Collision Avoidance and Path Point Tracking Control for Underactuated Autonomous Surface Vehicles With Unknown Model Nonlinearity

Abstract

While sailing along an expected route, autonomous surface vehicles (ASVs) may encounter the static or dynamic obstacles. Therefore, from the perspective of safety, ASVs must have the ability of automatic collision avoidance. In this paper, an obstacle avoidance and path point tracking control frame for ASVs is proposed. The frame includes two parts. One is the decision-making module, in which the desired sailing speed and course angular velocity are generated. More specifically, through adopting the velocity obstacle approach (VOA), the feasible obstacle avoidance action set is obtained. Then, considering the maneuvering characteristics of ASVs, the discrete optional obstacle avoidance actions are obtained by using the dynamic window approach (DWA). Finally, by introducing the International Regulations for Preventing Collisions at Sea (COLREGs), the obstacle avoidance actions that meet the rules can be screened out. An evaluation function is designed to select the final practical obstacle avoidance action. The second part is the dynamic controller module. A radial basis function (RBF) neural network-based path point tracking controller design is given in this part. The RBF neural network is designed to estimate the unknown model nonlinearity, and the stability of the closed-loop system is proved. Finally, simulations and experiments are carried out to illustrate the effectiveness of the presented algorithm.