Task Allocation Strategy For Platoons In Cooperative Driving

As a research subject with a long history,platoons are endowed with new research value by cooperative driving,namely,cooperative driving platoons.The leader in the cooperative driving platoon senses the traffic information ahead,makes driving decisions,and sends the instruction information to other vehicles of the platoon through relevant technologies of the Internet of Vehicles,so that the rear vehicles can reduce the perceived load and calculated load.However,this process will cause too much pressure on the resources of the first vehicle.Therefore,reasonable allocation of tasks for the cooperative driving platoon is a necessary measure to ensure the safe driving of the platoon.Also,mobile edge computing is an important development and construction direction of smart city and smart transportation.It is a significant research issue that how the cooperative driving platoon can be compatible with mobile edge computing equipment,and how to conduct more efficient task processing with the assistance of RSUs,so as to improve the stability of the platoon and the safety of autonomous driving.In view of the above problems and challenges,this thesis focuses on the task allocation of cooperative driving platoons in different scenarios.In order to effectively utilize the computational resources of the cooperative driving platoon,a task allocation strategy for the cooperative driving platoon is proposed in this thesis,which takes into account the characteristics of the autonomous driving task and the correlation between multiple cooperative vehicles in the platoon.The goal of the task allocation strategy for the cooperative driving platoon is to minimize the task processing delay.Firstly,the network model of the cooperative driving platoon is established based on the queuing theory,and the network delay analysis is carried out for the task processing process of the leader in the platoon.Secondly,the mathematical models of the cooperative driving platoon and the autonomous driving task are established,and the optimization problem of minimizing the task processing delay is proposed.Finally,based on the greedy algorithm,this thesis proposes a task allocation strategy for the cooperative driving platoon,which makes full use of the computing resources of different vehicles according to the heterogeneous characteristics of the system and alleviates the task processing pressure of the leader in the platoon.The simulation results show that the strategy proposed in this thesis can effectively improve the utilization rate of vehicle computing resources and reduce the processing delay of autonomous driving tasks.Furthermore,considering mobile edge computing scenario with roadside elements,this thesis extends the previous research content.In order to ensure the task delay and reduce the cost of task processing,a task allocation strategy for the cooperative driving platoon in the scene of moving edge computing was proposed.Firstly,the moving edge computing scenario is analyzed,and the system model and the structure of the platoon network based on software defined network are established.Secondly,in order to motivate self-driving cars and roadside edge mobile devices share the idle resources,this thesis established a resource sharing mechanism of the market to reward the computing resource server.In this mechanism,on the basis of detailed analysis of the response time and processing cost of tasks,this thesis proposes the optimization problem.In the end,this thesis designs a task allocation algorithm to solve the problem.Simulation experiments prove that the algorithm can realize the reasonable task allocation of the cooperative driving platoon in moving edge computing scenarios,and meet the task delay requirements.At the same time,the algorithm can also reduce the task execution cost.