APF-based intelligent navigation approach for USV in presence of mixed potential directions: Guidance and control design

Abstract

This paper investigates the real-time intelligent navigation approach for unmanned surface vehicle (USV), considering constraints of the randomly moving target and multiple static or moving obstacles. In the scheme, an intelligent guidance principle is developed by employing the traditional dynamic virtual ship (DVS) structure and the artificial potential field (APF) technique. The improved design for repulsive potential field can guarantee the reasonable obstacle avoidance capability. Especially, the amended repulsive terms are derived by the velocity and orientation factors of moving obstacles. Combining with the APF-based guidance, the robust neural path following control algorithm is proposed by employing the minimal learning parameter (MLP) and the nonlinear disturbance observer (DOB) technique. For merit of the improved design of DOB, the attitude of USV can be effectively stabilized to that of virtual ship. That can derive the state-of-art trade-off between the complicated control law and the hardware computing burden. Through the Lyapunov synthesis, all states of USV are with the semi-globally uniform ultimate bounded (SGUUB) stability. Two experiments have been illustrated to verify the obstacle avoidance and dynamic tracking performance of the proposed strategy.