Path Re-planning method of unmanned underwater vehicles based on dynamic bayesian threat assessment

Due to numerous uncertainties in the environment, unmanned underwater vehicle (UUV) sometimes deviate from their originally planned paths. To address this issue, a path replanning algorithm based on threat assessment using a dynamic Bayesian network is proposed. This ensures that UUV can adjust their paths to avoid danger when facing uncertain events. Initially, the UUV plans a path using the PSO-SMPC (Particle Swarm Optimization-Stochastic Model Predictive Control) algorithm, utilizing environmental data. Subsequently, a dynamic Bayesian network evaluates the likelihood of uncertain events occurring based on environmental and UUV state information. The algorithm then determines the level of threat posed by these events and decides whether to activate the PSO-SMPC algorithm for path replanning accordingly. Simulation results demonstrate the effectiveness of this approach in enhancing UUV operational safety and improving mission completion rates across various uncertain event scenarios. Furthermore, compared to alternative methods such as simulated annealing and traditional genetic algorithms, the proposed algorithm exhibits superior path planning capabilities.