Autonomous Obstacle Avoidance Algorithm for Unmanned Surface Vehicles Based on an Improved Velocity Obstacle Method

Focusing on the collision avoidance problem for Unmanned Surface Vehicles (USVs) in scenario of multi-vessel encounters, a USV autonomous obstacle algorithm based improved velocity method is proposed. The composed two parts: encounter detection model and path re-planning algorithm. draws idea through integration characteristics such as dynamic marine environment, encountering vessel motion model, International Regulations Preventing Collisions at Sea (COLREGS) to obtain region encounters. On this basis, constraint conditions state space behavior are added window complete risk assessment generate strategy set. premise minimum resource cost uses an particle swarm optimal control set re-planning. Simulation results show that can enable USVs safely evade multiple short-range targets under COLREGS.