Research On Computation And Task Scheduling Strategy Based On Platoon In Internet Of Vehicles

With the progress of wireless technology and Internet of Things,the rapid development of intelligent vehicles has also brought about a substantial increase in the demand for on-board applications.The emergence of some new computation-intensive and delay-sensitive IoV services poses new challenges to computing and task scheduling in Internet of Vehicle(IOV).Vehicular Edge Computing(VEC)will be an effective solution to meet both the scaling needs of vehicle computing resources and rapid interaction during task offloading by sinking cloud services to the edge of the mobile network.However,the computing power of a single vehicle is limited,and vehicles have high mobility in the IoV In this regard,platoons,formed by smart vehicles driving in the same patterns,can combine the resources of vehicles and maintain the relative stability between vehicles,which will effectively alleviate these problems and provide a promising paradigm to execute computation tasks with on-board computing resources.This paper analyzes the background and significance of IoV and edge computing which based on platoon,introduces the research status at home and abroad.On the basis of exploring the related theories of task scheduling,resource allocation and platoon,the computation and task scheduling strategy based on the characteristics of the platoon is studied.The specific work of this paper is as follows:(1)Propose a vehicular edge computing mechanism based on the stable platoon.This mechanism exploits computation capabilities of both platoons and edge computing enabled RSUs(Roadside Units),solving the vehicle edge computing problem in the process of vehicle travel in the context of single vehicle with limited resources,uneven road resources and high-speed vehicle movement,,with the main goal of minimizing the delay cost and energy consumption of the system task execution.In addition to factors affecting the actual environment,the model also takes into account conditions such as communication resource constraints.In this paper,a learning algorithm based on Deep Deterministic Policy Gradient(DDPG)is designed.The algorithm can effectively determine the target offload server and obtain the optimal resource scheduling strategy for computation and communication.Numerical results show that the proposed algorithm significantly reduces the delay and energy consumption compared with the benchmark algorithm.(2)Propose a vehicle edge computing mechanism based on the dynamic platoon.Based on the existing vehicle edge computing,an improvement is made on the stable platoon,and the problem of vehicle edge computing under dynamic formation of platoons is further considered.The model still takes into account the time variation of resources,vehicle dynamics and platoon characteristics in the real environment,and the main goal is to maximize the utility of vehicles.Firstly,a contract-based incentive mechanism was designed to motivate the vehicles to join the platoon.After the formation of the platoon,the task scheduling and resource allocation scheme is designed,The model takes the RSU as the server,the platoon as the server and the task relay node to the RSU.Finally,the convex optimization algorithm are combined to design the simulation according to the optimization problem.Compared with other schemes,the designed contract can motivate the participation of vehicles,and the proposed task offloading and resource allocation scheme can improve the utility of vehicles.