Simulated rescue and path optimization of submersible based on MCMC and GA algorithms

This study proposes an integrated approach for predicting the location of a missing submersible and optimizing the Search and Rescue (SAR) process. The method combines the Markov Chain Monte Carlo (MCMC) technique with the Metropolis-Hastings algorithm to iteratively generate the precise distribution of the submersible's potential locations. To enhance the efficiency of the SAR operations in real-world scenarios, a dynamic spiral search pattern is employed to optimize the search path, significantly improving detection effectiveness. In addition to the spiral search optimization, the study incorporates the Genetic Algorithm (GA) to further streamline the search process, reducing both the time and distance required for a successful rescue. Extensive simulation experiments validate the proposed methods, demonstrating their high success rate across diverse underwater environments. To address the challenges posed by multi-submersible rescue operations, a coordinated search strategy is implemented, which not only enhances overall efficiency but also ensures safety during the rescue. The results underscore the effectiveness of the integrated approach, offering a robust framework for advancing submarine rescue techniques in the future.