Multi-uav multi-target trajectory planning based on particle swarm optimization algorithm

Aiming at the problems of multi-UAV multi-target track path planning, such as local optimization, collision between aircrafts and low efficiency, etc. In this paper, a multi-UAV multi-objective improved particle swarm optimization (Mumoipso) algorithm is proposed, which combines the improved particle swarm optimization algorithm with Dubins algorithm. Firstly, the speed and position updating methods in particle swarm optimization are improved by target replacement and particle crossover. By replacing the target whose position changes due to its own speed, and crossing the particles whose position changes are affected by individual extremum and global extremum, the improved particle swarm optimization algorithm is suitable for multi-UAV multi-target track path planning. Secondly, the arctangent function is used to improve the inertia factor, and the linear decreasing function is used to improve the non-negative acceleration coefficient, so as to improve the UAV's global search ability in the early stage and the UAV's local search ability in the later stage to avoid falling into the local optimum. Finally, Dubins algorithm combined with Intersection Type method is used to plan a collision-free smooth path. The simulation results show that the proposed algorithm has better search effect and path planning mode on the premise of ensuring good stability, and compared with other algorithms, its fitness function and total voyage are improved by 16.3% and 10.2% respectively.