Multi-UAV Efficient Cooperative Forest Fire Patrol in 3-D Mountain Environments
Menghua Zhang, Honglun Wang, Yuebin Lun, Zhiyuan Yang, Yanxiang Wang · IEEE Transactions on Aerospace and Electronic Systems · 2025
To have efficient and comprehensive forest fire patrol in mountain environments, a multi-unmanned aerial vehicle (UAV) three-dimensional (3-D) cooperative forest fire patrol planning framework based on dynamic patrol information maps and terrain elevation is proposed in this paper. First, the forest fire patrol problem is formulated, and the patrol information maps are modeled according to the key patrol clue information. Second, to make the patrol information maps accurately reflect the real fire situation and improve patrol efficiency, a Bayesian inference framework and a particle filter (PF) are combined to estimate the posterior probability distribution of the forest fire source term (FFST), which includes the ignition point position and the fire spread time. Then, a Gaussian mixture model (GMM) is used to extract the potential FFST, thereby forest fire situation is estimated. On this basis, the dynamic updating mechanism of information maps is presented. Third, a 3-D cooperative forest fire patrol planning framework that allows for UAVs to follow terrain and avoid collisions is designed, which guarantees that the planned paths cover typical 3-D landforms such as peaks and hills and flight safety. Finally, comparative simulations and hardware-in-the-loop (HIL) experiments validate the effectiveness of the proposed method.