Path Planning Method for Multi-AUVs Patrol in Restricted Multizone Area

Conventional particle swarm optimization(PSO) may generate infeasible path due to lack of constraints during the particle swarm updating process when it is applied to cooperative path planning for autonomous undersea vehicle(AUV) in complex environment. In this paper, a cooperative path planning optimization method for multi-AUVs in complex environments is proposed. This method combines the predictive control with the particle swarm optimization by embedding the two-step prediction into the particle updating process, and the new particles are tested to avoid formation of infeasible particles. Moreover, the environment is modeled based on the grid method, and the coverage information and uncertainty of the environment are stored in the vector grid. Simulation is conducted to verify the improved algorithm in island monitoring scenario. The results show that the proposed algorithm achieves multi-AUV patrol path planning for complex environment, and different patrol frequency for the regions with different nature to increase the patrol efficiency