Study on Propeller Positioning and Trajectory Control of Underwater Robots Based on Bayesian Inference and Multi-Objective Optimization

Abstract

In ocean exploration, the positioning and trajectory optimization of underwater vehicles are critical research areas that directly impact the success of ocean resource exploration, environmental monitoring, and search and rescue missions. However, the complex and uncertain marine environment poses significant challenges. This study first determines the three-dimensional spatial position of the underwater vehicle, calculates the vehicle's speed in the marine environment through mechanistic analysis, traversing the initial position, velocity, acceleration, angular velocity and angular acceleration of the submersible, fits the motion trajectory using a random walk, and uses Bayesian inference to estimate maximum probability positions of submersibles and select the optimal initial deployment point. To better cope with real-world scenarios involving ocean currents and the diverse motion modes of underwater vehicles, this study further used an improved genetic algorithm to divide the motion trajectory of the submersible into a series of discrete path points, and constructed a multi-objective optimization model with the shortest search and rescue time and the maximum probability of successful search and rescue as the objective function to realize the precise positioning of the lost underwater vehicle.