Design of Adaptive Sliding Mode Controller Based on Neural Network for Robot Manipulator

This article presents a method for synthesizing a robot manipulator's adaptive sliding mode controller based on a neural network. In actual working conditions, the robot's dynamic equation has strong nonlinearity, the parameters change uncertainly, and in many cases, the robot is affected by unmeasured external disturbances. Using an RBF neuron network and adaptive control, we propose a solution to approximate and compensate for the uncertain components and external disturbances. The robust control term based on sliding mode control is designed to overcome approximation errors with chattering in the control signal reduced to a minimum. The simulation outcomes indicate that the robot controller suggested in this article possesses high quality, adaptability, and robust resistance to interference.