Abstract: In view of the limitation that only one side of UAVs and targets has long-distance detection capability in past studies, it can’t reflect the two-sides game relation of the battlefield environment. In response to this deficiency, this paper proposes a multi-step game strategy based on the stackelberg equilibrium and multi-step prediction, which realizes the game search with the long-distance detection capability. The path profit function is created, both sides can select the corresponding function according to the game state at different moments, which realizes the dynamic path planning of the UAVs. The simulation results show that the proposed strategy is fully applicable to the game model of this paper and has higher search efficiency than the greedy strategy, greatly improving target capture rate.

Key words: unmannned aerial vehicle, target search, games, dynamic path planning

摘要： 考虑到现有无人机搜索问题研究中无人机、移动目标仅有一方具有远距离探测能力的设定，已经无法体现出战场环境下双方的博弈关系。针对这一不足，基于stackelberg均衡策略，结合多步预测的思想，提出了stackelberg多步博弈策略，实现了无人机、目标都具有远距离探测能力的博弈搜索。通过建立无人机、目标各自的路径收益函数，使双方能够根据不同时刻的博弈状态选择相对应的函数，实现无人机的动态路径规划。仿真结果表明所提出策略完全适用于该博弈模型，比贪婪策略具有更高的搜索效率，大大提高了目标捕获率。

关键词: 无人机, 目标搜索, 博弈, 动态路径规划