A multi-submersible search and path optimization strategy based on intelligent algorithm

This paper presents a comprehensive study on optimizing submersible search tasks in complex underwater environments by combining advanced algorithms such as Kalman Filter, Simulated Annealing, Dijkstra's Algorithm, and Bayesian Search Theory. The fusion of these algorithms enhances the accuracy of prediction, optimizes the search path, and can dynamically adjust the search strategy in real time. The Kalman Filter improves prediction accuracy by mitigating noise in underwater environments, while Simulated Annealing helps avoid local minima in path optimization. Dijkstra’s Algorithm ensures efficient pathfinding in weighted graphs, and Bayesian Search Theory provides dynamic updates to adjust search probabilities based on new observations. Experiments conducted in simulated environments demonstrate the effectiveness of these algorithms in improving submersible search efficiency and avoiding obstacles. The results indicate the potential of this multi-algorithm approach to expand into areas such as environmental monitoring, autonomous navigation, and other complex decision-making systems. Future work will focus on testing in real-world environments and further integrating machine learning techniques to enhance real-time search efficiency.