Adaptive Neural Network Position/force Control of Robot Manipulators with Model Uncertainties*

In this paper, adaptive neural network position/force control of robot manipulators with model uncertainties is considered. The controller combines a neural network modeling technique with self-tuning fuzzy control which describes the relationship between force and position/velocity error. And robust control can be easily incorporated to suppress the neural network modeling errors and the bounded disturbances. Simulation results based on 2-DOF robot show the effectiveness of this approach

I. INTRODUCTION

Industrial robotic tasks are generally related to. manipulation, which require only controlling the position of the arm, but other tasks like assembly, pushing and polishing require interaction between the manipulator's end-effector and the environment, and consequently the development of more sophisticated force controlalgorithms^[1]~[2].