Joint Trajectory Planning and Resource Allocation for UAV-assisted Information Collection

Recently, unmanned aerial vehicles have been considered to implement information collection tasks by taking photos for points of interest in many scenarios like disaster relief. Most existing related works focused on maximizing the number of points of interest visited under the constraint of limited energy capacity of the Unmanned aerial vehicles (UAVs). In this paper, we consider the situation in which a UAV is deployed to collect information in a target area with limited energy capacity and storage capacity. Moreover, we consider the UAV can not only process the data collected with limited computation capacity but also send it back to the ground terminal for processing. To maximize the number of points of interest visited, we jointly optimize the trajectory and resource allocation of the UAV. Due to the NP-hardness of the formulated optimization problem, we propose a deep reinforcement learning based method to implement the decision-making. Experimental results show that our proposed algorithm is efficient in information collection tasks.