Environment-aware Path Planning for UAV-assisted Search and Rescue
Jianhui Chen, Shiqi Zeng, Xiaoli Xu · 2022
In this paper, we consider the UAV-assisted search and rescue (SAR) in a disaster area. The flying path for maximizing the expected number of detectable targets is designed under different amount of a-prior information on the environment. The target distribution is assumed to be modelled by the Neyman Scott Process (NSP). When the locations of target cluster centers are known, the probability map is constructed and the problem is formulated as a team orienteering problem (TOP) and solved by ant colony optimization (ACO) algorithm. When the locations of cluster centers are unknown, the correlation among targets can be utilized to optimize the UAV flying path in an online manner. The simulation results show that environment information is important for UAV path designing and the proposed online path planning by utilizing the spatial correlation among targets can greatly enhance the detection ratio, as compared with the random search.