Adaptive Nonsingular Terminal Sliding Mode Control of Underwater Manipulator Based on Neural Network and Combined Reaching Law

To address the trajectory tracking control of underwater manipulators with the interference of complex underwater environment, this paper proposes a nonsingular fast terminal sliding mode (NFTSM) control strategy based on an RBF neural network (RBFNN) and constructs a combined reaching law. The suggested NFTSM control strategy has the advantages of rapid convergence speed and finite convergence time. The constructed composite approaching law can accelerate the stabilization of the system. The neural network is chosen to maximally approximate the uncertain disturbances, including the model uncertainty error term and external disturbance term, which could reduce the chattering problem of the control system. Furthermore, the adaptive law is designed based on the Lyapunov theorem to make the entire system stable. This method is simulated on a 4-DOF manipulator for trajectory tracking, and its superior performance under random disturbances is demonstrated compared with the conventional sliding mode control (CSMC).