Positioning And Clustering Algorithm Research Of Intelligent Connected Vehicle Based On Cyber-Physical System And Fuzzy Theory

In recent years,a new round of technologic and industrial revolution is developing in depth and breath,the integration of new generation of information technology and automotive industry has lead to a profound change in characteristics and distribution of the automotive products based on the internet.And automobile is becoming intelligent and networked,which is also called the intelligent connected vehicle.The Cyber-Physical System(CPS)is a multidimensional and complex system by combining computing,network and physical environment,whose intention is to obtain a real-time perception and precise control.Now,CPS is commonly researched and applied in the field of intelligent traffic and advanced automotive systems.Since intelligent connected vehicle has the similar system architecture and features with CPS,It is possible to apply CPS technology to intelligent connected vehicle to make the car safer and smarter.Based on the analysis of the relationship between intelligent connected vehicle and CPS,this paper proposed an integreated framework of intelligent connected vehicle positioning and clustering based on CPS,and studied the positioning algorithm and clustering algorithm in this framework.The main contents of this paper summarized as follows:1)An integreated framework of intelligent connected vehicle positioning and clustering is proposed based on CPS.Firstly,the related knowledge of CPS technology and intelligent connected vehicle is introduced.Then,the relationship between intelligent connected vehicle and CPS is analyzed,and lists the technical requirements of CPS for intelligent connected vehicle.Finally,the integration of intelligent connected vehicle positioning and clustering is achieved by combining with CPS theory.2)A composite vehicle positioning algorithm based on fuzzy theory(CPFT)is proposed.Firstly,a fuzzy-weighting Locating mechanism is put forward to overcome the individual limitations in each positioning technology,which utilizes the information of the neighboring vehicles and fuzzy controlling system to obtain the composite position of the vehicle.Then,a fuzzy Kalman filter,which adopts another fuzzy controller to dynamically adjust the measurement noise covariance based on the credibility of each positioning technique,is used to filter the composite position.Finally,simulation results show that CPFT algorithm can effectively integrate various positioning techniques to enhance positioning accuracy,and has strong robustness as well.3)A fuzzy vehicle clustering algorithm based on the above localization information(FCLI)is proposed.Firstly,a parameter represents the ability of the vehicle node to be a cluster head is defined based on localization information and fuzzy clustering method.Based on this parameter,the cluster formation algorithm of FCLI is established.Then,a Kalman filter is used to predict the position of the cluster members.Based on the position prediction and edge cluster members,the cluster maintenance algorithm of FCLI is constructed.Finally,simulation results show that FCLI algorithm can make the clusters more stable,and reduce the number of undecided nodes and network overhead.