Age of Information Aware Trajectory Planning of UAV

This paper investigates the planning of Unmanned aerial vehicles (UAVs) trajectory in UAV-assisted Internet of Things (IoT) networks with a massive number of IoT devices (IoTDs). Existing UAV-assisted IoT network data collection schemes mostly focus on optimizing energy consumption and data collection throughput, while neglecting the temporal value of data collection. With the assistance of the age of information (AoI), the average AoI of data collected by the UAV from IoTDs is minimized to enhance information freshness. To strike a balance between trajectory planning and information freshness, a two-stage artificial intelligence (AI) algorithm is proposed in this paper. Firstly, to tackle the issue of prolonged flight time caused by the UAV sequentially collecting data from IoTDs, an improved clustering algorithm is introduced to determine the cluster centers of IoTDs. Secondly, considering that the UAV lacks prior knowledge of the IoT network environment, the AoI minimization problem is reformulated as a Markov decision process (MDP). A neural network algorithm based on twin-delayed deep deterministic policy gradient (TD3) is employed to optimize UAV trajectory. Simulation results show that the proposed algorithm is superior to the benchmark algorithms, particularly in scenarios involving a massive number of IoTDs.