Hybrid trajectory planning and tracking for automatic berthing: A grid-search and optimal control integration approach

Abstract

Automatic berthing is considered one of the most challenging problems for underactuated unmanned surface vehicles (USVs) due to the influence factors such as rudder effect, environmental disturbances, and control uncertainties. This paper focuses on trajectory planning and tracking to address this problem. The hybrid approach initially employs a reverse search strategy to plan the global berthing trajectory, with the terminal part of this trajectory subsequently being reconstructed and optimized by taking into account environmental loads and USV dynamic model. A nonlinear MPC controller is designed for real-time tracking of the berthing trajectory. The rudder effect constraints related to USV speed is supplemented to the dynamic model to alleviate the problem of poor maneuverability at low speed. The whole trajectory planning and tracking scheme of automatic berthing is simulated at Dalian Port considering environmental loads and uncertainties in control. The results show that the scheme has quite fast solving speed of trajectory planning and robust control of trajectory tracking.