| |
| UAV technology has emerged as a promising solution, introducing a new approach to smart grid inspection. This paper explores the safe and efficient task offloading of UAV cruise systems within smart grids. Specifically, it proposes a multi-UAV to multi-vehicle fog computing node task offloading framework named MU2MV. Firstly, a novel reputation incentive mechanism based on the behavior of vehicle fog computing nodes is introduced. This mechanism provides incentives or penalties based on the behaviors of these nodes to select suitable target fog computing nodes and encourage newly joined nodes to actively provide computing services as service providers. Secondly, a two-tier collaborative non-cooperative game model is proposed to simulate the cooperative-competitive relationships among UAVs and vehicle fog computing nodes, as well as among the vehicle fog computing nodes themselves, the optimal strategy combinations are determined by solving the Nash equilibrium solution within the game model. Finally, recognizing the challenges posed by the dynamic movement of UAVs and vehicle fog computing nodes in real environments, a deep reinforcement learning algorithm is proposed to seek the optimal strategy combination for UAVs and vehicle fog computing nodes. |
|
| Keywords:Computer application technology; Cyber physical system; UAVs; vehicle fog computing nodes;task offloading |
|
