UAV Localization in Non-Line-of-Sight Building Aera Using Genetic Algorithm

In modern urban scenarios, UAVs have become essential platforms for reconnaissance and surveillance due to their high mobility. However, building obstructions can cause UAVs to enter non-line-of-sight (NLOS) areas behind building corners, potentially endangering urban safety. The existing localization algorithms cannot accurately localize the target position, especially in the absence of building layout information. To address this issue, this paper proposes a ghost-matching localization method based on genetic algorithm, incorporating multipath propagation mechanisms such as diffraction and reflection of electromagnetic waves within buildings. The ghost positions of the target are first obtained using conventional Back Projection (BP) imaging. Subsequently, an objective function incorporating the unknown street width is established based on the geometric relationship between the target position and the ghost positions. This objective function is then optimized using genetic algorithm to estimate both the target location and street width. Simulation and experimental results demonstrate that the proposed method can accurately locate the target position without prior building layout information, validating the effectiveness of the algorithm.