UAV Control in Smart City Based on Space-Air-Ground Integrated Network

Abstract

Unmanned Aerial Vehicle (UAV) is an important part of the wireless network system of the future smart city. As a difficult point in the large-scale application of UAV, UAV control gradually attracts people's attention. Aiming at the problems of UAV control in smart city application, a near real time online learning architecture of UAV control based on the software-defined space-air-ground integrated network (SSAG) was proposed. This architecture uses the two-layer software defined network (SDN) controller architecture of SSAG framework to separate UAV control. The upper-tier SDN controller is responsible for the scheduling of UAV configuration, while the lower-tier SDN controller is responsible for regional coordination of UAV. The upper-tier SDN controller updates the tendency of network states by acquiring network states information in time interval. By simulating the network state in the next time interval, the optimal strategy of UAV scheduling of the next time interval is obtained by using the strategy iteration algorithm. Finally, an example is given to verify that the near real-time online learning architecture can accurately predict the UAV requirement, and increase the throughput of the network system compared with the traditional approach.