In this paper, a game-theoretical autonomous decision-making approach for efficient deployment of unmanned aerial vehicles (UAVs) in a multi-level and multi-dimensional assisted network is analyzed. The UAVs have directional antennas that work as wireless stations, which provide the best coverage for multiple ground mobile/fixed users. In general, UAVs work in a cooperative manner for achieving the suitable deployment with the optimal coverage values for the candidate region. In this paper, a game theory concept is used and the payoff function for each UAV is defined based on the coverage probability value, which depends on the altitude and the characteristic of antennas in the UAVs. We introduce a mathematical formulation for evaluating the payoff values based on a set of actions for each UAV, and the Nash equilibrium for this kind of game. This approach works in an intelligent way based on the interactions between the UAVs and their neighbors in a connected network and it might work even in harsh environments. In order to minimize interference, the UAVs’ altitudes are adjusted based on the antennas and other deployment requirements (i.e. search and surveillance purposes) by using the minimum number of UAVs to cover the candidate geographical region. The simulation results show that the proposed approach achieves the maximum coverage value, converges fast with the environmental changes based on the power levels, and robust for failure scenarios. Finally, we compare our approach against one of the traditional approaches called Collaborative Visual Area Coverage Approach (CVACA) based on uniform coverage quality. The simulation results show that the game approach outperforms the traditional approach in term of the coverage value and the computational time.

本文分析了一种基于博弈论的自主决策方法，用于在多层，多维辅助网络中有效部署无人机。UAV具有用作无线站的定向天线，可为多个地面移动/固定用户提供最佳覆盖。通常，无人机以协作的方式工作以实现具有候选区域的最佳覆盖值的适当部署。在本文中，使用了博弈论的概念，并根据覆盖概率值定义了每个无人机的收益函数，该概率取决于无人机的天线高度和特性。我们介绍了一种数学公式，用于基于每个无人机的一组动作以及这种游戏的纳什均衡来评估回报值。这种方法基于UAV及其相邻网络之间的交互以一种智能的方式工作，即使在恶劣的环境下也可能有效。为了最小化干扰，通过使用最少数量的无人机来覆盖候选地理区域，根据天线和其他部署要求（即搜索和监视目的）来调整无人机的高度。仿真结果表明，该方法可以实现最大覆盖值，并且可以根据功率水平快速适应环境变化，并且对于故障场景具有鲁棒性。最后，我们将我们的方法与基于统一覆盖质量的传统方法之一称为协作视觉区域覆盖方法（CVACA）进行了比较。