turning. The mobile robot IMPASS (Intelligent Mobility Platform with Active Spoke System) uses a unique mobility concept for locomotion, thus gait transition plays an important role in generating gait patterns to walk and turn. The primary focus of this paper is how to perform gait transition between gaits in walking direction. First, the basic gait patterns for steering and straight line walking are presented. More specifically, the critical gait parameterizations and the possible foot positions in different gait patterns to produce capable steering locomotion over terrain are presented. Since IMPASS is expected to utilize its metamorphic configurations to carry out gait transitions, the extending forward and inverse analyses are also presented based on previous work about topology classification and mobility analysis for IMPASS. Then the gait transition analysis and simulation of typical patterns are performed. The results from this work lay the foundation for the future research on trajectory and path planning for IMPASS.

Topics: Robots , Wheels
Commentary by Dr. Valentin Fuster
2009;():1027-1036. doi:10.1115/DETC2009-86925.

IMPASS (Intelligent Mobility Platform with Active Spoke System) is a novel mobile robot which is driven by a pair of rimless spoke wheels which can alter the length of any given spoke in the hub. A highly mobile robot such as IMPASS could prove very valuable in applications where the terrain is complex and dangerous, but for this platform to be practical for real world use, motion control of the actuated rimless wheel must be automated. This work discusses considerations for motion planning concerning the transitions from step to step in the two-dimensional sagittal plane. Each step transition can be defined by a switching angle, which is the angle made between the back spoke and a reference axis. Presented is a review of the types of step transitions that are advantageous for ascending and descending obstacles, as well as traversing terrain with minor irregularities. These step transitions have been tested in simulation and on a physical prototype, the results of which will be discussed.

Commentary by Dr. Valentin Fuster
2009;():1037-1045. doi:10.1115/DETC2009-86934.

Hyper-redundant robots are characterized by the presence of a large number of actuated joints, many more than the number required to perform a given task. These robots have been proposed and used for many application involving avoiding obstacles or, in general, to provide enhanced dexterity in performing tasks. Making effective use of the extra degrees of freedom or resolution of redundancy have been an extensive topic of research and several methods have been proposed in literature. In this paper, we compare three known methods and show that an algorithm based on a classical curve called the tractrix leads to a more ‘natural’ motion of the hyper-redundant robot with the displacements diminishing from the end-effector to the fixed base. In addition, since the actuators at the base ‘see’ the inertia of all links, smaller motion of the actuators nearer to the base results in a smoother motion of the end-effector as compared to other two approaches. We present simulation and experimental results performed on a prototype eight link planar hyper-redundant manipulator.

Commentary by Dr. Valentin Fuster
2009;():1047-1057. doi:10.1115/DETC2009-86946.

In this paper, overall manipulability measure and stroke of workspace are proposed and evaluated as design criteria for optimal kinematics design of a family of industrial robots. The object of study is a 6 degree of freedom serial robot manipulator where individual family members (robots) share arms from a common platform. The paper presents a formal mathematical framework where the product family design problem is stated as an optimization problem and where optimization is used to find an optimal product family. The paper illustrates how the proposed kinematic design criteria may be used to support the optimal kinematics design of a family of industrial robots, and it also visualizes the tradeoff between the size of the common platform and the kinematics performance of individual robots.

Topics: Robots , Optimization
Commentary by Dr. Valentin Fuster
2009;():1059-1067. doi:10.1115/DETC2009-86987.

A new method for obstacle avoidance of underactuated unmanned surface vessels is presented which combines trajectory planning with real-time tracking control. In this method, obstacles are approximated and enclosed by elliptic shapes which represent the stable limit cycle solution of a special class of ODEs (ordinary differential equation). The vessel trajectory at any moment is defined by the ODEs whose solution is the limit cycle defining the obstacle immediately on its path to the target. When no obstacle remains on the vessel’s path, the trajectory is defined by exponentially stable ODEs whose solution is the target trajectory. The planned trajectories are tracked by the vessel through a sliding mode control law which is robust to environmental disturbances and modeling uncertainties and can be computed in real time. One advantage of the method is that it allows for dynamic (moving and rotating) obstacles as well as a moving target. Another advantage is that only the current information about the obstacles and the target are required for real-time trajectory planning. Since the vessel current position is used as feedback to redefine the limit cycle trajectories, the method is also robust to large disturbance.

Commentary by Dr. Valentin Fuster
2009;():1069-1077. doi:10.1115/DETC2009-87046.

Nowadays, the stiffness mapping for robot manipulators is established in static condition by means of lumped stiffness model, which is not accurate enough to describe the actual deformation when the manipulator is working in extreme environment with heavy and dynamic loads. This paper presents a kineto-static stiffness mapping derived by means of the general equation of dynamics for robot manipulators, which takes into account the kinematic influence on the stiffness matrix. At each discrete point of the predefined trajectory of the end-effector, inertial forces and torques related to the kinematic parameters are introduced and flexible links are modeled by the finite element analysis (FEA) method to obtain the accurate deformation of each body. The formulation reveals that the kineto-static stiffness matrix consists of the stiffness properties of actuators, transmission mechanisms, force elements (spring and damping) and flexible links and is related to the inertial loads as well as the configuration and applied loads. A planar 3-RPR (revolute-prismatic-revolute) manipulator with flexible links is studied here as an example to show that the stiffness matrix computed by the proposed kineto-static stiffness map is both comprehensive and feasible.

Commentary by Dr. Valentin Fuster
2009;():1079-1087. doi:10.1115/DETC2009-87050.

The objective of this research is to create a movable, palpable virtual model of the dynamic human upper body, including the spine, shoulders, and arms skeletal structure, with dynamic pivot point and deformable skin. This Virtual Haptic Human Upper Body (VHHUB) model has realistic human motion with anatomically-accurate joint motion limits and a 71 degrees-of-freedom branching serial chain model. The aim is to provide realistic motions when an osteopathic medical student moves the virtual patient for palpatory diagnosis training. Medical trainees can thus practice feeling changes in human tissue due to motions, a common diagnostic technique.

Topics: Haptics
Commentary by Dr. Valentin Fuster
2009;():1089-1094. doi:10.1115/DETC2009-87056.

This paper presents the design of one degree-of-freedom spatial mechanisms that use non-circular gears to constrain the motion. In a spatial body-guidance problem, representing the motion by systems of polynomial equations restricts the number of end-effector positions and orientations (end-effector poses) that can be used as inputs for mechanism design. An approach has been developed that takes any number of desired poses as guide points and develops a mechanism that approximately attains the desired poses over the course of its motion. A problem with implementing this design strategy is the inherent difficulty in accounting for orientation and position errors. The approach described here addresses this problem by defining a new error functional, calculated in the joint space domain. As the mechanisms being dealt with are single degree-of-freedom closed chains, the starting position is a crucial decision in the design process. The method outlines the choice of the starting position and details how this error term can be used along with optimization techniques on either the mechanism parameters or the non-circular gears. A numerical example is presented.

Commentary by Dr. Valentin Fuster
2009;():1095-1103. doi:10.1115/DETC2009-87076.

Intelligent Mobility Platform with Active Spoke System (IMPASS) is a novel wheel-leg hybrid robot that can walk in unstructured environments by stretching in or out three independently actuated spokes of each wheel. This paper firstly introduces the latest prototype of the robot that has two actuated spoke wheels and one tail. Due to its metamorphic configurations, this robot can be considered as a mechanism with variable topologies (MVTs). The mobility analysis in previous work is briefly summarized and then the transformation relationships between different configurations are addressed. Thus, the motions of the robot on the ground, such as steering, straight-line walking and other combinations can be uniformly interpreted as series of configuration transformations. In order to maintain the coherence between the coordinate setups in different configurations, previous rigid body kinematic models are modified. Two types of turning, steady state turning and turning gait transition, are revisited with the characteristics being refined. Finally, the hardware design of the prototype is demonstrated, followed by the experimental verifications of the characteristic motions. The work presented in this paper lays the kinematics foundation for the future research on the motion planning algorithms of IMPASS.

Topics: Robots , Wheels
Commentary by Dr. Valentin Fuster
2009;():1105-1114. doi:10.1115/DETC2009-87415.

This paper formulates the kinematic specification of the synthesis problem for planar and spatial open serial chains in which the acceleration of the end-effector is specified. While the planar RR synthesis has been reported by the authors in some of their previous work, the synthesis of the spatial perpendicular RRS mechanical linkage with prescribed acceleration task is described in details in this paper and is a new contribution. Applications of this research focus on the design of planar RR and four-bar linkages to maintain specified local motion. In extending the research to spatial applications, a new strategy for failure recovery of a general six degree of freedom TRS robot arm is presented. It combines the second order effects of the task with the particular joint failure to yield free parameters that allow reconfiguration of the system to accommodate the failure.

Topics: Linkages , Design
Commentary by Dr. Valentin Fuster
2009;():1115-1121. doi:10.1115/DETC2009-87467.

A quadratic parallel manipulator refers to a parallel manipulator with a quadratic characteristic polynomial. This paper revisits the forward displacement analysis (FDA) of a quadratic spherical parallel manipulator: the Agile Eye. An alternative formulation of the kinematic equations of the Agile Eye is proposed. The singularity analysis of the Agile Eye is then dealt with. After an alternative solution to the FDA has been presented, a formula is revealed for producing a unique current solution to the FDA for a given set of inputs. A regular cube in the input space, which is singularity free, is also proposed for the Agile Eye. This work will facilitate the control of the Agile Eye.

Commentary by Dr. Valentin Fuster
2009;():1123-1137. doi:10.1115/DETC2009-87470.

Modeling protein molecules as kinematic chains provides the foundation for developing powerful approaches to the design, manipulation and fabrication of peptide based molecules and devices. Nevertheless, these models possess a high number of degrees of freedom (DOF) with considerable computational implications. On the other hand, real protein molecules appear to exhibits a much lower mobility during the folding process than what is suggested by existing kinematic models. The key contributor to the lower mobility of real proteins is the formation of Hydrogen bonds during the folding process. In this paper we explore the pivotal role of Hydrogen bonds in determining the structure and function of the proteins from the point of view of mechanical mobility. The existing geometric criteria on the formation of Hydrogen bonds are reviewed and a new set of geometric criteria are proposed. We show that the new criteria better correlate the number of predicted Hydrogen bonds with those established by biological principles than other existing criteria. Furthermore, we employ established tools in kinematics mobility analysis to evaluate the internal mobility of protein molecules, and to identify the rigid and flexible segments of the proteins. Our results show that the developed procedure significantly reduces the DOF of the protein models, with an average reduction of 94%. Such a dramatic reduction in the number of DOF can have has enormous computational implications in protein folding simulations.

Commentary by Dr. Valentin Fuster
2009;():1139-1146. doi:10.1115/DETC2009-87480.

Mobility analysis of multi-DOF multiloop planar linkages is much more complicated than the single-DOF planar linkages and has been little explored. This paper offers a unified method to treat the singularity (dead center position) and sub-branch identification of the planar two-DOF seven-bar linkages regardless of the choice of the inputs or fixed links. This method can be extended for the singularity analysis of other multi-DOF multiloop linkages. Based on the concept of joint rotation space and N-bar rotatability laws, this paper presents a general method for the sub-branch identification of the seven-bar linkages. It offers simple explanation and geometric insights for the formation of branch, singularity and sub-branch of the two-DOF seven-bar linkages. The presented algorithm for sub-branch identification is suitable for automated computer-aided mobility identification. Examples are employed to demonstrate the proposed method.

Commentary by Dr. Valentin Fuster
2009;():1147-1155. doi:10.1115/DETC2009-87505.

This paper presents a novel trilateration algorithm which estimates the position of a target object, such as a mobile robot, in a 2D or 3D space based on the simultaneous distance measurements from multiple reference points. The proposed algorithm is derived from the nonlinear least-squares formulation of trilateration, and provides a globally optimal position estimate from a general number of reference points and corresponding distance measurements. Using standard linear algebra techniques, the proposed algorithm has relatively low computational complexity and high operational robustness. Simulations have been conducted through representative examples to analyze the performance of the proposed trilateration algorithm in dealing with erroneous inputs of reference points and distance measurements. The results show that the proposed algorithm has lower systematic error and uncertainty in position estimation comparing with representative closed-form methods.

Commentary by Dr. Valentin Fuster
2009;():1157-1165. doi:10.1115/DETC2009-87577.

This paper investigates the mobility of a family of 3-DOF parallel manipulators using screw theory. Based on the 3-P SP, 3-P PS and 3-P CU manipulators and a new 3-CUP parallel manipulator, the paper obtains the branch motion-screws for these architectures and determines the sets of platform constraint-screws. The constraints are identified to be the forces acting on the S and U joints fixed in the platform and the couples acting about an axis perpendicular to the base. The mobility of a manipulator platform is thus obtained by determining the reciprocal screws to the platform constraint screw sets and the platforms are identified to have three instantaneous independent degrees of freedom which are: (i) a translation along an axis perpendicular to the base; and (ii) two rotations about two skew axes.

Topics: Screws , Manipulators
Commentary by Dr. Valentin Fuster
2009;():1167-1174. doi:10.1115/DETC2009-87759.

Finding optimal trajectory is critical in several applications for robots from payload transport between two given states in a prescribed time such that a cost functional is minimized. This paper is concerned with the path planning of flexible robotic arms in point-to-point motion, based on indirect solution of optimal control problem. Dynamic modelling technique based on the combined Euler–Lagrange formulation and assumed modes method is applied, then by implementing the Pontryagin’s minimum principle; necessary conditions for optimality are derived. Nonlinear states and control constraints are treated without any simplifications or transforming them into sequences of systems with linear equations. By these means, the modelling of the complete optimal control problem and the accompanying boundary value problem is automated to a great extent. Finally, the performance of method is illustrated through the computer simulations.

Commentary by Dr. Valentin Fuster
2009;():1175-1182. doi:10.1115/DETC2009-87806.

Mobility identification mainly refers to the problems with the motion continuity and smoothness of a potential design or plan. In any linkage synthesis or robot navigation, it is highly desirable that the ability of any of the numerous design candidates to reach the desired positions in a favorable manner can be determined in a single decisive step automatically rather than through a blind trial or even a physical experiment. Mobility of complex linkages has been one of the most troublesome problems in linkage synthesis and programming and the problem is further complicated with multiple degrees-of-freedom. For multiloop parallel manipulators this paper may represent the first mobility analysis method that can not only decisively and unambiguously rectify motion continuity between discrete positions but also provide clear geometric insight or interpretation regarding the formation of discontinuity. The treatment is based on the principle that the mobility of a multiloop linkage is affected by the mobility of each individual loop as well as the interaction between loops. Since the N-bar rotatability laws govern the mobility of an individual loop, the main mobility issue for multiloop linkages is how the mobility of these loops affects each other. One may find that the concept of joint rotation space (JRS) offers simple and intuitive explanation on how the mobility is affected by the combination of loops. The treatment is very suitable for an automated computer-aided mobility analysis. Examples are employed to demonstrate the proposed method. Continuity is a pivotal issue in linkage mobility analysis. Once the continuity can be rectified, problems with smoothness or singularity, which are discussed in the companion paper [28], can be resolved.

Topics: Linkages , Bifurcation
Commentary by Dr. Valentin Fuster
2009;():1183-1194. doi:10.1115/DETC2009-86106.

Based on previous research results presented by authors, this paper proposes a novel systematic approach for structure synthesis of all parallel mechanisms (excluding Bennett mechanism etc), which is totally different from the approaches based on screw theory and based on displacement subgroup. Main characteristics of this approach are: (a) the synthesized mechanisms are non-instantaneous ones, and (b) only simple mathematical tools (vector algebra, theory of sets, etc.) are used. Main steps of this approach include: (1) Determining functional and structural requirements of the parallel mechanism to be synthesized, such as position and orientation characteristic (POC) matrix, degree of freedom (DOF), etc. (2) Type synthesis of branches. (3) Assembling of branches (determining the geometry constraint conditions among the branches attached between the moving platform and the frame, and checking the DOF). (4) Identifying the inactive joints. (5) Selecting the actuating joints. In order to illustrate the whole procedure, the type synthesis of spherical parallel mechanisms is studied using this approach.

Commentary by Dr. Valentin Fuster
2009;():1195-1204. doi:10.1115/DETC2009-86489.

In recent years, there has been a good deal of controversy about the application of infinitesimal kinematics to the mobility determination of kinematic chains. On the one hand, there has been several publications that promote the use of the velocity analysis, without any additional results, for the determination of the mobility of kinematic chains. On the other hand, the authors of this contribution have received several reviews of researchers who have the strong belief that no infinitesimal method can be used to correctly determine the mobility of kinematic chains. In this contributions, it is attempted to show that velocity analysis by itself can not correctly determine the mobility of kinematic chains. However, velocity and higher order analysis coupled with some recent results about the Lie algebra, se(3) , of the Euclidean group, SE(3) , can correctly determine the mobility of kinematic chains.

Topics: Kinematics , Chain
Commentary by Dr. Valentin Fuster
2009;():1205-1213. doi:10.1115/DETC2009-86698.

In synthesizing a planar 4R linkage that can achieve four positions, the fixed pivots are constrained to lie on a center-point curve. It is widely known that the curve can be parameterized by a 4R compatibility linkage. In this paper, a slider crank is presented as a suitable compatibility linkage to generate the centerpoint curve. Further, the center-point curve can be parametrized by the crank angle of a slider crank linkage. It is observed that the center-point curve is dependent on the classification of the slider crank. Lastly, a direct method to calculate the focus of the center-point curve is revealed.

Topics: Linkages
Commentary by Dr. Valentin Fuster
2009;():1215-1222. doi:10.1115/DETC2009-86863.

Since the native conformation or the natural shape of a protein largely determines its function, a prediction of protein conformation can shorten the process of drug discovery. This prediction is an optimization search to locate a configuration associated with the global minimum energy for the molecule. Due to the complexity of the multidimensional energy landscape, the prediction process can be extensive, which leads to very long simulation run times. For example, a high-resolution structure prediction algorithm [1] refining 20,000 to 30,000 models of several 49 to 88 residue long molecules takes about 150 CPU days per molecule. This paper presents the method of modified energy landscape (MEL) that improves the efficiency of the Broyden–Fletcher–Goldfarb–Shanno (BFGS) method by 12.8% on average, and more than 30% in some cases for a representative sample of cases. Since the efficiency improvement allows the probabilistic search to cover more areas of the energy landscape, locating the global minimum is more likely. Also, in a practical protein prediction running coarse refinements on more decoys is more preferable than comprehensively refining few decoys because of the low accuracy of energy functions. Therefore, the MEL can significantly improve the protein prediction simulation even though it yields less average score improvement. The MEL is implemented in a refinement protocol in Rosetta [2].

Topics: Proteins
Commentary by Dr. Valentin Fuster
2009;():1223-1230. doi:10.1115/DETC2009-87205.

Variable topology mechanisms can serve many design functions with only one mechanism through changing their topology, these mechanisms have raised broad interest and attracted many studies in recent years, yet few have consolidated the different types of these mechanisms, nor discussed them in the light of the space they operate in. This work classified the variable topology mechanisms, and presented an expression of the mechanism’s working space. Variable topology mechanisms are classified into three types, topology changed by intrinsic constraints, topology changed by joint geometry change, and topology changed by external constraints. The causes and effects of various constraints inducing a topology change are described with the operating space, compatibility characteristics of joints, loops, and working stages with the operating space are established, verifying whether joints will constraint and lock up each other. The admissible operating space for a loop interface pairs so that loops are compatible, and the requisites of different working stages being workable with each other are identified. As a result, some basic requirements for admissible variable topology mechanisms are unveiled, laying a foundation stone for systematical synthesis of variable topology mechanisms.

Topics: Topology , Mechanisms
Commentary by Dr. Valentin Fuster
2009;():1231-1236. doi:10.1115/DETC2009-87241.

This paper deals with the position analysis of a multiple-mode 7R spatial linkage. The 7R linkage is constructed by combining two Bennett linkages with a common joint. As a result, the motions of the Bennett linkages are embedded in the multiple-mode 7R linkage, and there are three operation modes of the linkage: two Bennett modes and one 7R mode. The algorithms for solving the inverse kinematics of serial 6R manipulators are employed to find all possible configurations of the 7R linkage at a given value of the driving joint angle. The analysis is then conducted for a full rotation of the driving joint. The result of the analysis gives all feasible configurations of the linkage for a full rotation of the driving joint. According to the result of analysis, the motion of the linkage can be clearly divided into the three operation modes, and the linkage can switch between two modes at various connecting configurations. Furthermore, because of the two Bennett modes, the linkage can reach all feasible configurations without being taken apart and reassembled. This paper provides insight into the kinematics of multiple-mode single-loop 7R linkages constructed by combining paradoxic linkages.

Topics: Linkages
Commentary by Dr. Valentin Fuster
2009;():1237-1244. doi:10.1115/DETC2009-87260.

A method to calculate the volume of constant-orientation taskspace of parallel manipulators is developed and applied on the 6-RSS mechanism as an example. The taskspace is defined as a singularity-free reachable workspace. Quasi-singularity measure is used to measure the closeness to singular configurations and construct the quasi-singularity zone. Analyzed by DOE (Design of Experiment), effect weighs of design variables of a parallel manipulator are revealed and experiment results are applied in the optimization process. In order to obtain the maximum taskspace, the volume of the constant-orientation taskspace is optimized by a structured optimization process aided by iSIGHT-FD software. Finally, constant-orientation taskspace of greater volume is obtained.

Commentary by Dr. Valentin Fuster
2009;():1245-1254. doi:10.1115/DETC2009-87329.

This paper investigates geometry and kinematics of the Hoberman switch-pitch ball and its variant as an extended case of the ball. The paper starts from examining the geometry of the ball variant and its composition and decomposes it into loops each of which is an eight-bar radial linkage. Based on this, the paper investigates the geometry of the eight-bar radial linkage and the variant and subsequently extends the study to their kinematics. The Hoberman switch-pitch ball as a special case of the ball variant with bevel gears is investigated and a numerical example is employed to illustrate the kinematic characteristics of the eight-bar radial linkage and the Hoberman switch-pitch ball.

Topics: Switches
Commentary by Dr. Valentin Fuster
2009;():1255-1263. doi:10.1115/DETC2009-87372.

Spatial linkages are classified into four groups according to the number of fundamental equations or virtual loops that govern linkage displacement. The number of virtual loops represents the complexity of a spatial linkage as that of planar or spherical multiloop linkages. The concept of generalized branch points offers the explanation of how branches are formed in spatial group 2 linkages. In this paper, the mobility analysis is carried out based on the similarity of the mobility features rather than the specific or individual linkage structure. A branch rectification scheme is presented and demonstrated with examples.

Commentary by Dr. Valentin Fuster
2009;():1265-1273. doi:10.1115/DETC2009-87384.

A spatial linkage with the displacement governed by two fundamental equations can be regarded as a virtual double loop system. The mobility of the linkage is affected by the mobility of each individual “loop” as well as the interaction between the loops. The current use of branch points for branch identification is limited to linkages with simple topology, such as Stephenson-type linkages, which are simplified versions of group 2 mechanisms. However, in a general spatial group 2 linkage, both the fundamental equations are equivalent to virtual five-bar loops. Branch points in Stephenson-type linkages should be generalized to explain and define the interaction between two virtual five-bar loops. The concept of generalized branch points offers the explanation of how branches are formed in spatial group 2 linkages. This paper presents the theoretical background for the mobility analysis of complex spatial linkages.

Topics: Linkages , Bifurcation
Commentary by Dr. Valentin Fuster
2009;():1275-1284. doi:10.1115/DETC2009-87524.

This paper summarizes the concept of mobility used for holonomic and non-holonomic mechanisms. The mobility of mechanisms is considered from a geometric viewpoint starting with the variety generated by the constraint mapping as configuration space. While the local (finite) mobility is determined by the dimension of the configuration space, the differential mobility may be different. This is so for singular configurations, but also at regular configurations of underconstrained mechanisms. Overconstrained mechanisms are identified as those comprising manifolds of regular configurations that are critical points of the constraint mapping. The considerations include non-holonomic mechanisms. For such mechanisms the configuration space is the integral manifold of the kinematic constraints. Different types of singularities are discussed for non-holonomic mechanisms.

Commentary by Dr. Valentin Fuster
2009;():1285-1294. doi:10.1115/DETC2009-87653.

The current paper shows that all the planar linkages can be constructed from given components called, Assur Graphs, which can be ordered in a table with infinite numbers of rows and columns. In the paper we term this order the canonical form of the planar linkages. This canonical form is proved to be an ordered hierarchy of several levels, enabling systematic generation of all its members. The work has originated from the concept of Assur groups, long known in the field of kinematics, and used to decompose any linkage into basic kinematical atoms. In this paper we introduce a systematic procedure for generating all Assur groups thus finding all the topologies of plane linkages. The work in 2D is based upon known but new mathematical theorems which prove it to be complete and sound. The paper also indicates how this work can be extended into 3D linkages. The mathematical foundation of this work contains several new theorems that have been published by the rigidity theory community during the past six years.

Topics: Linkages
Commentary by Dr. Valentin Fuster
2009;():1295-1304. doi:10.1115/DETC2009-87718.

In the last twenty years, researchers have proposed a few different methods to establish a norm (or, metric) for both planar and spatial rigid body displacements. Desire to meaningfully quantify a displacement composed of rotation and translation stems from a requirement to ascertain “distance” between two given displacements in applications, such as motion approximation and interpolation, mechanism synthesis, collision avoidance, positioning, and robot calibration and control. In this paper, we show that the various seemingly different shape independent norm calculation methods based on approximating displacements with higher dimensional rotations via orthogonal matrices, or polar decomposition (PD) and singular value decomposition (SVD) can be reconciled and unified in the mathematically compact and elegant framework of biquaternions. In the process, we also propose an elegant and fast method for such norm calculations.

Commentary by Dr. Valentin Fuster
2009;():1305-1313. doi:10.1115/DETC2009-87812.

This paper explores the concept of kinematic convexity of planar displacements as an extension of the projective convexity in computational geometry to planar kinematics. This is achieved with the help of planar quaternions which converts planar displacements into points in the space of planar quaternions called the image space. In this way, projective convexity of points in the image space is developed and used as a representation of kinematic convexity of planar displacements. To address the issue of distance metric for planar displacements, we explored the connection between planar quaternions and quaternions and formulated the concept of kinematic convexity in the space of quaternions where a bi-invariant metric exists. An example is provided in the end to illustrate the use of kinematic convexity for estimating the “closest distance” from a fixed body to a moving body undergoing a rational Bézier motion.

Commentary by Dr. Valentin Fuster

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In