Reinforcement Learning and Robust Control for Robot Compliance Tasks

  • Authors:

  • Cheng-Peng Kuan;Kuu-Young Young

  • Affiliations:

  • Department of Electrical and Control Engineering, National Chiao-Tung University, Hsinchu, Taiwan/ e-mail: Email: kyoung@cc.nctu.edu.tw;Department of Electrical and Control Engineering, National Chiao-Tung University, Hsinchu, Taiwan/ e-mail: Email: kyoung@cc.nctu.edu.tw

  • Venue:
  • Journal of Intelligent and Robotic Systems
  • Year:
  • 1998

Quantified Score

Hi-index 0.00

Visualization

Abstract

The complexity in planning and control of robot compliance tasks mainlyresults from simultaneous control of both position and force and inevitablecontact with environments. It is quite difficult to achieve accuratemodeling of the interaction between the robot and the environment duringcontact. In addition, the interaction with the environment varies even forcompliance tasks of the same kind. To deal with these phenomena, in thispaper, we propose a reinforcement learning and robust control scheme forrobot compliance tasks. A reinforcement learning mechanism is used to tacklevariations among compliance tasks of the same kind. A robust compliancecontroller that guarantees system stability in the presence of modelinguncertainties and external disturbances is used to execute control commandssent from the reinforcement learning mechanism. Simulations based ondeburring compliance tasks demonstrate the effectiveness of the proposedscheme.