Dynamic environment UAV deployment algorithm based on potential game theory

Abstract

To address the issue of low coverage resulting from the challenge of acquiring the optimal deployment position in commonly used distributed deployment algorithms, this study presents a three-dimensional deployment algorithm for Unmanned Aerial Vehicles (UAVs) based on potential games. First, a local mutually beneficial game model is designed to demonstrate the existence of exact potential games and Nash equilibrium. The Nash equilibrium solution corresponds to the maximum coverage. Next, drawing inspiration from exploration, a solution method called Exploration Spatial Adaptive Play is proposed. It utilizes the maximum utility function value from multiple step sizes in the exploration direction to update the action selection probability, thereby ensuring the optimal deployment position in each decision cycle. To address the issue of sensor position error, a method for processing sensor position errors is proposed. The simulation results demonstrate that the proposed distributed deployment algorithm achieves higher coverage compared to commonly used methods.