Multi-USV cooperative path planning via FSM-based distance field and enhanced hippopotamus optimization

Abstract

With single Unmanned Surface Vehicle (USV) no longer meeting growing mission demands, cooperative multi-USV systems have become essential, particularly in obstacle-rich waters where task allocation deviates and path planning struggles to balance distance and safety. To address these challenges, an integrated framework combining a constrained-distance-based Hungarian assignment algorithm and an improved Hippopotamus Optimization algorithm (CDH-IHO) is developed to achieve simultaneous target assignment and path planning (STAPP). The CDH module achieves globally optimal assignment by exploiting a constrained distance matrix constructed through the Fast Sweeping Method (FSM), while the IHO module introduces a Cubic chaotic map and a mutation operator to enhance convergence and avoid local optima. Distance, turning angle, safety, and penalty terms are jointly considered for collision-free path generation. Simulations in five scenarios verify global optimality in assignment and superior performance in path length, smoothness, and safety. CDH-IHO provides an efficient and robust solution for STAPP.