Multi-AUV Dynamic Cooperative Path Planning with Hybrid Particle Swarm and Dynamic Window Algorithm in Three-Dimensional Terrain and Ocean Current Environment

Aiming at the cooperative path-planning problem of multiple autonomous underwater vehicles in underwater three-dimensional terrain and dynamic ocean current environments, a hybrid algorithm based on the Improved Multi-Objective Particle Swarm Optimization (IMOPSO) and Dynamic Window (DWA) is proposed. The traditional particle swarm optimization algorithm is prone to falling into local optimization in high-dimensional and complex marine environments. It is difficult to meet multiple constraint conditions, the particle distribution is uneven, and the adaptability to dynamic environments is poor. In response to these problems, a hybrid initialization method based on Chebyshev chaotic mapping, pre-iterative elimination, and boundary particle injection (CPB) is proposed, and the particle swarm optimization algorithm is improved by combining dynamic parameter adjustment and a hybrid perturbation mechanism. On this basis, the Dynamic Window Method (DWA) is introduced as the local path optimization module to achieve real-time avoidance of dynamic obstacles and rolling path correction, thereby constructing a globally and locally coupled hybrid path-planning framework. Finally, cubic spline interpolation is used to smooth the planned path. Considering factors such as path length, smoothness, deflection Angle, and ocean current kinetic energy loss, the dynamic penalty function is adopted to optimize the multi-AUV cooperative collision avoidance and terrain constraints. The simulation results show that the proposed algorithm can effectively plan the dynamic safe path planning of multiple AUVs. By comparing it with other algorithms, the efficiency and security of the proposed algorithm are verified, meeting the navigation requirements in the current environment. Experiments show that the IMOPSO–DWA hybrid algorithm reduces the path length by 15.5%, the threat penalty by 8.3%, and the total fitness by 3.2% compared with the traditional PSO algorithm.