A discrete‐time optimal adaptive control law for a robot arm

Abstract

Abstract The equations of motion describing a robot's dynamics are coupled and non‐linear, making the design of an optimum controller difficult using classical techniques. In this work an explicit adaptive control law is proposed based on a discrete linear model for each link and on the minimization of a quadratic performance criterion for position error and total control effort. The system parameters are recursively estimated at each control step by use of least squares, with a typical sample time of 0.02 s. A computer simulation of the resulting scheme is performed to evaluate the controller. The simulation model, based on the first three links of an existing robot, includes detailed motor dynamics and treats the wrist assembly as a load mass. Simulated test paths requiring movement of the outer two links indicate that the controller adapts and that its behaviour is stable and convergent.