ASME Conference Presenter Attendance Policy and Archival Proceedings

2013;():V003T00A001. doi:10.1115/DSCC2013-NS3.

This online compilation of papers from the ASME 2013 Dynamic Systems and Control Conference (DSCC2013) represents the archival version of the Conference Proceedings. According to ASME’s conference presenter attendance policy, if a paper is not presented at the Conference, the paper will not be published in the official archival Proceedings, which are registered with the Library of Congress and are submitted for abstracting and indexing. The paper also will not be published in The ASME Digital Collection and may not be cited as a published paper.

Commentary by Dr. Valentin Fuster

Nonlinear Control

2013;():V003T34A001. doi:10.1115/DSCC2013-3722.

The paper addresses the invariance control for a class of cascade nonlinear systems with unmodeled dynamics appearing at the input. A sufficient condition for the robust invariance control of the system under consideration is derived. Based on the methods of passivity and switching control of states of the linear subsystem, both the stabilization of the linear subsystem and the positive invariance of the prespecified region in state space can be ensured. Under some additional assumptions, the whole system is semi-global asymptotically stable. A simulation example is given to demonstrate the effectiveness of the proposed design procedure.

Commentary by Dr. Valentin Fuster
2013;():V003T34A002. doi:10.1115/DSCC2013-3741.

Measures to reduce control performance for greater robustness (e.g. reduced bandwidth, shallow loop roll-off) must be enhanced if the plant or actuators are known to have nonlinear characteristics that cause variations in loop transmission. Common causes of these nonlinear behaviors are actuator saturation and friction/stiction in the moving parts of mechanical systems. Systems with these characteristics that also have stringent closed loop performance requirements present the control designer with an extremely challenging problem. A design method for these systems is presented that combines very aggressive Nyquist-stable linear control to provide large negative feedback with nonlinear feedback to compensate for the effects of multiple nonlinearities in the loop that threaten stability and performance. The efficacy of this approach is experimentally verified on a parallel kinematic mechanism with multiple uncertain nonlinearities used for vibration suppression.

Topics: Actuators , Feedback
Commentary by Dr. Valentin Fuster
2013;():V003T34A003. doi:10.1115/DSCC2013-3761.

We introduce, through an analysis overall restricted, for the sake of simplicity, in two-dimensions, the class of proportional systems, a nice subclass of the ΣΠ-algebraic nonlinear systems that we formerly introduced in another paper as a sort of ‘non-linear paradigm’ linking nonlinear to bilinear systems. Also we define a decomposition, which every ΣΠ-algebraic system undergoes, into the cascade of a driver, medial and final bilinear subsystem, having the same input-output behavior as the original. We show that a systematic way for global feedback stabilization can be developed for the class of proportional systems, leading to the global feedback exponential stabilization of the medial part under some ‘natural’ condition of non singularity. We show in an example the capability of the proposed method to achieving global feedback stabilization for the original system as well.

Commentary by Dr. Valentin Fuster
2013;():V003T34A004. doi:10.1115/DSCC2013-3807.

In this paper, the UDE (uncertainty and disturbance estimator) based robust control is investigated for a class of non-affine nonlinear systems in a normal form. Control system design for non-affine nonlinear systems is one of the most difficult problems due to the lack of mathematical tools. This is also true even for the exact known non-affine systems because of the difficulty in explicitly constructing the control law. It is shown that the proposed UDE-based robust control strategy leads to a stable system. The most important features of the approach are that (i) by adding and subtracting the control term u, the original non-affine form is transformed into a semi-affine form, which not only simplifies the control design procedure, but also avoids the singularity problem of the controller; (ii) the employment of UDE makes the estimation of the lumped uncertain term which is a function of control input, states and disturbances possible, rather than states alone; and (iii) it does not require any knowledge (e.g., bounds) about the uncertainties and disturbances, except the information about the bandwidth, during the design process. The stability of the closed-loop system is established. Effectiveness of the proposed approach is demonstrated through application to the hard disk driver control problem.

Commentary by Dr. Valentin Fuster
2013;():V003T34A005. doi:10.1115/DSCC2013-4039.

The development and implementation of an experimental sliding mode control law for a 2.5 meter long unmanned amphibious vehicle (the DUKW-Ling) when waterborne is presented. A first-order sliding control surface is used for surge tracking error when a P controller is used to minimize the heading error in the system. The state of the vehicle is measured using onboard sensors with the ability to record surge, sway, yaw and position of the vehicle in real-time. Experimental data collected show the ability of the vessel to maintain a desired heading and speed. This article emphasizes the ability of sliding mode controller to respond to the unpredictable and random water and wind currents acting on the vehicle.

Topics: Vehicles
Commentary by Dr. Valentin Fuster
2013;():V003T34A006. doi:10.1115/DSCC2013-4047.

In this paper, a novel approach to controller design for non-linear multi–input/multi–output (MIMO) systems is presented based on the Contoured Robust Controller Bode (CRCBode) plot. CRCBode plots show level–sets of a robust metric and identify certain “forbidden regions” on the controller Bode magnitude and phase plots such that intersections of the controller frequency response with these forbidden regions indicate that a robust stability and performance criterion is violated. Nonlinear system dynamics are included as a structured uncertainty set consisting of linearizations about several operating points. To demonstrate this approach, we design a controller for a MIMO high–speed, low–tension magnetic tape drive memory system. A preliminary approximate inverse step is described, followed by several loop–shaping design iterations to eliminate all intersections with the forbidden regions on the CRCBode diagrams. Finally, the CRCBode compensator is compared to one generated using an automated H synthesis algorithm.

Commentary by Dr. Valentin Fuster

Nonlinear Estimation and Control

2013;():V003T35A001. doi:10.1115/DSCC2013-3702.

An adaptive control methodology with a low-resolution encoder feedback is presented for a biomedical application, the Ros-Drill (Rotationally Oscillating Drill). It is developed primarily for ICSI (Intra-Cytoplasmic Sperm Injection) operations, with the objective of tracking a desired oscillatory motion at the tip of a microscopic glass pipette. It is an inexpensive set-up, which creates high-frequency (higher than 500 Hz) and small-amplitude (around 0.2 deg) rotational oscillations at the tip of an injection pipette. These rotational oscillations enable the pipette to drill into cell membranes with minimum biological damage. Such a motion control procedure presents no particular difficulty when it uses sufficiently precise motion sensors. However, size, costs and accessibility of technology on the hardware components severely constrain the sensory capabilities. Consequently the control mission and the trajectory tracking are adversely affected. This paper presents a dedicated novel adaptive feedback control method to achieve a satisfactory trajectory tracking capability. We demonstrate via experiments that the tracking of the harmonic rotational motion is achieved with desirable fidelity.

Commentary by Dr. Valentin Fuster
2013;():V003T35A002. doi:10.1115/DSCC2013-3783.

This paper is concerned with disturbance rejection performance in single-input single-output (SISO) nonlinear systems that are described by uncertain linear dynamics and bounded nonlinearities. First, the nonlinear terms are transformed into an equivalent bounded disturbance at the output of a linear system. Then, a disturbance observer (DOB) is added to the closed loop to achieve robust disturbance rejection. The DOB design is formulated as an extended Luenberger observer having internal dynamics with at least an eigenvalue at the origin. The synthesis of a (sub)optimal DOB is carried out by solving multi-objective H sensitivity optimization. The design approach is applied to an inverted pendulum with actuator backlash. Closed loop response shows that tracking performance is indeed greatly enhanced with the DOB.

Commentary by Dr. Valentin Fuster
2013;():V003T35A003. doi:10.1115/DSCC2013-3884.

This paper presents the modeling and study of dynamic behavior of a backhoe machine for tuning of PID controller. The tuning procedure of PID controller is performed, in detailed, for the case of a typical operation, digging a foundation and truck loading. This tuning procedure guarantees the local asymptotic stability in the sense of Lyapunov of origin of the closed-loop equation of PID controller. Besides the tuning procedure requires the knowledge of certain properties of dynamic model, which are dependent on the desired trajectory. Finally it is demonstrated that this tuning procedure proves to be effective, and also robust in the execution of other tasks performed by the backhoe machine.

Commentary by Dr. Valentin Fuster
2013;():V003T35A004. doi:10.1115/DSCC2013-3966.

In this paper, we present a technique for estimating the input nonlinearity of a Hammerstein system by using multiple orthogonal ersatz nonlinearities. Theoretical analysis shows that by replacing the unknown input nonlinearity by an ersatz nonlinearity, the estimates of the Markov parameters of the plant are correct up to a scalar factor, which is related to the inner product of the true input nonlinearity and the ersatz nonlinearity. These coefficients are used to construct and estimate the true nonlinearity represented as an orthogonal basis expansion. We demonstrate this technique by using a Fourier series expansion as well as orthogonal polynomials. We show that the kernel of the inner product associated with the orthogonal basis functions must be chosen to be the density function of the input signal.

Commentary by Dr. Valentin Fuster
2013;():V003T35A005. doi:10.1115/DSCC2013-3973.

In this paper, a model-predictive control (MPC) method is detailed for the control of nonlinear systems with stability considerations. It will be assumed that the plant is described by a local input/output ARX-type model, with the control potentially included in the premise variables, which enables the control of systems that are nonlinear in both the state and control input. Additionally, for the case of set point regulation, a suboptimal controller is derived which has the dual purpose of ensuring stability and enabling finite-iteration termination of the iterative procedure used to solve the nonlinear optimization problem that is used to determine the control signal.

Topics: Stability
Commentary by Dr. Valentin Fuster
2013;():V003T35A006. doi:10.1115/DSCC2013-4104.

This paper discusses optimal and robust observer design for the Lipschitz nonlinear systems. The stability analysis for the Lure problem is first reviewed. Then, a two-DOF nonlinear observer is proposed so that the observer error dynamic model can be transformed to an equivalent Lure system. In this framework, the difference of the nonlinear parts in the vector fields of the original system and observer is modeled as a nonlinear memoryless block that is covered by a multivariable sector condition or an equivalent semi-algebraic set defined by a quadratic polynomial inequality. Then, a sufficient condition for asymptotic stability of the observer error dynamics is formulated in terms of the feasibility of polynomial matrix inequalities (PMIs), which can be solved by Lasserre’s moment relaxation. Furthermore, various quadratic performance criteria, such as H2 and H, can be easily incorporated in this framework. Finally, a parameter adaptation algorithm is introduced to cope with the parameter uncertainty.

Commentary by Dr. Valentin Fuster

Optimization and Optimal Control

2013;():V003T36A001. doi:10.1115/DSCC2013-3735.

In order to solve the problems arising from the manual calibration method in the developing process of vehicle automatic transmission control unit (TCU), known as time-consuming, heavy workload, high cost and over-dependence on subjective experience, this article researches on a virtual calibration method based on an approximate model to obtain optimal parameters for TCU. The neural network approximate model is established from the test data chosen with the method of DoE (Design of Experiment). The virtual calibration method is then conducted through Optimal Latin Hypercube Design (OLHD) and multi-island genetic algorithm (MIGA) to search the optimal parameters. By comparing the new calibration method with original manual one on the condition of gear 1 up to gear 2, the result shows that the new method can increase the efficiency significantly.

Commentary by Dr. Valentin Fuster
2013;():V003T36A002. doi:10.1115/DSCC2013-3821.

This paper revisits the sensitivity minimization approach with new formulation and synthesis methodology. The new robust control synthesis approach is presented here with a numerical example of an application to a header height control of combine harvester machine. The proposed synthesis methodology is compared with the H control design by comparing the performance of controllers designed using both techniques. The results indicate that the proposed methodology provides a viable alternative for robust controller synthesis and can even be better in robust performance in many cases.

Commentary by Dr. Valentin Fuster
2013;():V003T36A003. doi:10.1115/DSCC2013-3845.

The paper treats a class of parameter-dependent optimization/root finding problems where the minimizer or a real root need to be determined as a function of parameter. Applications of parameter-dependent optimization include spacecraft debris avoidance, adaptive control of Hybrid Electric Vehicles, engine mapping and model predictive control. In these and other problems, the parameter changes can be controlled either directly or indirectly. In this paper, the error analysis of a dynamic predictor-corrector Newton’s type algorithm is presented. Based on this analysis, an approach to govern the changes in the parameter to enable the algorithm to track the minimizer within an acceptable error bound is described. Two simulation examples are presented. In the first example the objective is to minimize the distance between points on a curve and a given set and simultaneously move as fast as possible along the given curve. In the second example we illustrate the use of this technique for aircraft flight envelope estimation. Specifically, we estimate maximum speed of an aircraft as a function of its altitude and flight path angle.

Topics: Optimization
Commentary by Dr. Valentin Fuster
2013;():V003T36A004. doi:10.1115/DSCC2013-3916.

We formulate the problem of an autonomous agent team facing the attack of an adversarial agent as a single-pursuer-multiple-evader pursuit-evasion game, with the assumption that the pursuer is faster than all evaders. In this game, the pursuer aims to minimize the capture time of the last surviving evader, while the evaders as a team cooperate to maximize this time. We present a gradient-based approach that quickly computes the controls for the evaders as a team under an open-loop formulation that is conservative towards the evader team by deriving analytical formulas. We demonstrate the advantage of the gradient-based approach by comparing performance both in computation time and in optimality with the iterative open-loop method studied in our previous work. Multiple heuristics have been designed to deal with the inherent intractability of evaluating all possible capture sequences. Extensive simulations have been performed, with results discussed.

Topics: Teams
Commentary by Dr. Valentin Fuster
2013;():V003T36A005. doi:10.1115/DSCC2013-3969.

The Prandtl-Ishlinskii (PI) model is a popular hysteresis model that has been widely applied in smart materials-based systems. Recently, a generalized PI model is formulated that is capable of characterizing asymmetric, saturated hysteresis. The fidelity of the model hinges on accurate representation of envelope functions, play operator radii, and corresponding weights. For a given number of play operators, existing work has typically adopted some predefined play radii, the performance of which could be far from optimal. In this paper, novel schemes based on entropy and relative entropy (Kullback-Leibler divergence) for optimal compression of a generalized PI model are proposed to best represent the original hysteresis model subject to a given complexity constraint, i.e., the number of play operators. The overall compression performance is expressed as a cost function, and is optimized using dynamic programming. The proposed compression schemes are applied to the modeling of the asymmetric hysteresis between resistance and temperature of a vanadium dioxide (VO2) film, and the effectiveness is further demonstrated in a model verification experiment. In particular, under the same complexity constraint, an entropy-based compression scheme and a Kullback-Leibler divergence-based compression scheme result in modeling errors around 37% and 48%, respectively, of that under a uniform compression scheme.

Topics: Modeling , Compression
Commentary by Dr. Valentin Fuster
2013;():V003T36A006. doi:10.1115/DSCC2013-4096.

A major class of extremum seeking control is based on the use of periodic dither perturbation of plant input for extracting the gradient information. Presence of the dither input into the steady state operation is undesirable in practice due to the possible excessive wear of actuators. It is thus beneficial to stop the dithering action after the extremum seeking process reaches its steady state. In this paper, we propose a method for automatically discriminate between the steady state and the transient state modes of extremum seeking control process using the sinusoidal detection techniques. Some design guidelines are proposed for the parameter selection of the relevant sinusoidal detection scheme. The proposed scheme is validated with simulation study.

Topics: Steady state
Commentary by Dr. Valentin Fuster

Piezoelectric Actuation and Nanoscale Control

2013;():V003T37A001. doi:10.1115/DSCC2013-3865.

This paper describes a nonlinear command tracking scheme for an electrostatic laser scanning micromirror assembly. The results are based on an innovative gimballed comb transducer concept developed at the Fraunhofer Institute for Photonic Microsystems. The outer mirror axis is designed as a Staggered Vertical Comb (SVC) in out-of-plane configuration and it shall provide a quasistatic operation with large deflection angles for triangular trajectories. The challenges for trajectory design and open loop command tracking are determined by the inherently nonlinear transducer characteristics and the lightly damped mass-spring dynamics. In this paper a flatness-based trajectory design is presented that considers the nonlinear transducer dynamics as well as the nonlinear elastic mechanical suspension with model parameters derived from ANSYS analysis. The paper discusses design constraints and detailed design considerations and it shows proof of concept performance results based on experimental verification with a real microscanner assembly.

Commentary by Dr. Valentin Fuster
2013;():V003T37A002. doi:10.1115/DSCC2013-3869.

A methodology for designing a low-computation, high-bandwidth strategy for closed-loop control of a hysteretic system without a priori knowledge of the desired trajectory is presented. The resulting two degree-of-freedom hysteresis control strategy is applied to a dynamic mirror with antagonistic piezoelectric stack actuation. Hysteresis compensator is performed by a finite state machine switching polynomials for hysteresis inversion based on the input signal slope. Residual error after hysteresis compensation is corrected by an LQR feedback controller. Experimental results demonstrate effectiveness of the hysteresis compensator and closed-loop system under the proposed hysteresis control strategy. For the triangular input signal tested, the closed-loop system achieves a 91.5% reduction in hysteresis uncertainty with 60 kHz sample rate.

Commentary by Dr. Valentin Fuster
2013;():V003T37A003. doi:10.1115/DSCC2013-3945.

Mechanical indentation and plowing is one of the most widely used methods in probe-based nanolithography. Compared to other probe-based nanolithography techniques such as the Dip-pen and the milliped, mechanical plowing is not restrictive to conductive materials and/or soft materials. However, like other probe-based nanolithgraphy techniques, the low-throughput has hindered the implementation of this technique in practices. The fabrication throughput, although can be increased via parallel-probe, is ultimately limited by the tracking precision of the probe relative to the sample during the plowing process. In this paper, a new iterative learning control technique is proposed and utilized to account for the adverse effects encountered in high-speed, large-range mechanical plowing nanolithography, including the hysteresis, the vibrational dynamics, and the cross-axis dynamics-coupling effects. Moreover, vertical (normal) ultrasonic vibration of the cantilever is introduced during the fabrication process to improve the fabrication quality. This approach is implemented to directly fabricate patterns on a mask with a tungsten layer deposited on a silicon dioxide substrate. The experimental results demonstrated that a relatively large-size pattern of four grooves (20 μm in length) can be fabricated at a high-speed of ∼5 mm/sec, with the line width and line depth at ∼95 nm and 2 nm, respectively. A fine pattern of the word ‘NANO’ is also achieved at the speed of ∼5 mm/sec. Such a high-speed direct lithography of mask with nanoscale line width and depth points the use of mechanical-plowing technique in strategic-important applications such as mask lithography for semiconductor industry.

Commentary by Dr. Valentin Fuster
2013;():V003T37A004. doi:10.1115/DSCC2013-3974.

A novel threshold sensing strategy for improving accuracy of a tracking controller used in calibration of an Inertial Measurement Unit (IMU) with a piezoelectric micro-actuator is presented in this paper. An asynchronous threshold sensor is hypothesized as a way to improve state estimates obtained from analog sensor measurements of micro-actuator motion. In order to produce accurate periodic signals using the proposed piezoelectric actuator and sensing arrangement, an Iterative Learning Control (ILC) is employed for control system design. Three sensing strategies: (i) an analog sensor alone with a Kalman filter, (ii) an analog sensor and threshold sensor with a Kalman filter and (iii) an analog sensor and threshold sensor with a Kalman smoother are compared in simulation. Results show that incorporating threshold sensors into the piezoelectric micro-actuation system should allow at least certain angular positions or rates to be known with much higher accuracy than from analog sensing alone, which can be useful for identifying calibration curves from the linear region of IMU operation.

Commentary by Dr. Valentin Fuster
2013;():V003T37A005. doi:10.1115/DSCC2013-3983.

Design and analysis for an efficient and force dense piezoelectric poly-actuated linear motor is presented. A linear motor is constructed with multiple piezoelectric actuator units engaging a rod having gear teeth. The multiple piezo-units are placed along the geared rod with a particular phase difference such that a near constant force is generated regardless of the rod position by coordinating the multiple piezo-units. Rolling contact buckling mechanisms within the piezo-units provide large displacement amplification with high energy transmission and low loss properties from the piezo-units to the geared rod. This piezo-based motor has capacitive actuator characteristics which allow it to bear static loads efficiently. Furthermore, the poly-actuator architecture presented provides for scalability through modular design.

First, the basic design principle describing the engagement of buckling amplification mechanisms to a phased array-shaped gear rod is presented, and the resulting force and displacement characteristics are analyzed. Design methods for creating a piezoelectric poly-actuated linear motor are then summarized. A prototype design is presented for which a maximum mean force of 213 N, a maximum velocity of 1.125 m/s, and a force density of 41 N/kg is calculated.

Topics: Linear motors , Design
Commentary by Dr. Valentin Fuster
2013;():V003T37A006. doi:10.1115/DSCC2013-4019.

Multi-axis (z, θx, θy) micro-actuators based on thin-film lead-zirconate-titanate (PZT) for use in dual axes confocal microscopy are presented with their static and dynamic models. Prototype actuators have achieved as much as 430 μm of vertical displacement and ±10° of mechanical tilting angles in both θx and θy directions in a footprint of 3.2×3.2 mm. The experimental static displacements and transient response of the actuator were used to identify residual stresses in the thin films, dimensional variance due to fabrication limitation, and damping coefficients in the model. With the identified parameters, the model predicts the static displacements of the four corners of the stage with an average absolute error of 17.4 μm over five different voltage levels and shows a reasonable agreement with the experimentally measured transient dynamic data. These results will be used to develop closed-loop controller for the system.

Commentary by Dr. Valentin Fuster

Robotic Manipulators

2013;():V003T38A001. doi:10.1115/DSCC2013-3740.

The solution of the inverse kinetics problem of soft manipulators is essential to generate paths in the task space to perform grasping operations. To address this issue, researchers have proposed different iterative methods based on Jacobian matrix. However, although these methods guarantee a good degree of accuracy, they suffer from singularities, long-term convergence, parametric uncertainties and high computational cost. To overcome intrinsic problems of iterative algorithms, we propose here a neural network learning of the inverse kinetics of a soft manipulator. To our best knowledge, this represents the first attempt in this direction. A preliminary work on the feasibility of the neural network solution has been proposed for a conical shape manipulator driven by cables. After the training, a feed-forward neural network (FNN) is able to represent the relation between the manipulator tip position and the forces applied to the cables. The results show that a desired tip position can be achieved quickly with a degree of accuracy of 0.73% relative average error with respect to total length of arm.

Commentary by Dr. Valentin Fuster
2013;():V003T38A002. doi:10.1115/DSCC2013-3805.

Some manipulation tasks have directions of end-effector acceleration of a manipulator and directions for which dynamic accuracy is required in the motion. This paper proposes an index (DAIT: Dynamic Accuracy Index for Task-directions) that allows us to evaluate the dynamic accuracy of manipulators considering the task-directions. Firstly, we derive the DAIT. Secondly, we evaluate some postures of a 2-degrees of freedom (DOF) planar manipulator on the basis of some indices that have been proposed (condition number, dynamic manipulability measure and task compatibility) and the DAIT, respectively. Thirdly, we show a manipulator’s design example based on the DAIT. Finally, from the evaluation and design results, we discuss the usefulness of the DAIT in determining the suitable postures of the manipulator for a given task and in designing the suitable manipulators for a given task.

Topics: Design , Manipulators
Commentary by Dr. Valentin Fuster
2013;():V003T38A003. doi:10.1115/DSCC2013-3818.

A comparison of serial, parallel, and dual Passive Assist Devices (PADs) designed using energy minimization based on a known maneuver is presented. Implementation of a PAD can result in an improvement in system performance with respect to efficiency, reliability, and/or safety. In previous work we demonstrated this concept experimentally on a single link robot arm augmented with a torsional spring in parallel. Here we show that the concept can extended to serial and dual systems as well. To make the optimization converge more quickly we introduce a new initial design using a weighted force displacement curve fit. We provide engineering insight into why different types of PADs perform differently depending on the maneuver and offer guidelines on how to select a specific type based on the application. Finally, we demonstrate this design process and selection procedure on a 3-link manipulator arm and show that a combination of parallel and dual PADs could reduce energy consumption by up to 78%.

Topics: Design
Commentary by Dr. Valentin Fuster
2013;():V003T38A004. doi:10.1115/DSCC2013-4029.

In this paper, we consider the dimensional synthesis of one degree-of-freedom multi-loop planar linkages such that they do not violate normal direction and second order curvature constraints imposed by contact with objects. Our goal is in developing minimally actuated multi-loop mechanical devices for human-robot interaction, that is, devices whose tasks will happen in a human environment.

Currently no systematic method exists for the kinematic synthesis of robotic fingers that incorporate multi-loop kinematic structure with second order task constraints, related to curvature. We show how to use these contact and curvature effects to formulate the synthesis equations for the design of a planar one-degree-of-freedom six-bar linkage. An example for the design of a finger that maintains a specified contact with an object, for an anthropomorphic task, is presented at the end of the paper.

It is important to note, that the theoretical foundation presented in this paper, assists in solving some of the open problems of this field, providing preliminary results on the synthesis of kinematic chains with multi-loop topology and the use of novel task specifications that incorporate curvature constraints with future applications in grasping and object manipulation.

Commentary by Dr. Valentin Fuster
2013;():V003T38A005. doi:10.1115/DSCC2013-4052.

Parallel manipulators are well known for their superior stiffness, higher accuracy, lower inertia and faster response compared to the serial counterparts and hence is widely used for high-speed machining and heavy load applications. However, controller limitations as well as design constraints can result in un-optimized designs causing unsettling residual vibrations at the end effector and limit their applications. Though many have discussed improving the structural design, augmenting with redundant sensors/ dampers and advanced feedback control methods for serial and mobile manipulators for vibration attenuations, very few investigated such techniques for parallel manipulators (PM). In this manuscript, we evaluate a specific type of feed-forward technique for planar PM. Lagrangian based dynamic models of platform manipulators and a simple trajectory level proportional-derivative control will be used with the gains tuned to force oscillations at the end effector to ensure stability. We will then demonstrate the applicability of basic input shapers for PM based on computation of natural frequencies and damping ratio for each mode, and resulting improvements in terms of appropriate performance measures.

Topics: Design , Manipulators
Commentary by Dr. Valentin Fuster
2013;():V003T38A006. doi:10.1115/DSCC2013-4094.

In this paper, we present an optimal design and control algorithm for multi-input binary-segmented Shape Memory Alloy (SMA) actuator arrays applied to a multi-degree-of-freedom (DOF) robot manipulator as it tracks a desired trajectory. The multi-DOF manipulator used for this paper is a 3-DOF-robot finger. A multi-input binary-segmented SMA actuator drives each DOF. SMA wires are embedded into a compliant vessel, such that both electric and fluidic (hot/cold) input can be applied to the actuators. By segmenting the SMA actuators, each segment can be controlled in a binary fashion (fully contracted/extended) to create a set of discrete displacements for each joint of the manipulator. To design the number of segments and length of each segment, an algorithm is developed to optimize the workspace. To optimize the workspace, it is desired to have a uniform distribution of reachable points in Cartesian space. Moreover, the number of neighbors (points that can be reached just by one control command from the current configuration) and the computational cost are important in workspace optimization. Graph theory search techniques based on the A* algorithm are employed to develop the control algorithm. A path-cost function is proposed to optimize the cost, which is a combination of actuation time, energy usage, and kinematic error. The kinematic error is estimated as the deviation between the actual and desired trajectory. The performance of the search algorithm and cost function are validated through simulation.

Commentary by Dr. Valentin Fuster

Robotics and Manipulators

2013;():V003T39A001. doi:10.1115/DSCC2013-3737.

Safety is one of important factors in control of mechatronic systems interacting with humans. In order to evaluate the safety of such systems, mechanical impedance is often utilized as it indicates the magnitude of reaction forces when the systems are subjected to motions. Namely, the mechatronic systems should have low mechanical impedance for improved safety. In this paper, a methodology to design controllers for reduction of mechanical impedance is proposed. For the proposed controller design, the mathematical definition of the mechanical impedance for open-loop and closed-loop systems is introduced. Then the controllers are designed for stable and unstable systems such that they effectively lower the magnitude of mechanical impedance with guaranteed stability. The proposed method is verified through case studies including simulations.

Commentary by Dr. Valentin Fuster
2013;():V003T39A002. doi:10.1115/DSCC2013-3767.

This paper presents an approach for fast modeling and identification of robot dynamics. By using a data-driven machine learning approach, the process is simplified considerably from the conventional analytical method. Regressor selection using the Lasso (l1-norm penalized least squares regression) is used. The method is explained with a simple example of a two-link direct-drive robot. Further demonstration is given by applying the method to a three-link belt-driven robot. Promising result has been demonstrated.

Commentary by Dr. Valentin Fuster
2013;():V003T39A003. doi:10.1115/DSCC2013-3836.

For robots with joint elasticity, discrepancies exist between the motor side information and the load side (i.e., end-effector) information. Therefore, high tracking performance at the load side can hardly be achieved when the estimate of load side information is inaccurate. To minimize such inaccuracies, it is desired to calibrate the load side sensor (in particular, the exact sensor location). In practice, the optimal placement of the load side sensor often varies due to the task variation necessitating frequent sensor calibrations. This frequent calibration need requires significant effort and hence is not preferable for industries which have relatively short product cycles. To solve this problem, this paper presents a sensor frame identification algorithm to automate this calibration process for the load side sensor, in particular the accelerometer. We formulate the calibration problem as a nonlinear estimation problem with unknown parameters. The Expectation-Maximization algorithm is utilized to decouple the state estimation and the parameter estimation into two separated optimization problems. An overall dual-phase learning structure associated with the proposed approach is also studied. Experiments are designed to validate the effectiveness of the proposed algorithm.

Topics: Elasticity , Sensors , Robots
Commentary by Dr. Valentin Fuster
2013;():V003T39A004. doi:10.1115/DSCC2013-3896.

This paper presents the dynamics modeling and dynamic identification of a dual-blade wafer handling robot. An explicit form dynamic model for this 8-link parallel robot is proposed. The dynamic model is transformed into a decoupled form to enable dynamic parameters identification with least-square regression. A well conditioned trajectory is chosen for identification experiment. Both viscous friction and Coulomb friction are considered to make the model more reliable. Model has been validated by experiments.

Commentary by Dr. Valentin Fuster
2013;():V003T39A005. doi:10.1115/DSCC2013-3950.

Traditional kinematic analysis of manipulators, built upon a deterministic articulated kinematic modeling often proves inadequate to capture uncertainties affecting the performance of the real robotic systems. While a probabilistic framework is necessary to characterize the system response variability, the random variable/vector based approaches are unable to effectively and efficiently characterize the system response uncertainties. Hence in this paper, we propose a random matrix formulation for the Jacobian matrix of a robotic system. It facilitates characterization of the uncertainty model using limited system information in addition to taking into account the structural inter-dependencies and kinematic complexity of the manipulator. The random Jacobian matrix is modeled such that it adopts a symmetric positive definite random perturbation matrix. The maximum entropy principle permits characterization of this perturbation matrix in the form of a Wishart distribution with specific parameters. Comparing to the random variable/vector based schemes, the benefits now include: incorporating the kinematic configuration and complexity in the probabilistic formulation, achieving the uncertainty model using limited system information (mean and dispersion parameter), and realizing a faster simulation process. A case study of a 6R serial manipulator (PUMA 560) is presented to highlight the critical aspects of the process. A Monte Carlo analysis is performed to capture the deviations of distal path from the desired trajectory and the statistical analysis on the realizations of the end effector position and orientation shows how the uncertainty propagates throughout the system.

Topics: Manipulators
Commentary by Dr. Valentin Fuster
2013;():V003T39A006. doi:10.1115/DSCC2013-4108.

This paper presents the formulation of a reduced-order linear discrete–path approximation in state space and its solution as a function of path lengths for a 3D curvature-based beam model (CBM). Solutions to both forward and inverse problems are discussed; the former is essential for real-time deformed shape visualization whereas the latter is much needed for haptic force feedback. The method is illustrated with an application example where a 2D beam is characterized by a 6th order CBM. Practical implementation shows that when external forces as system input are expressed in global coordinates, the CBM can be decoupled into two 2nd order systems enabling parallel computing of the deformed shape and the orientation and moment, and effectively reducing the table size for storing the operating conditions. The proposed real-time computation method has been validated by verifying results against published experimental and MSM simulated data.

Commentary by Dr. Valentin Fuster


2013;():V003T40A001. doi:10.1115/DSCC2013-3772.

This paper proposes a contactless vibration testing system for rotating disks based on an impulse response excited by a laser ablation. High power YAG pulse laser is used in this system for producing an ideal impulse force on structural surface without contact. The contactless vibration testing system is composed of a YAG laser, laser Doppler vibrometer and spectrum analyzer. This system makes it possible to measure vibration characteristics of structures under operation, such as vibration measurement of a rotating disk. The effectiveness of this system is confirmed by experimental and theoretical analyses. In this paper, a platter of hard disk drive is employed as an experimental object. Vibration characteristics of a rotating and non-rotating platter are measured and compared with the results of theoretical analysis.

Commentary by Dr. Valentin Fuster
2013;():V003T40A002. doi:10.1115/DSCC2013-3822.

This paper presents a new rotary position-control system using a color sensor. The angle sensing mechanism is based on the working principle of a red-green-blue (RGB) sensor that measures the radiant-intensity variation of the light reflected on the color surface. The optical-power propagation mechanism from a light-emitting diode (LED) source to the RGB sensor’s voltage is investigated using the light reflected on the designated-RGB codes of the cylindrical color-track surface. The nonlinearity due to a color printer and a paper roughness is compensated for through iterative comparisons with the reference angle achieved from a precision potentiometer. The performance of this new absolute angle sensor is validated using a rotary mechanical system with a cylindrical inertia controlled by a lead compensator. The stability of the positioning system is also investigated with the frequency response. Eventually, the feasibility of this new rotary angle sensor with the cost-effective and non-contact sensing mechanism is demonstrated.

Commentary by Dr. Valentin Fuster
2013;():V003T40A003. doi:10.1115/DSCC2013-3833.

This paper presents a control scheme of visual servoing. Real-time vision guidance is necessary in many desirable applications of industrial manipulators. Challenge comes from the limitations of visual sensing and robot dynamics. Typical industrial machine vision systems have low sampling rate and large latency. In addition, due to the large inertia of industrial manipulators, a proper consideration of robot dynamics is important. In particular, actuator saturation may cause undesirable response. In this paper, an adaptive tracking filter is used for sensing compensation. Based on the compensated vision feedback, a two-layer controller is formulated using multi-surface sliding control. System kinematics and dynamics are decoupled and handled by the two layers of the controller respectively. Further, a constrained optimal control approach is adopted to avoid actuator saturation. Validation is conducted using a SCARA robot.

Commentary by Dr. Valentin Fuster
2013;():V003T40A004. doi:10.1115/DSCC2013-3852.

A human motion capture system is becoming one of the most useful tools in rehabilitation application because it can record and reconstruct a patient’s motion accurately for motion analysis. In this paper, a human motion capture system is proposed based on inertial sensing. A microprocessor is implemented on-board to obtain raw sensing data from the inertial measurement unit (IMU), and transmit the raw data to the central processing unit. To reject noise in the accelerometer, drift in the gyroscope, and magnetic distortion in the magnetometer, a time-varying complementary filter (TVCF) is implemented in the central processing unit to provide accurate attitude estimation. A forward kinematic model of the human arm is developed to create an animation for patients and physical therapists. Performance of the hardware and filtering algorithm is verified by experimental results.

Topics: Filters
Commentary by Dr. Valentin Fuster
2013;():V003T40A005. doi:10.1115/DSCC2013-3873.

We present a framework for robust estimation of the configuration of an articulated robot using a large number of redundant proprioceptive sensors (encoders, gyros, accelerometers) distributed throughout the robot. Our method uses an Unscented Kalman Filter (UKF) to fuse the robot’s sensor measurements. The filter estimates the angle of each joint of the robot, enabling the accurate estimation of the robot’s kinematics even if not all modules report sensor readings. Additionally, a novel outlier detection method allows the the filter to be robust to corrupted accelerometer and gyro data.

Commentary by Dr. Valentin Fuster
2013;():V003T40A006. doi:10.1115/DSCC2013-3913.

The goal of this paper is to perform a parametric study on a newly developed visual odometry algorithm for use with color-depth (RGB-D) camera pairs, such as the Microsoft Kinect. In this algorithm, features are detected in the color image and converted to 3D points using the depth image. These features are then described by their 3D location and matched across subsequent frames based on spatial proximity. The visual odometry is then calculated using a one-point inverse kinematic solution. The primary contribution of this work is the identification of critical operating parameters associated with the algorithm, the analysis of their effects on the visual odometry performance, and the verification of the analysis using experimentation.

Commentary by Dr. Valentin Fuster

System Identification and Estimation for Automotive Applications

2013;():V003T41A001. doi:10.1115/DSCC2013-3775.

In the design of vehicle stability control (VSC) systems for ground vehicles, sideslip angle plays a vital role and its estimation has long been an active research topic. Accurate estimation of sideslip angle is more difficult for lightweight vehicles (LWVs) because their parameters are prone to significant changes with loading conditions — the amount and position of the payload. In this paper, a robust sideslip angle estimator based on a recently emerging smooth variable structure filter (SVSF) is presented. This sideslip angle estimator is suitable for LWVs because it is almost non-sensitive to the changes of the system parameters. A four-state vehicle lateral dynamic model including a pseudo-Burckhardt tire model is employed in the filter design. Compared with the widely utilized extended Kalman filter (EKF), the SVSF shows much better robustness against modeling errors. It is also more favorable in terms of tuning effort and computational speed. Simulation studies were conducted based on a high-fidelity vehicle model in CarSim®, where the vehicle took the form of a lightweight electric ground vehicle with independent in-wheel motors. The performance of the SVSF was shown by comparisons against the EKF under different settings for model parameters.

Topics: Vehicles , Filters
Commentary by Dr. Valentin Fuster
2013;():V003T41A002. doi:10.1115/DSCC2013-3776.

Battery state estimation (BSE) is one of the most important design aspects of an electrified propulsion system. It includes important functions such as state-of-charge estimation which is essentially for the energy management system. A successful and practical approach to battery state estimation is via real time battery model parameter identification. In this approach, a low-order control-oriented model is used to approximate the battery dynamics. Then a recursive least squares is used to identify the model parameters in real time. Despite its good properties, this approach can fail to identify the optimal model parameters if the underlying system contains time constants that are very far apart in terms of time-scale. Unfortunately this is the case for typical lithium-ion batteries especially at lower temperatures. In this paper, a modified battery model parameter identification method is proposed where the slower and faster battery dynamics are identified separately. The battery impedance information is used to guide how to separate the slower and faster dynamics, though not used specifically in the identification algorithm. This modified algorithm is still based on least squares and can be implemented in real time using recursive least squares. Laboratory data is used to demonstrate the validity of this method.

Topics: Batteries
Commentary by Dr. Valentin Fuster
2013;():V003T41A003. doi:10.1115/DSCC2013-3777.

A new estimation method for estimating the vehicle sideslip angle, mainly based on a linear parameter varying (LPV) model with independently estimated tire friction forces, is proposed for electric ground vehicles (EGVs) with four independent in-wheel motors. By utilizing the individual wheel dynamics, the longitudinal ground friction force is estimated from a PID observer based on a descriptor linear system approach. Moreover, the lateral ground friction force for each wheel is estimated through the friction ellipse relationship given the estimated longitudinal friction force, without relying on explicit tire models. Since the estimation errors of friction forces may bring parameter uncertainty for the LPV system, robust analysis with desired H-infinity performance is given for the observer design of the LPV modeling. This method is specially proposed for large tire slip angles and lateral friction forces. Simulation results for different maneuvers validate this novel sideslip angle estimation method.

Commentary by Dr. Valentin Fuster
2013;():V003T41A004. doi:10.1115/DSCC2013-3831.

In order to optimize the use of fault tolerant controllers for unmanned or autonomous aerial vehicles, a health diagnostics system is being developed. To autonomously determine the effect of damage on global vehicle health, a feature-based neural-symbolic network is utilized to infer vehicle health using historical data. Our current system is able to accurately characterize the extent of vehicle damage with 99.2% accuracy when tested on prior incident data. Based on the results of this work, neural-symbolic networks appear to be a useful tool for diagnosis of global vehicle health based on features of subsystem diagnostic information.

Topics: Vehicles
Commentary by Dr. Valentin Fuster
2013;():V003T41A005. doi:10.1115/DSCC2013-3984.

This work presents two advances to the estimation of homogeneous charge compression ignition (HCCI) dynamics. Combustion phasing prediction in control-oriented models has been achieved by modeling the in-cylinder temperature and composition dynamics, which are dictated by the large mass of residuals trapped between cycles. As such, an accurate prediction of the residual gas fraction as a function of the variable valve timing is desired. Energy and mass conservation laws applied during the exhaust valve opening period are complemented with online in-cylinder pressure measurements to predict the trapped residual mass in real time. In addition, an adaptive parameter estimation scheme uses measured combustion phasing to adjust the residual mass prediction. Experimental results on a multicylinder gasoline HCCI engine demonstrate the closed loop residual estimation’s ability to compensate for modeling errors, cylinder to cylinder variations, and engine wear. Additionally it is shown that using the adaptive parameter estimation reduces the model parameterization effort for a multicylinder engine.

Commentary by Dr. Valentin Fuster
2013;():V003T41A006. doi:10.1115/DSCC2013-4005.

The in-cylinder temperature information is critical in the field of auto-ignition control in advanced combustion modes. However, the in-cylinder temperature is hard to be directly measured at low cost in production engines. In this paper, a cycle-by-cycle estimation method is proposed for the in-cylinder temperature at the crank angle of intake valve closing (IVC), referred to as Tivc. Through investigating the thermodynamics of Tivc, an Extended Kalman Filter (EKF) based method was devised by utilizing the measurable temperature information from the intake and exhaust manifolds. The proposed method was validated through high-fidelity GT-Power engine model simulation.

Commentary by Dr. Valentin Fuster

System Identification and Modeling

2013;():V003T42A001. doi:10.1115/DSCC2013-3798.

Stringent emission regulations mandated by California air regulation board (CARB) require monitoring the upstream exhaust gas oxygen (UEGO) sensor for any possible malfunction causing the vehicle emissions to exceed certain thresholds. Six faults have been identified to potentially cause the UEGO sensor performance to deteriorate and potentially lead to instability of the air-fuel ratio (AFR) control loop. These malfunctions are either due to an additional delay or an additional lag in the transition of the sensor response from lean to rich or rich to lean. Current technology detects the faults the same way (approximated by a delay type fault) and does not distinguish between the different faults. In the current paper, a statistics based approach is developed to diagnose these faults. Specifically, the characteristics of a non-normal distribution function are estimated based on the UEGO sensor output and used to detect and isolate the faults. When symmetric operation is detected, a system identification process is employed to estimate the parameters of the dynamic system and determine the type of operation. The proposed algorithm has been demonstrated on real data obtained from both Ford F150 and Mustang V6 vehicles.

Topics: Sensors
Commentary by Dr. Valentin Fuster
2013;():V003T42A002. doi:10.1115/DSCC2013-3803.

Ionic polymer metal composite (IPMC), categorized as an ionic electroactive polymer (EAP), can exhibit conspicuous deflection with low external voltages (∼5 V). This material has been commonly applied in robotic artificial muscles since reported in 1992 because it can be fabricated in various sizes and shapes. Researchers developed numerous IPMC models according to its deflection in response to the corresponding input stimulation. In this paper, an IPMC strip is modeled (1) as a cantilever beam with a loading distribution on the surface, and (2) with system identification tools, such as an autoregressive with exogenous (ARX)/autoregressive moving average with exogenous (ARMAX) model and an output-error (OE) model. Nevertheless, the loading distribution is non-uniform due to the imperfect surface conductivity. Finally, a novel linear time-variant (LTV) modeling method is introduced and applied to an IPMC electrical model on the basis of the internal environment such as surface resistance, thickness, and water distribution related to the unique working principle of IPMC. A comparison between the simulated and the experimental deflections demonstrates the benefits and accuracy of the LTV electrical model.

Commentary by Dr. Valentin Fuster
2013;():V003T42A003. doi:10.1115/DSCC2013-3808.

Artificial cilia systems are used for microfluidic manipulation. By analogy to the biological cilia, such systems seek to mix, separate, or propel fluids, particularly in the low-Reynolds-number regime, without damaging sensitive samples. An important category of artificial cilia systems is magnetically-actuated artificial cilia, since the driving magnetic field does not interact with many samples of interest. Simulation results are presented to show that linear modeling fails to adequately predict the optimal location due to strong nonlinear effects; using the linear result to select magnet placement results in amplitudes 84% lower than the amplitude with the optimal placement found using the nonlinear model. This represents a substantial loss in efficacy. Since large amplitudes are desirable to enhance flow manipulation, the results illustrate the importance of nonlinear dynamics models in the design of magnet-cilia devices.

Commentary by Dr. Valentin Fuster
2013;():V003T42A004. doi:10.1115/DSCC2013-3953.

Health management of Li-ion batteries requires knowledge of certain battery internal dynamics (e.g., lithium consumption and film growth at the solid-electrolyte interface) whose inputs and outputs are not directly measurable with noninvasive methods. Therefore, identification of those dynamics can be classified as an inaccessible subsystem identification problem. To address this problem, the retrospective-cost subsystem identification (RCSI) method is adopted in this paper. Specifically, a simulation-based study is presented that represents the battery using an electrochemistry-based battery charge/discharge model of Doyle, Fuller, and Newman augmented with a battery-health model by Ramadass. The solid electrolyte interface (SEI) film growth portion of the battery-health model is defined as the inaccessible subsystem to be identified using RCSI. First, it is verified that RCSI with a first-order subsystem structure can accurately estimate the film growth when noise or modeling errors are ignored. Parameter convergence issues are highlighted. Second, allowable input and output noise levels for desirable film growth tracking performance are determined by studying the relationship between voltage change and film growth in the truth model. The performance of RCSI with measurement noise is illustrated. The results show that RCSI can identify the film growth within 1.5% when the output measurement noise level is comparable to the change in output voltage between successive cycles due to film growth, or when the input measurement noise is comparable to the difference in current that results in a difference in voltage that is the same as the voltage change between successive cycles. Finally, the sensitivity of the performance of RSCI to initial condition errors in the battery charge/discharge model is investigated. The results show that when the initial conditions have an error of 1%, the identified results change by 7%. These results will help with selecting the appropriate sensors for the experiments with the hardware.

Commentary by Dr. Valentin Fuster
2013;():V003T42A005. doi:10.1115/DSCC2013-4007.

In many engineering applications, diagnostic techniques are needed to characterize the mechanical properties of internal components that are not readily visible at the surface of an object, as in the use of nondestructive testing to detect sub-surface damage in composite materials. Understanding the role of structural interfaces between two bodies is a key factor in developing these diagnostic techniques because the mechanical and geometric properties of the interface determine the degree to which measurements on the surface can be used to interrogate sub-surface components. In this paper, vibration measurements on a polycarbonate material, henceforth referred to as the buffer, are used to characterize the mechanical properties of a polymer particulate composite, henceforth referred to as the target, which is located beneath the buffer. To this end, a three-dimensional laser Doppler vibrometer and piezoelectric inertial actuator are used to measure the broadband response of the two-body structural dynamic system. Because of the importance of the actuator dynamics to the diagnostic measurements, a descriptive model is developed to better understand these dynamics and interpret the results. The longitudinal dynamics of the two-body system are shown to involve stronger coupling between the target and buffer materials as compared to the transverse dynamics. A Complex Mode Indicator Function is then used to extract the modal deflection shapes, and it is shown that the interface between the bodies introduces complexity in the dynamic response. Changes in the surface velocity of the buffer material are also studied as a function of a key mechanical property — the volume fraction of crystals in the target composite material. It is demonstrated that both the linear and nonlinear vibration characteristics of the buffer material change as a function of the composition of the target material, suggesting that a compositional diagnostic procedure is possible using surface vibration measurements.

Commentary by Dr. Valentin Fuster
2013;():V003T42A006. doi:10.1115/DSCC2013-4081.

Arising technologies in vehicle-to-vehicle (V2V) communication allow vehicles to obtain information about the motion of distant vehicles. Such information can be presented to the driver or incorporated in advanced autonomous cruise control (ACC) systems. In this paper, we investigate the effects of multi-vehicle communication on the dynamics of connected vehicle platoons and propose a motif-based approach that allows systematical analysis and design of such systems. We investigate the dynamics of simple motifs in the presence of communication delays, and show that long-distance communication can stabilize the uniform flow when the flow cannot be stabilized by nearest neighbor interactions. The results can be used for designing driver assist systems and communication-based cruise control systems.

Commentary by Dr. Valentin Fuster

System Identification and Therapeutic Control in Bio-Systems

2013;():V003T43A001. doi:10.1115/DSCC2013-3768.

This work aims to predict in-hospital mortality in the open-source Physionet ICU database from features extracted from the time series of physiological variables using neural network models and other machine learning techniques. We developed an effective and efficient greedy algorithm for feature selection, reducing the number of potential features from 205 to a best subset of only 47. The average of five trials of 10-fold cross validation shows an accuracy of (86.23±0.14)%, a sensitivity of (50.29±0.22)%, a specificity of (92.01 ± 0.21)%, a positive prediction value of (50.29±0.50)%, a negative prediction value of (92.01±0.00)%, and a Lemeshow score of 119.55±9.87. By calibrating the predicted mortality probability using an optimization approach, we can improve the Lemeshow score to 27.51±4.38. The developed model has the potential for application in ICU machines to improve the quality of care and to evaluate the effect of treatment or drugs.

Commentary by Dr. Valentin Fuster
2013;():V003T43A002. doi:10.1115/DSCC2013-3806.

A hybrid modeling framework integrating a highly specific mechanistic model with highly abstract empirical model is presented. With the growing interest in the scientific and medical community for identification of therapeutic targets in treatment of disease, it is necessary to develop predictive models that can describe cellular behavior in response to environmental cues. Intracellular signaling pathways form complex networks that regulate cellular response in both health and disease. Mechanistic (or white-box) models of biochemical networks are often unable to explain comprehensive cellular response due to lack of knowledge and/or intractable complexity (especially in events distal from the cell membrane). Empirical (or black-box) models may provide a less than accurate representation of cellular response due to data deficiency and/or loss of mechanistic detail. In the proposed framework, we use a mechanistic model to capture early signaling events and apply the resulting generated internal signals (along with external inputs) to a downstream empirical sub-model. The key construct in the approach is the treatment of a cell’s biochemical network as an encoder that creates a functional internal representation of external environmental cues. The signals derived from this representation are then used to inform downstream behaviors. Using this idea, we are able to create a comprehensive framework that describes important mechanisms with sufficient detail, while representing complex or unknown mechanisms in a more abstract form. The model is verified using published biological data describing T-Cells in immune response.

Commentary by Dr. Valentin Fuster
2013;():V003T43A003. doi:10.1115/DSCC2013-3848.

This paper seeks to determine the validity of two distinct tube-load models relating central aortic blood pressure to peripheral blood pressure in humans. Specifically a single-tube model (1-TL) and a serially connected two-tube (2-TL) model, both terminating in a Windkessel load, are considered as representations of the central aortic-peripheral arterial path. The validity and fidelity of the two models was assessed and compared quantitatively by fitting central aortic, radial and femoral blood pressures collected from 8 patients. Both models fitted the BP waveform pairs effectively, and were capable of estimating pulse travel time (PTT) accurately; also the model derived frequency responses were close to the empiric transfer function estimates derived from central and peripheral BP measurements. The 2-TL model was consistently better than 1-TL with statistical significance in terms of accuracy of the central aortic BP waveform, the average waveform RMSE were 2.52 mmHg versus 3.24 mmHg respectively (p<0.05).

Commentary by Dr. Valentin Fuster
2013;():V003T43A004. doi:10.1115/DSCC2013-3850.

In this paper, we present an innovative active non-intrusive system identification approach to cardiovascular monitoring. The proposed approach is based on a dual collocated actuator-sensor system for cardiovascular system identification, in which the actuators actively excite the arterial tree to create rich and informative trans-mural pressure waves traveling in the arterial tree, which are then non-intrusively measured by the collocated sensors. In our previous work, we developed a mathematical model to reproduce the propagation of intra-vascular (arterial) and extra-vascular (artificial) pressure waves along the arterial tree. Then, we used a dual (radial-femoral) blood pressure cuff system as a prototype dual collocated actuator-sensor system to demonstrate the proposed methodological framework to create rich trans-mural pressure waves as well as to non-intrusively reconstruct them from sensor measurements. In this follow-up work, we propose a novel system identification algorithm to derive cardiovascular system dynamics and reconstruct central aortic blood pressure waveform from the trans-mural pressure waves observed at the peripheral locations. It was successfully demonstrated that the system identification algorithm was able to reconstruct the central aortic blood pressure accurately, and that its performance was superior to the passive non-intrusive approach.

Commentary by Dr. Valentin Fuster
2013;():V003T43A005. doi:10.1115/DSCC2013-3879.

Cardiopulmonary resuscitation (CPR) is a commonly used procedure and plays a critical role in saving the lives of patients suffering from cardiac arrest. This paper is concerned with the design of a dynamic technique to optimize the performance of CPR and to consequently improve its outcome, the survival rate. Current American Heart Association (AHA) guidelines treat CPR as a static procedure with fixed parameters. These guidelines set fixed values for CPR parameters such as compression to ventilation ratio, chest compression depth, etc., with an implicit assumption that they are somehow “optimal,” which has not been really substantiated. In this study, in a quest to improve this oft-used procedure, an interactive technique has been developed for dynamically changing the CPR parameters. Total blood gas delivery which is combination of systemic oxygen delivery and carbon dioxide delivery to the lungs has been defined as the objective function, and a sequential optimization procedure has been explored to optimize the objective function by dynamically adjusting the CPR parameters. The results of comparison between the sequential optimization procedure and the global optimization procedure show that the sequential optimization procedure could significantly enhance the effectiveness of CPR.

Commentary by Dr. Valentin Fuster
2013;():V003T43A006. doi:10.1115/DSCC2013-4060.

This paper reports preliminary results on the effects of ankle muscle fatigue on ankle mechanical impedance. The experiment was designed to induce fatigue in the Tibialis Anterior and Triceps Surae muscle group by asking subjects to perform isometric contractions against a constant ankle torque generated by the Anklebot, a backdriveable robot that interacts with the ankle in two degrees of freedom. Median frequencies of surface electromyographic signals collected from Tibialis Anterior and Triceps Surae muscle group were evaluated to assess muscle fatigue. Using a standard multi-input and multi-output stochastic impedance identification method, multivariable ankle mechanical impedance was measured in two degrees of freedom under muscle fatigue. Preliminary results indicate that, for both Tibialis Anterior and Triceps Surae muscle group, ankle mechanical impedance decreases in both the dorsi-plantarflexion and inversion-eversion directions under muscle fatigue. This finding suggests that decreasing ankle impedance with muscle fatigue may help to develop joint support systems to prevent ankle injuries caused by muscle fatigue.

Commentary by Dr. Valentin Fuster

Variable Structure/Sliding-Mode Control

2013;():V003T44A001. doi:10.1115/DSCC2013-3859.

Control of an inverted pendulum using horizontal force (e.g. cart-pole) is a well-established problem. However, there is no work in the literature that addresses controlling an inverted pendulum using only vertical force without horizontal movement. In this paper, control algorithm is presented that can control the inverted pendulum via only vertical force within the operating range of the given system. In the process, the dynamic stabilizing effect of vertical harmonic oscillation acting on inverted pendulum is exploited and previous works on the analysis of the stabilizing effect is utilized. The result is a nonlinear controller that combines sliding mode and energy shaping control which is capable of stabilizing the pendulum upright as well as at any titled angle within the region of attraction described. This is done for pendulum starting at any initial angle while maintaining vertical movement of the inverted pendulum within the prescribed space limit.

Commentary by Dr. Valentin Fuster
2013;():V003T44A002. doi:10.1115/DSCC2013-3872.

This paper introduces an accurate position control algorithm based on Backward-Euler discretization of a second-order sliding mode control (SOSMC) and the super-twisting observer (STO). This position control algorithm does not produce numerical chattering, which has been known to be a major drawback of explicit implementation of SOSMC and STO. It is more accurate than the conventional PID control that is also free of chattering. In contrast to conventional Backward-Euler discretization schemes of SOSMC and STO, the presented discretization method does not require any special solvers for computation. The accuracy and implementation of this algorithm are illustrated by simulations.

Commentary by Dr. Valentin Fuster
2013;():V003T44A003. doi:10.1115/DSCC2013-3926.

This paper examines the performance of an optimal sliding mode Gaussian (OSG) controller for the regulation of a hydropower plant implemented in the modeling program SIMSEN. The controller is designed to regulate grid frequency and includes the dynamics of the wicket gate servo system, turbine, and grid. Simulation results for OSG control are compared to those of more traditional LQG and PI controllers. Simulation shows that OSG control provides superior performance for the nominal system and for the system with parametric uncertainties.

Commentary by Dr. Valentin Fuster
2013;():V003T44A004. doi:10.1115/DSCC2013-3987.

This paper presents design and implementation of a super twisting sliding mode control for superheat temperature and evaporating temperature of refrigerant fluid in an evaporator of HVAC (Heating-Ventilation and Air Conditioning)-Refrigeration system. Based on a nonlinear model of the evaporator two control approaches are presented. The first approach is based on a Multi-Input Multi Output (MIMO) system in which there are two control inputs; inlet mass flow and outlet mass flow rate, and the outputs are the length of two phase flow and evaporating temperature of refrigerant. The second approach considers the system as a Single input single output (SISO) one, and by using inlet mass flow, superheat temperature is controlled. In the first approach, by implementing a feedback linearization method the two control inputs are decoupled. By decoupling the effects of both inputs, the two state variables of system are controlled separately and effectively. By applying sliding mode control robustness against the disturbances and uncertainties is guaranteed. Super-twisting algorithm is applied as a remedy for chattering problem in classical sliding mode control and achieving finite time convergence. Controller and model of systems are simulated using MATLAB and Simulink. The results of simulations show the effectiveness of designed controller in presence of uncertainties.

Commentary by Dr. Valentin Fuster
2013;():V003T44A005. doi:10.1115/DSCC2013-4067.

A novel combination of model predictive control (MPC) and sliding mode control (SMC) is presented in this paper. The motivation is to inherit the ability to explicitly deal with state and input constraints from MPC, and the good robustness property from SMC. The design of the finite-time optimal control problem and the conditions for the persistent feasibility and the closed-loop stability are discussed. Simulation results are shown to demonstrate the nominal and robust performance of the proposed control algorithm.

Commentary by Dr. Valentin Fuster

Vehicles and Human Robotics

2013;():V003T45A001. doi:10.1115/DSCC2013-3834.

As robots are increasingly used in human-cluttered environments, the requirement of human-likeness in their movements becomes essential. Although robots perform a wide variety of demanding tasks around the world in factories, remote sites and dangerous environments, they are still lacking the ability to coordinate with humans in simple, every-day life bi-manual tasks, e.g. removing a jar lid. This paper focuses on the introduction of bio-inspired control schemes for robot arms that coordinate with human arms in bi-manual manipulation tasks. Using data captured from human subjects performing a variety of every-day bi-manual life tasks, we propose a bio-inspired controller for a robot arm, that is able to learn human inter- and intra-arm coordination during those tasks. We embed human arm coordination in low-dimension manifolds, and build potential fields that attract the robot to human-like configurations using the probability distributions of the recorded human data. The method is tested using a simulated robot arm that is identical in structure to the human arm. A preliminary evaluation of the approach is also carried out using an anthropomorphic robot arm in bi-manual manipulation task with a human subject.

Topics: Robots , Biomimetics
Commentary by Dr. Valentin Fuster
2013;():V003T45A002. doi:10.1115/DSCC2013-3851.

Best practices in product design require engineers to perform preliminary hazard analyses on the most promising conceptual designs, as well as a more rigorous hazard analysis when the details of the product are being finalized. When the product is a complex dynamic system that interacts directly with a human, the engineers must consider the wide range of possible motions and forces that the device could create. Such an analysis goes beyond a simple thought exercise and requires detailed knowledge about the system dynamics and operating environment. This paper presents such an analysis of an inverted-pendulum human transporter. The list of hazards is constructed by using fundamental knowledge of the dynamics and the mechanical design obtained through simulation and experimentation. However, the dynamics are so complex that the list is augmented with hazards that are revealed by searching through accident videos posted on the Internet. The severity of each hazard is estimated using an energy-based measurement of the hazard onset conditions. While this case study is interesting, it also provides a systematic approach to hazard analysis that can be applied to other complex and dangerous dynamic systems.

Commentary by Dr. Valentin Fuster
2013;():V003T45A003. doi:10.1115/DSCC2013-3920.

This paper presents an interactive multimedia framework for introducing students to vehicle electrification/hybridization. The framework familiarizes its target audience with: (i) the societal factors driving the development of hybrid and plug-in hybrid electric vehicles (HEVs/PHEVs); (ii) the differences between conventional vehicles, HEVs, and PHEVs; and (iii) the high-level performance constraints and tradeoffs inherent in hybrid vehicle design. The framework consists of two coupled components: (i) a set of educational videos on vehicle electrification; and (ii) a 3D videogame built around physics-based models of conventional and series hybrid ambulances. The paper presents both the above education framework and the specific principles from the pedagogy literature guiding its design.

Commentary by Dr. Valentin Fuster
2013;():V003T45A004. doi:10.1115/DSCC2013-3933.

An algorithm to estimate positions, orientations, linear velocities and angular rates of an Underwater Remotely Operated Vehicle (UROV), based on the Extended Kalman Filter (EKF), is presented. The complete UROV kinematic and dynamic models are combined to obtain the process equation, and measurements correspond to linear accelerations and angular rates provided by an Inertial Measurement Unit (IMU). The proposed algorithm is numerically validated and its results are compared with simulated UROV states. A discussion about the influence of the covariance matrices on the estimation error and overall filter performance is also included. As a conclusion, the proposed algorithm estimates properly the UROV linear velocities and angular rates from IMU measurements, and the noise in estimated states is reduced in about one order of magnitude.

Commentary by Dr. Valentin Fuster
2013;():V003T45A005. doi:10.1115/DSCC2013-4049.

For the purpose of developing robot-assisted human walking systems, human and robot walking dynamics are modeled using models of different complexity depending on simulation scenarios in different phases of robotic system development and selected walking parameters to be analyzed. This paper addresses the early modeling and simulation phase of the development of a novel mobile robot-assisted gait rehabilitation system to be used as a demonstrator for a cognitive robot control architecture currently under development. For simulation purposes dynamical models of walking human and powered orthosis are developed in multi-body simulation software (MSC Adams) using the LifeMod plug-in while the control algorithms are developed in Matlab. The paper introduces a novel ROS (Robot Operating System) based communication established between the real system software modules and the simulation environment. The performance evaluation was performed by running the simulation with motion data which were obtained using marker-based motion capture system and which were implemented as ROS node.

Topics: Simulation , Modeling
Commentary by Dr. Valentin Fuster
2013;():V003T45A006. doi:10.1115/DSCC2013-4056.

One of the aims of the Colombian Ministry of Defense in the field of science and technology is to develop and build in-house simulators for training. An important prerequisite in the development of these types of simulators is to have accurate knowledge about the forces that act on the particular type of ship being considered. In the pursuit of this objective, the Science & Technology Corporation for the Development of the Shipbuilding Industry in Colombia — COTECMAR has established a research program for the development of physics-based models to predict the generalized forces acting on maneuvering ships.

The following article proposes a mathematical model capable of providing the simulator with calculations for the hydrodynamic forces acting on three types of ships: displacement ships, submarines and planning hulls. Derived from slender-body theory (SBT), the mathematical model presented minimizes computational time and eliminates the need for experimental data, making it possible to use the calculation of hydrodynamics forces at the initial stages of design when the geometry of the ship is constantly revised and the effect of those changes in the dynamic performance of the ship needs to be assessed.

The article explains the mathematical model proposed and its modular nature, compares existing numerical and experimental data with results obtained from this study for the three case studies selected: displacement ships, submarines and planing hulls.

Topics: Simulation , Ships
Commentary by Dr. Valentin Fuster

Vehicle Dynamics and Control

2013;():V003T46A001. doi:10.1115/DSCC2013-3829.

A variable stiffness architecture is used in the suspension system to counteract the body roll moment, thereby enhancing the roll stability of the vehicle. The variation of stiffness concept uses the “reciprocal actuation” to effectively transfer energy between a vertical traditional strut and a horizontal oscillating control mass, thereby improving the energy dissipation of the overall suspension. The lateral dynamics of the system is developed using a bicycle model. The accompanying roll dynamics are also developed and validated using experimental data. The positions of the left and right control masses are optimally allocated to reduce the effective body roll and roll rate. Simulation results show that the resulting variable stiffness suspension system has more than 50% improvement in roll response over the traditional constant stiffness counterparts. The simulation scenarios examined is the fishhook maneuver.

Commentary by Dr. Valentin Fuster
2013;():V003T46A002. doi:10.1115/DSCC2013-3886.

Many agricultural tasks, such as harvesting, are labor intensive. With the interests in autonomous farming, a method to rapidly generate trajectories for agricultural robots satisfying different realistic constraints becomes necessary. A hierarchical cooperative planning method is studied in this paper for a group of agricultural robots with a low computational cost. Two parts are involved in the method: once a reconfiguration event is confirmed, all the possible formation configurations will be evaluated and ranked according to their feasibility and performance index; a local pursuit strategy based cooperative trajectory planning algorithm is designed to generate optimal cooperative trajectories for robots to achieve and maintain their desired formation. To help reduce the computation cost associated with the cooperative planning algorithm, early termination conditions are proposed. The capabilities of the proposed cooperative planning algorithm are demonstrated in a simple citrus harvesting problem.

Commentary by Dr. Valentin Fuster
2013;():V003T46A003. doi:10.1115/DSCC2013-3890.

In this paper we study the lateral motion control and torque allocation for four-wheel-independent-drive electric vehicles (4WID-EVs) with combined active front steering (AFS) and direct yaw moment control (DYC) through in-vehicle networks. It is well known that the in-vehicle networks and x-by-wire technologies have considerable advantages over the traditional point-to-point communications, and bring great strengths to 4WID-EVs. However, there are also bandwidth limitations which would lead to message time delays in network communication. We propose a method on effectively utilizing the limited bandwidth resources and attenuating the adverse influence of in-vehicle network-induced time delays, based on the idea of dynamic message priority assignment according to the vehicle states and control signals. Simulation results from a high-fidelity vehicle model in CarSim® show that the proposed vehicle lateral control and torque allocation algorithm can improve the 4WID-EV lateral motion control performance, and the proposed message priority dynamic assignment algorithm can significantly reduce the adverse influence of the in-vehicle network-induced time delays.

Commentary by Dr. Valentin Fuster
2013;():V003T46A004. doi:10.1115/DSCC2013-3929.

The designs of commercial Anti-Lock Braking Systems often rely on assumptions of a torsionally rigid tire-wheel system. However, variations in tire/wheel technologies have resulted in lower torsional stiffnesses that cannot be captured well using these rigid wheel assumptions. This paper presents an adaptive nonlinear controller based on a model that incorporates sidewall flexibility, and transient & hysteretic tread-ground friction effects. The sidewall stiffness and damping and as well as tread parameters are assumed to be unknown and subsequently estimated through a set of gradient-based adaptation laws. A virtual damper is introduced via a backstepping controller design to address difficulties associated with tires with low torsional damping.

Topics: Traction , Tires , Wheels
Commentary by Dr. Valentin Fuster
2013;():V003T46A005. doi:10.1115/DSCC2013-4021.

Unmanned ground vehicles (UGVs) are gaining importance and finding increased utility in both military and commercial applications. Although earlier UGV platforms were typically exclusively small ground robots, recent efforts started targeting passenger vehicle and larger size platforms. Due to their size and speed, these platforms have significantly different dynamics than small robots, and therefore the existing hazard avoidance algorithms, which were developed for small robots, may not deliver the desired performance. The goal of this paper is to present the first steps towards a model predictive control (MPC) based hazard avoidance algorithm for large UGVs that accounts for the vehicle dynamics through high fidelity models and uses only local information about the environment as provided by the onboard sensors. Specifically, the paper presents the MPC formulation for hazard avoidance using a light detection and ranging (LIDAR) sensor and applies it to a case study to investigate the impact of model fidelity on the performance of the algorithm, where performance is measured mainly by the time to reach the target point. Towards this end, the case study compares a 2 degrees-of-freedom (DoF) vehicle dynamics representation to a 14 DoF representation as the model used in MPC. The results show that the 2 DoF model can perform comparable to the 14 DoF model if the safe steering range is established using the 14 DoF model rather than the 2 DoF model itself. The conclusion is that high fidelity models are needed to push autonomous vehicles to their limits to increase their performance, but simulating the high fidelity models online within the MPC may not be as critical as using them to establish the safe control input limits.

Commentary by Dr. Valentin Fuster

Vehicle Path Planning and Collision Avoidance

2013;():V003T47A001. doi:10.1115/DSCC2013-3858.

This paper presents a nonlinear Model Predictive Control approach to controlling a tractor-trailer system. Using a nonlinear tractor-trailer model, the controller determines the optimal steer angle, based on the trailer’s measured position and heading, as well as information about the path geometry in front of it. Then, the computer determines the amount of voltage to apply to the steering wheel motor to achieve the necessary steer angle. In the simulation study, the controller algorithm is capable of guiding a 2-1/2 meter long trailer around a 5-meter radius turn, towed by a four wheel drive off-road utility vehicle, with a maximum error of 8.5 centimeters.

Commentary by Dr. Valentin Fuster
2013;():V003T47A002. doi:10.1115/DSCC2013-3909.

Surveillance missions that involve unmanned ground vehicles (UGVs) include situations where a UGV has to choose between alternative paths to complete its mission. Currently, UGV missions are often limited by the available on-board energy. Thus, we propose a dynamic most energy-efficient path planning algorithm that integrates mission prior knowledge with real-time sensory information to identify the mission’s most energy-efficient path. Our proposed approach predicts and updates the distribution of energy requirement of alternative paths using recursive Bayesian estimation through two stages: (1) exploration — road segments are explored to reduce their energy prediction uncertainty; (2) exploitation — the most energy-efficient path is selected using the collected information in the exploration stage and is traversed. Our simulation results show that the proposed approach outperforms offline methods, as well as a method that only relies on exploitation to identify the most energy-efficient path.

Commentary by Dr. Valentin Fuster
2013;():V003T47A003. doi:10.1115/DSCC2013-3947.

This paper demonstrates that an autonomous vehicle can perform emergency lane changes up to the limits of handling through real-time generation and evaluation of bi-elementary paths. Path curvature and friction limits determine the maximum possible speed along the path and, consequently, the feasibility of the path. This approach incorporates both steering inputs and changes in speed during the maneuver. As a result, varying path parameters and observing the maximum possible entry speed of resulting paths gives insight about when and to what extent a vehicle should brake and turn during emergency lane change maneuvers. Tests on an autonomous vehicle validate this approach for lane changes at the limits of handling.

Topics: Vehicles
Commentary by Dr. Valentin Fuster
2013;():V003T47A004. doi:10.1115/DSCC2013-3951.

This paper presents a multi-objective safety system that is capable of avoiding unintended collisions with stationary and moving road obstacles, vehicle control loss as well as unintended roadway departures. The safety system intervenes only when there is an imminent safety risk while full control is left to the driver otherwise. The problems of assessing wether an intervention is required as well as controlling the vehicle motion in case an intervention is needed are jointly formulated as a single optimization problem, that is repeatedly solved in receding horizon. The novelty of the formulation lies in the ability of simultaneously avoiding moving obstacles while assessing the necessity thereof. The versatility of the proposed formulation is demonstrated through simulations showing its ability of avoiding a wide range of accident scenarios.

Commentary by Dr. Valentin Fuster
2013;():V003T47A005. doi:10.1115/DSCC2013-4000.

This research focuses on determining the minimum preview time needed to predict and prevent vehicle rollover. Statistics show that although rollover only occurs in 2.2% of total highway crashes, it accounts for 10.7% of total fatalities. There are several dynamic rollover metrics in use that measure a vehicle’s rollover propensity under specified conditions. However, in order to prevent a rollover event from occurring, it is necessary to predict a vehicle’s future rollover propensity. This research uses a novel vehicle rollover metric, called the zero-moment point (ZMP), to predict a vehicle’s rollover propensity. Comparing different amounts of preview, the results show that short-range predictions — as little as 0.75 seconds ahead of the vehicle — are sufficient to prevent nearly all dynamics-induced rollovers in typical shoulders and medians.

Topics: Vehicles
Commentary by Dr. Valentin Fuster
2013;():V003T47A006. doi:10.1115/DSCC2013-4106.

Racecar drivers are skilled at tracking a path, avoiding accidents, and controlling their vehicles at the limits of handling. Better understanding of how a skilled driver selects and drives a racing line, could potentially lead to a new technique for obstacle avoidance. To investigate this, the characteristics of a racecar driver’s line must be captured mathematically. This paper describes an algorithm for fitting a path to the GPS data of a driver’s racing line. A family of path primitives composed of straights, clothoids, and constant radius arcs are used to describe the racing line. The fitted paths provide a method for analyzing racing lines and the different techniques used by skilled drivers to navigate the track.

Topics: Algorithms , Fittings
Commentary by Dr. Valentin Fuster

Vibrational and Mechanical Systems

2013;():V003T48A001. doi:10.1115/DSCC2013-3910.

One of the predominant difficulties in the theory of distributed structure control systems comes from the fact that these resonant structures have a large number of active modes in the working band-width. Among the different methods for vibration control, Positive Position Feedback (PPF) is of interest, which uses piezoelectric actuation to overcome the vibration as a collocated controller. Modified Positive Position Feedback (MPPF) is later presented by adding a first-order damping compensator to the conventional second-order compensator, to have a better performance for steady-state and transient disturbances. In this paper, Multivariable Modified Positive Position Feedback (MMPPF) is presented to suppress the unwanted resonant vibrations in the structure. This approach benefits the advantages of MPPF, while it controls larger number vibration modes. An optimization method is introduced, consisting of a cost function that is minimized in the area of the stability of the system. LQR problem is also used to optimize the controller performance by optimized gain selection. It is shown that the LQR-optimized MMPPF controller provides vibration suppression in more efficiently manner.

Commentary by Dr. Valentin Fuster
2013;():V003T48A002. doi:10.1115/DSCC2013-3955.

Cranes are used in manufacturing facilities, throughout nuclear sites, and in many other applications for various heavy-lifting purposes. Unfortunately, the flexible nature of cranes makes fast and precise motion of the payload challenging and dangerous. Certain applications require the coordinated movement of multiple cranes. Such tasks dramatically increase the complexity of the crane operation. This paper studies the dynamic behavior of a dual-hoist bridge crane. Simulations and experiments are used to develop an understanding of the dynamic response of the system. Various inputs and system configurations are analyzed and important response characteristics are highlighted.

Commentary by Dr. Valentin Fuster
2013;():V003T48A003. doi:10.1115/DSCC2013-4068.

The dynamic control of stacked-plate mechanical systems such as circuit board assemblies is a common technical problem that often requires a complete description of the open loop system dynamics prior to controller development. Often, a preliminary finite element model (FEM) of the test article is developed to understand the dynamics of the system to perform a modal test. The results of this modal test must then be used to update the stiffness, mass and damping matrices to yield correct FEM frequencies mode shapes and damping. This work describes the mathematical development of a finite element model of a multi-plate test article and proceeds with a model update using differentiated velocity data collected at discrete points on the structure with a laser Doppler vibrometer (LDV) and drive point measurements collected at the excitation location with an impedance head. Using these data, accelerance FRFs were computed and the first three flexible mode shapes were estimated and these shapes were compared to the corresponding FEM shapes using both percent frequency difference and modal assurance criterion (MAC). Several parameters of the system model were modified yielding improved correlation with the experimental results.

Commentary by Dr. Valentin Fuster
2013;():V003T48A004. doi:10.1115/DSCC2013-4075.

This paper develops governing equations for material strain and tension based on a temperature distribution model when the flexible materials (often called webs) are transported on rollers through heat transfer processes within roll-to-roll (R2R) processing machines. Heat transfer processes are employed widely in R2R systems that contain process operations such as printing, coating, lamination, etc., which require heating/cooling of the moving web material. The heat transfer processes introduce the thermal expansion/contraction of the material and changes in the elastic modulus. Thus, the temperature distribution in the moving material affects the strain distribution in the material. Because of change in strain as well as modulus as a function of temperature, tension in the material resulting from elastic strain is also affected by heating/cooling of the web. To obtain the temperature distribution, two basic heat transfer modes are considered: web wrapped on a heat transfer roller and the web span between two consecutive rollers. The governing equations for strain is then obtained using the law of conservation of mass considering the temperature effects. Subsequently, a governing equation for web tension is obtained by assuming the web is elastic with the modulus varying with temperature; an average modulus is considered for defining the constitutive relation between web strain and tension. Since it is difficult to obtain measurement of tension using load cell rollers within heat transfer processes, a tension observer is designed. To evaluate the developed governing equations, numerical simulations for a single tension zone consisting of a heat transfer roller, a web span, and a driven roller are conducted. Results from these numerical model simulations are presented and discussed.

Commentary by Dr. Valentin Fuster
2013;():V003T48A005. doi:10.1115/DSCC2013-4089.

We present an alternative to averaging methods for vibrational control design of second-order systems. This method is based on direct application of the stability map of the linearization of the system at the desired operating point. The paper focuses on harmonic forcing, for which the linearization is Mathieu’s equation, but somewhat more general periodic forcing functions may be handled. When it is applicable, this method achieves significantly greater functionality than previously reported approaches. This is demonstrated on two sample systems. One is the vertically driven inverted pendulum, and the other is an input-coupled bifurcation control problem arising from electrostatic MEMS comb drives.

Commentary by Dr. Valentin Fuster
2013;():V003T48A006. doi:10.1115/DSCC2013-4095.

Statically-balanced mechanisms have been widely used for passive compensation of gravity loads in many applications including neurological rehabilitation and micro-/reduced-gravity simulation. For these applications, it is desirable that the used mechanism has minimal impedance the interacting human can feel. Impedance of a statically-balanced mechanism is contributed by many factors including the payload on the end-effector and the joint friction. This paper studies the relation between the end-effector impedance and the load-dependent joint friction for statically-balanced mechanisms. In the study a load dependent joint friction force model was developed. With the model, contribution of the end-effector load or payload on the joint friction can be evaluated, from which the end-effector impedance of the mechanism caused by the joint friction can be computed. The study results are applied to the analysis of a reduced-gravity simulator to evaluate the effect of the joint friction on the end-effector impedance of the mechanism. The findings of the study can help the assessment of the dynamic performance and also help the optimal design of statically-balanced mechanisms.

Commentary by Dr. Valentin Fuster

Wind Energy Systems and Control

2013;():V003T49A001. doi:10.1115/DSCC2013-3796.

This paper presents a control strategy that combines altitude and crosswind motion control for tethered wind energy systems with airborne turbines and generators. The proposed algorithm adjusts altitude and induces an appropriate level of crosswind motion to present the system with an apparent wind speed that most closely meets, but does not exceed, the rated wind speed of the on-board turbine(s), thereby tracking the turbine’s optimal power point. The adjustment of both altitude and motion control, along with the reduction in crosswind motion and altitude when the rated wind speed is exceeded, differentiates the proposed control architecture from other strategies proposed in the literature. Initial control laws and simulation results are presented for the Altaeros lighter-than-air wind energy system.

Commentary by Dr. Valentin Fuster
2013;():V003T49A002. doi:10.1115/DSCC2013-3826.

An optimal control approach for a wind turbine drivetrain with the objective of maximizing energy harvesting and minimizing noise emission is presented. One of the major challenges facing the public acceptance for continuous growth of wind turbine installation is its noise emission. However, reducing the noise emission could lead to decreased wind energy harvesting. As a result, a tradeoff arises between power generation and noise emission, especially when a wind turbine operates under the partial-load condition.

This paper will show that through controlling the generator electromagnetic torque and/or the blade pitch angle, an optimal tradeoff between wind turbine energy harvesting and noise emission can be obtained. The dynamic model of a wind turbine drivetrain and a noise emission prediction model are also presented. Simulation results of using the proposed control design for different wind speed ranges are analyzed and compared.

Commentary by Dr. Valentin Fuster
2013;():V003T49A003. doi:10.1115/DSCC2013-3840.

Auto-rotation or autogyro is a well-known phenomenon where a rotor in a wind field generates significant lift while the wind induces considerable aerodynamic torque on the rotor. The principle has been studied extensively for applications in aviation. However, with recent works indicating immense, persistent, and pervasive, available wind energy at high altitudes, the principle of autogyro could potentially be exploited for energy harvesting. In this paper, we carry out a preliminary investigation on the viability of using autogyros for energy extraction. We mainly focus on one of the earliest documented works on modeling of autogyro and extend its use to explore energy harvesting. The model is based on blade element theory. We provide simulation results of the concept. Although the results are encouraging, there are various practical aspects that need to be investigated to build confidence on this approach of energy harvesting. This work aims to build a framework upon which more comprehensive research can be conducted.

Topics: Wind energy
Commentary by Dr. Valentin Fuster
2013;():V003T49A004. doi:10.1115/DSCC2013-3957.

Transient and harmonic stresses in wind turbine rotor shafts contribute to gearbox failure. This paper investigates the reduction of rotor shaft torsional vibrations through active control of the generator torque. A 5 MW turbine model is used to test the procedure. A model of a permanent magnet synchronous generator is included as part of the wind turbine simulation. The simulations are carried out using the software FAST from the National Renewable Energy Laboratory (NREL). The PI and feedback linearized controller for the generator is derived together with the means for vibration isolation. Examples of steady, time varying, and turbulent wind are presented which all show significant reduction in the torsional oscillations.

Commentary by Dr. Valentin Fuster
2013;():V003T49A005. doi:10.1115/DSCC2013-3971.

This paper proposes a novel control approach for optimizing wind farm energy capture with a nested-loop scheme of extremum seeking control (ESC). Similar to Bellman’s Principle of Optimality, it has been shown in earlier work that the axial induction factors of individual wind turbines can be optimized from downstream to upstream units in a sequential manner, i.e. the turbine operation can be optimized based on the power of the immediate turbine and its downstream units. In this study, this scheme is illustrated for wind turbine array with variable-speed turbines for which torque gain is controlled to vary axial induction factors. The proposed nested-loop ESC is demonstrated with a 3-turbine wind farm using the SimWindFarm simulation platform. Simulation under smooth and turbulent winds show the effectiveness of the proposed scheme. Analysis shows that the optimal torque gain of each turbine in a cascade of turbines is invariant with wind speed if the wind direction does not change, which is supported by simulation results for smooth wind inputs. As changes of upstream turbine operation affects the downstream turbines with significant delays due to wind propagation, a cross-covariance based delay estimate is proposed as adaptive phase compensation between the dither and demodulation signals.

Topics: Wind farms
Commentary by Dr. Valentin Fuster
2013;():V003T49A006. doi:10.1115/DSCC2013-4069.

Maintaining the accumulator pressure regardless of its energy level and tracking the power demanded by the electrical grid are two potential advantages of the Compressed Air Energy Storage (CAES) system proposed in [1, 2]. In order to achieve these goals, a nonlinear controller is designed motivated by an energy-based Lyapunov function. The control inputs of the storage system include displacement of the pump/motor in the hydraulic transformer and displacement of the liquid piston air compressor/expander. While the latter has a relatively low bandwidth, the former is a faster actuator with a higher bandwidth. In addition, the pneumatic path of the storage vessel that is connected to the liquid piston air compressor/expander has a high energy density, whereas the hydraulic path of the storage vessel is power dense. The nonlinear controller is then modified to achieve a better performance for the entire system according to these properties. In the proposed approach, the control effort is distributed between the two pump/motors based on their bandwidths: the hydraulic transformer reacts to high frequency events, while the liquid piston air compressor/expander performs a steady storage/regeneration task. As a result, the liquid piston air compressor/expander will loosely maintain the accumulator pressure ratio and the pump/motor in the hydraulic transformer will precisely track the desired generator power. This control scheme also allows the accumulator to function as a damper in the storage system by absorbing power disturbances from the hydraulic path generated by wind gusts.

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