Design And Implementation Of Cooperative Sensing And Control Through Fog Computing

With the development of vehicular technologies,more and more vehicles have accesses to networks,and more and more services become dependent on networks.This puts a large burden on the network infrastructure.Meanwhile,the pressure of the cloud data center is getting larger and larger,which leads to low computing efficiency and long delay for services.At the same time,the unmanned driving,a hot research topic currently,dependents mainly on the sensors of the vehicles,making it difficult for vehicles to handle the complicated situation on the road.Fog computing,a new distributed computing paradigm close to the edge of netwoks,migrates the computing from the cloud to the edge of the network.It can achieve cooperation between vehicles to help implementing the computation locally,relieving the computing pressure of the cloud data center and reducing the latency of services.Thus,the introduction of fog computing to unmanned vehicular environment will help solving the complicated situation on the road,expend the scope of sensing and meet the low latency requirement of services.Based on the deep analysis of the problems encountered in vehicular networks,the thesis combined the idea of fog computing and clustering technology,and designed the cooperative sensing and cooperation control mechanism based on fog computing to solve the problems of the limited perception range of the current unmanned vehicles,and difficult to deal with the complex road conditions.The mechanism divides vehicles into different computing clusters based on path information and destinations.Within each computing cluster,the fog computing is used to achieve cooperation among vehicles,and expend the sensing scope of each vehicle to deal with the complex road conditions,satisfying the requirement of short delay.Among clusters,the fog computing can be achieved in cluster heads to realize the virtual distribution of lane resources,and the member vehicles refer to the virtual lane for maneuver control,which enhances the safety and reliability of vehicles.This thesis elaborates the mechanism of cooperative sensing and control based on fog computing,including flows,modules and communication procedures.The clusting algorithm based on path information and destination has also been described.In addition,based on the maps of real-world,we simulate and evaluate the proposed mechanism and algorithms.Simulation results show that the proposed mechanism and algorithm can well solve the problems of complicated situation on the road,and meet the low latency and effectiveness requirement of services of the unmanned drive.