Online Learning of Virtual Impedance Parameters in Non-Contact Impedance Control Using Neural Networks

Tsuji Toshio

Terauchi Mutsuhiro

Tanaka Yoshiyuki

IEEE Transactions on Systems, Man, and Cybernetics-Part B: Cybernetics

Impedance control is one of the most effective methods forcontrolling the interaction between a manipulator and a task environment.In conventional impedance control methods, however, the manipulatorcannot be controlled until the end-effector contacts task environments. Anoncontact impedance control method has been proposed to resolve such aproblem. This method on only can regulate the end-point impedance, butalso the virtual impedance that works between the manipulator and theenvironment by using visual information. This paper proposes a learningmethod using neural networks to regulate the virtual impedance parametersaccording to a given task. The validity of the proposed method wasverified through computer simulations and experiments with a multijointrobotic manipulator.

Impact control

impedance control

noncontact impedance

neural networks (NN)

robot manipulator

機械工学 [ 530 ]

英語

IEEE

Copyright (c) 2004 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.

[ISSN] 1083-4419

[DOI] http://dx.doi.org/10.1109/TSMCB.2004.829133 ～の異版である