First Steps Toward Translating Robotic Walking To Prostheses: A Nonlinear Optimization Based Control ApproachReport as inadecuate


First Steps Toward Translating Robotic Walking To Prostheses: A Nonlinear Optimization Based Control Approach


First Steps Toward Translating Robotic Walking To Prostheses: A Nonlinear Optimization Based Control Approach - Download this document for free, or read online. Document in PDF available to download.

This paper presents the first steps toward successfullytranslating nonlinear real-time optimization based controllersfrom bipedal walking robots to a self-contained poweredtransfemoral prosthesis: AMPRO, with the goal of improvingboth the tracking performance and the energy efficiencyof prostheses control. To achieve this goal, a noveloptimal control strategy combining control Lyapunov functionCLF based quadratic programs QP with impedancecontrol is proposed. This optimal controller is first verifiedon a human-like bipedal robot platform, AMBER. The resultsindicate improved compared to variable impedancecontrol tracking performance, stability and robustness tounknown disturbances. To translate this complete methodologyto a prosthetic device with an amputee, we begin by collectingreference human locomotion data via Inertial measurementUnits IMUs. This data forms the basis for anoptimization problem that generates virtual constraints, i.e.,parameterized trajectories, specifically for the amputee andthe prosthesis. A online optimization based controller is utilized to optimally track the resulting desired trajectories. The parameterization of the trajectories is determined througha combination of on-board sensing on the prosthesis togetherwith IMU data, thereby coupling the actions of the user withthe controller. Importantly, the proposed control law displaysremarkable tracking and improved energy efficiency,outperforming PD and impedance control strategies. Thisis demonstrated experimentally on the prosthesis AMPROthrough the implementation of the holistic sensing, algorithmand control framework, with the end result being stableprosthetic walking by an amputee.



Advanced Mechanical Bipedal Experimental Robotics Lab AMBER Publications - Advanced Mechanical Bipedal Experimental Robotics Lab AMBER -



Author: Zhao, Huihua - Horn, Jonathan - Reher, Jacob - Paredes, Victor - Ames, Aaron D. - -

Source: https://smartech.gatech.edu/







Related documents