Multi-UAV cooperative reconnaissance planning problem refers to the problem that multiple UAVs reconnaissance multi-targets to solve the problem of minimizing the number of UAVs and the shortest total driving path. The bat algorithm is an emerging swarm intelligence algorithm with broad prospects and significant effects in many fields. In this paper, for the multi-UAV cooperative reconnaissance planning problem, the scene is coded, and the bat algorithm is used to solve the problem, and the obtained results are compared with the results obtained by genetic algorithm and particle swarm algorithm. The experimental results show that the bat algorithm has better stability and faster processing speed than the genetic algorithm and particle swarm optimization algorithm when solving the multi-UAV cooperative reconnaissance mission planning problem.