Abstract:

In view of the problem that the number of Pareto optimal solution sets in the unmanned aerial vehicle (UAV) multi objective path planning method increases with iteration, it is difficult to choose the cooperative path suitable for the task, a multi-UAV cooperative path planning based on improved multi-objective firefly algorithm (MOFA) evolution of interactive strategy is proposed. First, the variable decomposition strategy is used to decompose large scale variables in the firefly algorithm into multiple subpopulations to reduce algorithm search complexity. Then, the Tent chaos initialization strategy and multiple population cycle split merge strategy are used to improve the search performance of the algorithm. Bipolar preference dominance is used and designed the cooperation index to select the cooperative path suitable for the task in the Pareto optimal solution set. The simulation experiments show that the proposed algorithm can accurately find the optimal route planning scenarios that satisfies the focus and synergy according to the task setting, and demonstrate that the efficiently of the proposed algorithm.

Key words: UAV path planning, swarm cooperation, multi objective optimization, firefly algorithm, population diversity