Nonlinear Robot ARM Control Through Third Order Motor Model

Abstract

Design of robot controllers are typically based on the second order model of rigid robot arms. However, the robot joint motor dynamics constitutes an important part of the overall robot dynamics, and should be included in robot controller design. A third order robot dynamic model is obtained in this paper which includes the robot arm dynamics as well as the actuator dynamics. Differential geometric control theory is used to linearize and decouple the resultant nonlinear and coupled dynamic system, yielding linear subsystems for each degree of freedom in the robot task space. Linear system control theory and optimal control theory are then used to obtain desired system performance. Extensive computer simulations are carried out using the joint actuators. They demonstrate that the controller designed in this paper produces excellent steady state as well as transient system performance, and it is robust in the presence of robot model parameter inaccuracy. The importance of robot motor dynamics is further illustrated by computer simulations, where a controller based on the second order robot arm dynamical model is adopted to control the third order robot dynamics and produces unacceptable transient state response.