MATOPO:A multi-UAV assisted task offloading and path optimization method for moving edge computing
Aiming at solving the task offloading and path planning challenge of multi-UAV assisted mobile edge computing,a multi-agent deep reinforcement learning method for task offloading and path optimization is proposed to reduce the total energy consumption of the system and improve computing performance.Firstly,the model of multi-UAV assisted mobile edge computing system is designed,and the UAV network is centrally managed by software-defined network technology.Then,on the basis of considering the load of the UAV and the fairness of the associated service of the user equipment,taking the total energy consumption of the system as the optimization goal,the multi-agent depth deterministic strategy gradient algorithm is designed to complete the task unloading and the path management optimization of the UAV,so as to achieve load balancing and reduce the total energy consumption of the whole system.Simulation results show that compared with other benchmark algorithms,the proposed method can reduce system energy consumption and computing delay to a certain extent,ensure the efficiency,stability and reliability of the whole system,and better meet the service requests of mobile edge users on the basis of making full use of the computing resources of UAV-assisted mobile edge computing systems.
mobile edge computingmulti-UAV networktask offloadingpath optimizationmulti-agent deep reinforcement learning
万方数据
维普
