A Potential Game Approach to Multi-UAV Accurate Coverage Based on Deterministic Radio Wave Propagation Model in Urban Area

Abstract

Existing methods based on statistical models ignore the detail effect of the real environment on path loss, leading to some discrepancies with the actual results in real urban environments. In order to reduce the discrepancies and improve the unmanned aerial vehicle (UAV) coverage effect, a multi-UAV coverage method based on deterministic radio wave propagation is proposed. First, the ray-tracing method, as one of the deterministic models, is used to simulate the path loss. Next, the cooperative search and coverage problem of multiple unmanned aerial vehicles is modeled as a potential game model, and the utility function is designed by path loss. After that action set is designed to constrain the motion of unmanned aerial vehicles. Then, the coverage problem is solved using a binary log-linear learning algorithm, which combines ray-tracing method and deterministic model. Finally, experiments are conducted in local areas of two cities, Haikou and Guangzhou by proposed method and statistic models. The experimental results show that the proposed method is capable of multi-UAV cooperative search and coverage tasks in urban areas and can achieve better performance than the statistical model-based methods.

Keywords

• Multi-UAV cooperative coverage
• deterministic radio wave propagation model
• potential game
• binary log-linear learning algorithm
• urban areas