Comparison of Adaptive Control Methods for Trajectory Tracking Robots

A. Slobodinsky, Y. Strassberg, and J. Dayan (Israel)

Keywords

Adaptive Control, Robotics, Trajectory following, Costfunctions, Performance criteria

Abstract

The present paper compares sixteen different adaptive control methods for robot manipulators representatives of the different approaches published during the last two decades. The comparison is based on simulated results of the selected algorithms implemented on an industrial robot, the Mitsubishi RV-M2. The simulation emulates tracking control along two trajectories with different conditions: ideal implementation without disturbances, changes in system parameters, external disturbances, unmodeled dynamics and speed of motion. Two different cost functions were defined and weights assigned to errors, moments, computation cost and margin of stability. These functions are calculated for the different control algorithms considered; these are then served as the basis for the comparison. Several algorithms performed better than the others, achieving better results for certain operations under different conditions, and, therefore, were selected and recommended for similar tasks.

Important Links:



Go Back