Neural Network-Based Funnel Adaptive Surge Motion Tracking Control of Autonomous Underwater Vehicle with Unknown Deadzone

This paper investigates the asymptotic state tracking control with guaranteed transient performance for the surge motion of autonomous underwater vehicles with unknown input deadzone. A neural network-based funnel adaptive controller is designed based on Lyapunov theory, with funnel control strategy used to guarantee tracking performance. Neural network is used to approximate uncertainty. Robust strategy is adopted for the compensation of input deadzone. To reduce the design complexity, a differentiator is constructed to estimate the time derivative of virtual control law. The control algorithm can not only guarantee asymptotic state tracking, but also transient performance. The simulation results show the effectiveness of the proposed algorithm.