Energy Minimization for Cellular-Connected UAV-MEC Patrol Inspection Systems

Abstract

In this paper, we consider a cellular-connected unmanned aerial vehicle (UAV)-based mobile-edge computing (MEC) in patrol inspection scenarios, where the UAV needs to traverse multiple predetermined waypoints for patrol inspection and then offloads the part of the computation task to the ground base stations (GBSs) for relieving its workload, during which the related energy cost is a significant issue due to the long flight period and the large amount of data to be processed. This paper aims to minimize the total energy consumption by jointly optimizing the task completion time, communication scheduling, computation resource allocation, and UAV's trajectory. Firstly, we apply the path discretization method and alternating optimization method to decompose the original problem into two tractable subproblems: 1) optimal transmission strategy between two consecutive waypoints; and 2) the association sequence among waypoints. Then, by involving the energy efficiency and location relationship among GBSs and waypoints, a hybrid minimum ratio traveling salesman problem (MRTSP) is proposed to design the UAV's initial trajectory. Finally, the successive convex approximation (SCA) technique and block coordinate descent (BCD) scheme are adopted to optimize the UAV's trajectory and wireless resource allocation. The simulation results demonstrate that our proposed schemes can significantly reduce energy consumption.