An autonomous collision avoidance system unified with the TFMS and the FCS-MPC strategy for USV

In order to realize the autonomous collision avoidance of unmanned surface vehicle(USV), a collision avoidance system(CAS) for USV based on the Tunable Fast Marching Square method(TFMS) and the Finite Control Set-Model Predictive Control (FCS-MPC) strategy is proposed. First, the navigation simulation model of the USV is established. Then, the TFMS is adopted to get the optimal path from start point to destination point in given environment. FCS-MPC is used under the constraints of local environment and dynamic model of USV to ensure collision avoidance during navigation. Finally, the track point in the global planned path is served as the local target points of collision avoidance strategy, so the USV can arrive at the final target points safely and independently. Simulation experiments are carried out with real coastal map as environmental constraints to validate the effectiveness of proposed CAS.