Bayesian Network-Based Threat Assessment and TARAPSO Algorithm for UUV Path Planning in Complex Environments

Path planning is crucial for unmanned underwater vehicles (UUVs) navigating in complex marine environments, which are often affected by multiple threat factors. Traditional path planning methods often fail to account for these factors, suffer from local optimal solutions, and lack adequate exploration and exploitation capabilities. To overcome these limitations, this paper proposes a target attraction rule-based adaptive particle swarm optimization (TARAPSO) algorithm, driven by threat assessment. Firstly, a 3D model of the complex marine environment is established, incorporating multiple threat factors. Secondly, a Bayesian network (BN)-based threat assessment model is developed to perform probabilistic inference on the states of potential threats encountered by the UUV. Based on the results of this assessment, a cost function incorporating the threat degree index is designed to evaluate the impact of various threat factors on the UUV’s path. Finally, the TARAPSO algorithm is introduced and applied to solve the multi-objective optimization problem of path planning, incorporating the threat assessment-driven cost function. Simulation results show that the TARAPSO algorithm, when combined with the BN-based threat assessment strategy, demonstrates superior efficiency in evaluating multiple threat factors, generates competitive paths with shorter lengths, and effectively ensures safe navigation of the UUV.