Improved Multi-Objective Particle Swarm Optimization Algorithm Based on Area Division With Application in Multi-UAV Task Assignment

This paper concerns the multi-UAV task assignment problem, which is solved by a multi-objective particle swarm optimization algorithm for adaptive region partitioning. Since the traditional multi-objective optimization algorithms tend to fall into local optimum solutions when dealing with optimization problems, this paper establishes an improved multi-objective particle swarm optimization (MOPSO) algorithm based on the adaptive angle area division. This paper proposes a new multi-UAV task assignment model where the threat constraint is concerned. To solve this model, the algorithm first preprocesses solution spatial information, including normalization of solutions and area division of space. Further, global optimal particle selection strategy is improved based on angle of division. In order to improve the global searching ability, some infeasible solution is used. Finally in the implementation stage of the algorithm, we set multiple nodes for the trajectory of the UAVs to increase the stability of the algorithm. The simulation experiments results demonstrate that the improved algorithm can provide a flyable solution for the UAVs and achieve better convergence and diversity.