UAV Trajectory Planning for AoI-Minimal Data Collection in UAV-Aided IoT Networks by Transformer

Abstract

Maintaining freshness of data collection in Internet-of-Things (IoT) networks\nhas attracted increasing attention. By taking into account age-of-information\n(AoI), we investigate the trajectory planning problem of an unmanned aerial\nvehicle (UAV) that is used to aid a cluster-based IoT network. An optimization\nproblem is formulated to minimize the total AoI of the collected data by the\nUAV from the ground IoT network. Since the total AoI of the IoT network depends\non the flight time of the UAV and the data collection time at hovering points,\nwe jointly optimize the selection of hovering points and the visiting order to\nthese points. We exploit the state-of-the-art transformer and the weighted A*,\nwhich is a path search algorithm, to design a machine learning algorithm to\nsolve the formulated problem. The whole UAV-IoT system is fed into the encoder\nnetwork of the proposed algorithm, and the algorithm's decoder network outputs\nthe visiting order to ground clusters. Then, the weighted A* is used to find\nthe hovering point for each cluster in the ground IoT network. Simulation\nresults show that the trained model by the proposed algorithm has a good\ngeneralization ability to generate solutions for IoT networks with different\nnumbers of ground clusters, without the need to retrain the model. Furthermore,\nresults show that our proposed algorithm can find better UAV trajectories with\nthe minimum total AoI when compared to other algorithms.\n