Multi-UAV Task Assignment Based on Improved VPPSO

Task assignment is a critical factor for multi-UAVs to efficiently accomplish missions. In this paper, two task assignment models are established under different optimization objectives. Based on the particle swarm optimization (PSO) algorithm, a particle encoding scheme that satisfies model constraints is designed. To address the issue of PSO easily falling into local optima, improvements are made by revising the particle velocity update strategy, introducing chaotic initialization for particle positions, and incorporating hill-climbing mutation operations to enhance the algorithm’s search capability. An improved velocity pausing particle swarm optimization algorithm (VPPSO) is proposed to solve multi-UAV task assignment problems. Simulation validations using two case studies demonstrate that the algorithm achieves accurate and stable solutions, with faster convergence and stronger optimization capabilities compared to the basic PSO algorithm, indicating better practicality.