Joint UAV Trajectory Optimisation and QoS Consideration for IoT Device Data Collection

In this paper, we examine a scenario where a single unmanned aerial vehicle (UAV) is used to perform data collection on multiple IoT devices. Our goal is to minimise the energy consumption and service time of the UAV by planning its trajectory, while we focus on the quality of service(QoS) of the IoT devices. To achieve this, we begin by selecting hover points for the UAV and then the UAV visits all hover points in a certain order for data collection. Specifically, we employ a K-means clustering algorithm to cluster the IoT devices and utilize the resulting cluster centers as hover points. We then transform the problem of UAV visits to hover points into a travelling salesman problem (TSP) and design a genetic algorithm (GA)-based path selection algorithm to solve the problem. Finally, We display the results of our experiments, compared to random and greedy algorithm, our algorithm can reduce energy consumption and drone service time by approximately 18% and 7% respectively.