AI-based Regression Analysis for Optimizing the Performance of Robot Manipulator Trajectory Tracking

Abstract

Robotics is one of the major research fields of smart manufacturing. With the support of artificial intelligence (AI), robot manipulators can be more automated and can even learn from their own behaviors to achieve certain goals. The human maneuvers can be reduced more since the robot is more generalized and can be used in various scenarios. In this paper, an AI-based compensation method is proposed to minimize the trajectory error of the robot manipulator tool center point (TCP) under simulated external excitations and interferences during milling. A simulated KUKA LBR iiwa robot model is used as the analysis object, several different trajectories and external excitations are generated and are applied on this robot model. The simulation results show that the proposed compensation method is simple to deploy with low cost. Besides, the proposed method is significantly effective in reducing the TCP translation errors and it has good generalization ability under different motion scenarios.