基于人工蜂群优化的无人机协同MEC网络中卸载算法OA

Mobile Edge Computing(MEC)technology is increasingly employed in scenarios demanding stringent low latency and resource stability,such as disaster rescue and forest fire warning.However,the scarcity of ground infrastructure often restricts its effec-tiveness.Unmanned Aerial Vehicles(UAV),leveraging their deployment flexibility and high mobility,serve as an ideal platform to address this challenge.This study innovatively proposes a Balanced Multi-UAV Coverage Path Planning(BmUCPP)method,combining the Spanning Tree Coverage(STC)algorithm with the Minimum Spanning Tree(MST)algorithm,with a primary focus on resolving the load imbalance problem in multi-UAV cooperative operations.To tackle the multi-objective optimization challenges inherent in edge computing models,an Improved Artificial Bee Colony(IABC)-Genetic Algorithm(GA)—IABC-GA is developed.This enhanced algo-rithm efficiently minimizes key objectives while ensuring MEC service quality.Evaluation results demonstrate that the IABC-GA exhib-its distinct advantages in optimization capability,convergence speed,and stability.Aiming to meet practical requirements of field or disaster-stricken environments,a dynamic UAV-assisted MEC model is established.This model comprehensively considers UAV com-munication,computing,and endurance constraints,environmental communication quality,and capabilities of ground User Equipment(UE).The core objective is to minimize the average weighted energy efficiency for both UE and UAVs(integrating energy consumption and latency).By deeply integrating the proposed BmUCPP path planning algorithm with task scheduling algorithms,multi-dimensional simulations are conducted.The results substantiate that this collaborative strategy effectively reduces the overall cost of dynamic UAV-assisted edge offloading services.