关键词: 舰载无人机, 双种群优化算法, 任务规划

Abstract: Mission planning of multiple shipborne UAVs for coordinated attacks on maritime targets is one of the key technologies to effectively improve the maritime combat capability of UAVs. Different from land-based UAVs, the take-off and landing platform of shipborne UAVs is a mobile platform at sea, and there are more options for take-off and landing points, which increases the complexity of the solution. Furthermore, shipborne UAVs work in a complex marine environment. During the navigation of the ship at sea, due to the different marine environments of the sea, excessive waves will affect the state of the ship and pose a safety threat to the departure and landing of the UAV. Therefore, the study of shipboard UAV mission planning is of important military significance. Based on the principle of multi-objective optimization, this paper focuses on the UAV mission planning modeling and solution problem with the background of a shipborne UAV coordinated attack mission against targets at sea.
In mission planning modeling, a multiple takeoff and landing point is considered for the feature that the takeoff and landing platform of the shipborne UAV is a mobile platform. The time window constraint is added to each takeoff and landing point for the feature of the complex marine working environment of the shipborne UAV. In addition, the shipborne UAV mission planning model is established with the goal of maximum attack revenue and minimum UAV damage by combining various factors such as the air time required for the UAV to complete the mission and the survival threat.
In the aspect of the solution, due to the strong constraints of this model, the feasible solutions are likely to evolve into infeasible solutions in the process of solving the traditional algorithm, which easily makes the algorithm fall into local optimum and makes it difficult to find the solution set of Pareto front. This paper adopts a dual population optimization algorithm, in which feasible and infeasible solutions are not directly compared while evolving in parallel, so that the algorithm avoids falling into local optimum, the reasonableness of the algorithm is verified by giving different hypothetical parameters, and the simulation results show the feasibility and effectiveness of the method by comparing it with the classical multi-objective optimization algorithms such as NSGA-2 and SPEA-2.

Key words: shipborne UAV, dual population optimization algorithm, mission planning