Page 31 - ASME DSCC 2015 Program
P. 31

Technical Program




              combined optimal design and control of a Near Isothermal Liquid   A Control Strategy for an Active Alignment Transtibial Prosthesis
              Piston Air Compressor/expander for a Compressed Air Energy Storage   Invited session paper. DSCC2015-9948
              (caes) System for Wind Turbines                        Andrew laPre, frank Sup, University of Massachusetts Amherst, Amherst,
              Invited session paper. DSCC2015-9957                   MA, United States
              Mohsen Saadat, Perry Y. li, University of Minnesota, Minneapolis, MN,   This paper presents a control approach for an experimental transtibial prosthesis
              United States
                                                                     that can actively realign the residual limb in relation to prosthetic foot during the
              The key component of a Compressed Air Energy Storage (CAES) system    stance phase of gait. The realignment objective is to inject positive power into the
              is the air compressor/expander. The roundtrip efficiency of this energy   gait cycle while actively reducing the magnitude of the sagittal moment trans-
              storage technology highly depends on the efficiency of the air compressor/  ferred to the residual limb. The altered gait dynamics of this new type of prosthe-
              expander. Decreasing compression/expansion time Increases power-den-  sis require a control approach that coordinates its function with a user’s gait cycle.
              sity which reduces cost. There is a trade off between the thermal efficiency   This paper overviews the mechanical design of the prosthesis development, the
              and power density of this component. Different ideas and approaches were   proposed finite-state adaptive controller, and presents experimental results for
              introduced and studied in the previous works to improve this trade off by   constant cadence walking and adaptation while changing walking speeds
              enhancing the heat transfer between air and its environment. In the present
                                                                     Wheel-slip Based Inverse Solution for differential drive Mobile robots
              work, a combination of optimal compression/expansion trajectories, optimal
                                                                     Invited session paper. DSCC2015-9854
              chamber shape and optimal heat exchanger material distribution in the
              chamber is considered to maximize the power density of a compression/   alan Whitman, garrett clayton, alex poultney, hashem ashrafiuon,
                                                                     Villanova University, Villanova, PA, United States
              expansion chamber for a given desired efficiency. Results show that the
              power density can be improved by more than 20 folds if the optimal combi-  An approximate inverse dynamic solution for differential drive mobile robots
              nation of flow rate, shape and porosity are used together.  is presented where the wheel speeds are analytically calculated for a spec-
                                                                     ified desired trajectory. The differential equations of motion of the mobile
              InVITED SESSIon                                        robot are nondimensionalized and a small parameter is defined based on
              2-20-1  WM5  Control of Robotic Systems                the ratio of wheel slip and wheel speed variation time scales. Approximate
              Emerson Burkhart A                      1:30pm–3:30pm
                                                                     dominant and first order inverse solutions are derived for general trajecto-
                                                                     ries. Next, a trajectory planning strategy is devised to ensure continuous
              Session Organizer: Kevin fite, Clarkson University     wheel speed profiles in achieving any desired target trajectory. The case of
              Session Organizer: Mo Rastgaar, Michigan Technological University  target circular trajectories is specifically considered and with inverse solu-
              Session Organizer: frank Sup, University of Massachusetts, Amherst  tions which are verified both through simulations and experiments.
              Session Organizer: Robert Gregg, University of Texas, Dallas
              Session Chair: Jim Dabney, University of Houston-Clear Lake  Trajectory Tracking of a Pendulum-Driven Spherical Robot
                                                                     Contributed regular paper. DSCC2015-9727
              Session Co-Chair: hashem ashrafiuon, Villanova University
                                                                     francois R. Hogan, McGill University, Montreal, QC, Canada,
              Torque estimation in a Wrist rehabilitation robot using a Nonlinear   James R. forbes, University of Michigan, Ann Arbor, MI, United States
              Disturbance observer
                                                                     In this paper, a controller is designed to achieve tracking of a pendu-
              Invited session paper. DSCC2015-9717
                                                                     lum-driven spherical robot about a desired trajectory in the presence of
              Mohammadhossein Saadatzi, David C. long, Colorado School of Mines,   model uncertainty and external disturbances. A methodology is proposed to
              Golden, CO, United States, ozkan Celik, Colorado School of Mines, XX, CO,   design a feedforward controller for the spherical robot by approximating the
              United States
                                                                     desired trajectory using circles as basis functions. A linear state-feedback
              In this paper, we present implementation of a nonlinear disturbance    controller is used to drive the perturbations from the desired trajectory to
              observer algorithm to estimate disturbance torques on a wrist rehabilitation   zero. Numerical simulations show that the controller realizes tracking of the
              robot. The ultimate goal is to enable accurate estimation of user interaction   desired trajectory with perturbed initial conditions.
              torques without force/torque sensors. A dynamic model of the Wrist Gimbal
                                                                     Complete Dynamic Modelling of flexible Joint Robots
              two degree-of-freedom forearm and wrist exoskeleton is developed. A non
                                                                     Contributed short paper. DSCC2015-9813
              linear disturbance observer (NDO) algorithm that incorporates the robot
                                                                     yu Zhao, xiaowen yu, Masayoshi Tomizuka, University of California,
              dynamics is implemented. Friction models for both joints of the robot are
                                                                     Berkeley, CA, United States, Cong Wang, New Jersey Institute of
              identified. The robot dynamic model and the accuracy of the disturbance
                                                                     Technology, Newark, NJ, United States
              torque estimations are experimentally verified under closed-loop trajectory
              tracking scenarios.                                    Joint flexibility is common in industrial robots that have geared joints. In order
                                                                     to design a precision motion controller that compensates the effects of joint
                                                                     elasticity, an accurate dynamic model of flexible joint robots is required. The
                                                                     models that are commonly used ignore the gyroscopic interactions be-
                                                                     tween the motors and links. In order to evaluate the influence of the ignored
                                                                     gyroscopic interaction, a complete dynamic model for flexible joint robots is
                                                                     derived in this paper. It is shown that when to realize high accuracy for high
                                                                     velocity trajectory tracking, the motor inertia is non-negligible compared to link   31
                                                                     inertia, and that the neglected interaction terms must be taken into account.
   26   27   28   29   30   31   32   33   34   35   36