Research On Dual-objective Genetic Algorithm For UAV Elastic Scheduling Based On Edge Computing

In the past decade,mobile edge computing has received extensive attention and research.It shifts computation-intensive tasks from SDs to a nearby MEC facility for flexible and efficient computing services.Typically,MEC devices are static at the edge of a wireless network,meaning that once they are deployed,they are rarely moved.The layout of MEC equipment in a given area depends on the distribution of workload in the area,so unmanned aerial vehicle(UAV)-based MEC systems have attracted a lot of attention.In this paper,the UAV elastic scheduling problem based on edge computing is studied,various scheduling methods are investigated in detail,and the advantages and disadvantages of existing scheduling schemes are analyzed.For the UAV elastic scheduling problem based on edge computing,the following two aspects are the focus of this paper.In this thesis,the UAV elastic scheduling problem based on edge computing is studied,the idea of scheduling algorithm is summarized,and the advantages and disadvantages of existing scheduling schemes are analyzed.For the UAV elastic scheduling problem based on edge computing,the following two aspects are the focus of this thesis:1.Task scheduling problem with different deadlines under limited number of UAVs.In this problem,the UAV flies from the base station to the node and hovers over the node to help calculate.After the calculation,it goes to the next node and returns to the base station after all tasks are completed.Due to the limited number of drones,it is impossible to meet all the requirements of the mission,so it can only accomplish the goal as much as possible,convert the overtime into punishment,and try to minimize the completion time of the mission and the total cost of the UAV.2.When the task must be completed,the UAV flies to the node from the base station to receive the task,and can help calculate above the node or when flying to other nodes and return the calculation result within the deadline.In order to minimize the cost and the shortest completion time,the task scheduling problem in UAV-based MEC system was studied,and the concept of UAV leasing was proposed to solve the problem of UAV resource shortage.To solve the problem of insufficient computing power,a new algorithm framework based on NSGA-II was proposed.