Multi-UAV Task Allocation Based on Improved Algorithm of Multi-objective Particle Swarm Optimization

Abstract

With the development of the technology of unmanned aerial vehicle (UAV), the multi-UAV task allocation has become a hot topic in recent years. Recently, many classical intelligent optimization algorithms have been applied to this problem, because the multi-UAV task allocation problem can be formalized as a NP-hard issue. However, most research treat this problem as a single objective optimization problem. In view of this situation, we use an improved algorithm of multi-objective particle swarm optimization (MOPSO) to solve the task allocation problem of multiple UAVs. We will take two stages of SMC resampling to improve the disadvantages in the MOPSO algorithm. In the first stage, resampling is used to improve the slow convergence of the particle swarm optimization in the middle and late stages. In the second stage, resampling is used to expand the search area of the particle swarm optimization algorithm and to prevent the algorithm from falling into the local optimal solution. The simulation results show that the improved algorithm has a good performance in solving the task allocation problem of multiple UAVs.