关键词: 无人机集群, 多智能体建模, Carlton-Cellular, 体系效能评估, 森林消防

Abstract: Unmanned aerial vehicle (UAV) swarm represent a new type of firefighting force for forest fire fighting mission. The system-of-systems (SoS) effectiveness evaluation of UAV swarm helps optimize the design of UAVs and enhance their capabilities in performing forest fire fighting mission. To improve the interpretability of simulation evaluation results, this paper establishes a multi-level decoupling framework oriented towards the SoS based on DoDAF (Department of Defense Architecture Framework) and the combat cycle theory. Considering the combustion and dynamic characteristics of forest fire, this paper first constructs a cellular model of forest fire combustion. Then, based on this combustion model, a Carlton-Cellular firefighting effectiveness calculation model that reflects the spreading characteristics of fire extinguishing bomb is proposed. Finally, this paper applies this calculation model to a multi-agent simulation platform centered on the multi-level decoupling framework to solve the SoS effectiveness of UAV swarm in forest fire fighting mission. Experimental results show that basic performance indicators such as UAV speed, bomb load, and effective coverage radius of fire extinguishing bomb can effectively enhance SoS effectiveness. When the UAV speed increases from 60m/s to 85m/s, the firefighting effectiveness improves by 0.46. Increasing the bomb load from 2 to 3 increases the success rate of the UAV swarm's firefighting mission by 70%. An increase of 3m in the effective coverage radius improves the firefighting effect of the swarm by 0.3. This research can provide a reference for the model design of firefighting UAVs.

Key words: Unmanned Aerial Vehicle swarm, multi-agent modeling, Carlton-Cellular, effectiveness evaluation, forest fire fighting