Intelligent UAV Charging Station Deployment and Path Planning in Smart City

With the proliferation of Internet of Things (IoT) devices, computation-intensive and delay-sensitive applications are prevalent in people's daily work and life. Unmanned aerial vehicles (UAVs) can compensate for the mobility and flexibility shortcomings of traditional base stations (BSs) and remote clouds, and provide services for the above applications in the case of limited or no coverage of infrastructure. However, the limited size and load capacity of UAVs determine that they only have limited battery capacity, which makes them unable to fly for long periods of time. The availability of UAV edge services and the UAV flight safety are difficult to guarantee. Motivated by this, it is significant to discuss the problem of UAV charging station (CS) deployment and path planning in smart city. In this paper, we first study the UAV CS deployment problem, which not only finds the actual locations of the UAV CSs but also minimizes the number of CSs. Then, we model the UAV path planning problem as a traveling salesman problem to determine the shortest path, and an improved ant colony optimization (ACO) algorithm is proposed as our solution. Finally, the experimental results verify that our scheme performs well in both timeliness and availability.