Neural Network-Based Adaptive Optimal Terminal Sliding Model Controller for Robot Manipulator Trajectory Tracking

With the development of robotic technology, robots are increasingly used in many manufacturing and production fields. Position control is fundamental for robot operation, ensuring accurate performance. However, the absence of a precise robot dynamics model often limits the precision of position control. To address this issue, a neural network-based adaptive optimal terminal sliding mode controller is proposed in this research. Firstly, a neural network is established to approximate unknown dynamics. Then, a terminal sliding mode surface is designed to promote the convergence of tracking errors. An adaptive terminal sliding mode position controller is subsequently constructed. Finally, we propose a whale optimization algorithm (WOA)-based method to optimize the control parameters. Simulations and experiments are conducted to verify the effectiveness and superiority of the designed controller.