Autonomous path planning for unmanned surface vehicles using the fast-sweeping method in island and reef waters

This paper proposes a fast path planning scheme for Unmanned Surface Vehicles (USVs) operating in complex marine environments. First, a Variable Fast-Sweeping Method (VFSM) is introduced, based on the principles of the Fast-Sweeping Method. The VFSM aims to optimize the planning route by balancing the length of the USV's path and addressing collision risks, achieving a compromise between the two objectives. Compared to conventional path planning approaches, the proposed VFSM significantly enhances computational efficiency and is particularly well-suited for environments with complex obstacles. Second, the ship Tam model is utilized to establish an Angle-Guided Ship Domain Model (AFSM) and a State Constraint Model (ACFSM), which account for the USV's status messages during navigation under the influence of state constraints. The impact of the USV's state, combined with other obstacles and the challenges of adhering to the planned route, is effectively addressed through dynamic route re-planning. The re-planned path complies with both the USV's state requirements and dynamic obstacle constraints. Finally, once the optimal collision-free path is generated, the moving obstacle's constraints are incorporated using the AFSM, which expands restricted regions based on ship speed calculations. This allows the planned route to be dynamically adjusted in response to moving obstacles. Simulations using simunps demonstrate that the proposed method generates continuous, safe, and realistic paths, effectively handling complex maritime traffic conditions. The results confirm the scheme's practicality and reliability in real-world applications.