Building up a Control Algorithm Desired Compensation Adaptive with Reduced On-Line Computation for Robot Almega 16

Abstract

This paper presents a build of improving algorithm adaptive control in joint space for the motion system of the Almega16 manipulator. The proposed controller eliminates the need for on-line computation of the regression matrix by replacing joints position and velocity with the desired joints position and velocity. In a adaptive control algorithm, the kinetic parameters identification system always provides state information update on the time-variant parameters. The true value is compared to the reference set-point and its evaluation result is input to the controller for adjustment. The results from Matlab - Simmechanic simulations and experiments show that the motion system of Robot Almega16 satisfies the requirement of a control system: the errors of rotating joints quickly converge to zero within a short transient time, so that closed-loop system is stable based on Lyapunov method.