Neuroadaptive Control for Trajectory Tracking of Indirect Drive Robots

Most industrial robots are indirect drive robots, which utilize low torque and high speed motors. Indirect drive robots have gear reducers between the motors and links. Due to the flexibility of transmission units, it is difficult to achieve high accuracy for trajectory tracking. In this paper, a neuroadaptive control, which is essentially a neural network (NN) based adaptive back-stepping control approach, is proposed to deal with this problem. The stability of the proposed approach is analysed using Lyapunov stability theory. A data-driven approach is also proposed for the training of the neural network. The effectiveness of the proposed controller is verified by simulation of both single joint and 6-axis industrial robots.