MTSP-NSGA-III based path planning for underwater glider swarm executing multi-point exploration missions

The underwater glider swarm can efficiently and reliably execute water surface multi-point exploration missions through collaboration among individual gliders. To further improve the efficiency of a glider swarm in executing multi-point exploration missions under complex ocean currents, it is necessary to optimize the detection sequence of target exploration points, which is clearly a path planning problem. To address this issue, this paper establishes performance evaluation models for glider based on kinematic relationship and force analysis, enabling the quantitative analysis of energy consumption, motion accuracy and voyage time. On this basis, a point-to-point path planning method for glider is developed by combining the depth-averaged current velocity prediction method and the principle of motion synthesis. Furthermore, combining the point-to-point path planning method with the solution approach for the multiple traveling salesman problem (MTSP), a mathematical model for multi-point exploration path planning of glider swarm is proposed, aiming to improve the motion accuracy and reduce energy consumption and voyage time of the gliders. Based on the non-dominated sorting genetic algorithm III (NSGA-III), the solution method of the mathematical model is proposed. Finally, the effectiveness of the proposed method is validated by several numerical cases considering different ocean currents, summarizing some conclusions. The research work aims to provide theoretical guidance for the practical application of the underwater glider swarm.