Research On Vehicles Collaboration Approach And Compressed Sensing Data Acquiring By Vehicle Infrastructure Integration

Traffic safety and congestion are often derived from incompleteness vehicles information,including access and process the information untimely and inaccurately.So the vehicle infrastructure integration system has become the research focus of the field of intelligent transportation,which has a huge potential to solve the above problems.Vehicles collaboration is the fundamental requirement of the vehicle infrastructure integration system.However,how to collect the massive traffic information effectively and transfer simply are great challenges.Therefore,this paper is aim to studying vehicles collaboration under the vehicle infrastructure integration system,and exploring vehicles sensing information collection and information exchange methods with the help of compressed sensing information interaction.The study relies on the national natural fund project "Research on the mechanisms of vehicles' fractal collaboration under holographic transportation environment "(NO:61174176),Firstly,according to the car following and lane changing behavior considering the traffic conditions deficiency,the algorithm of car cooperation change model is researched and the car cooperation change method is established,which take into account the road resource utilization rate and the safety of driving behavior.Then the problem of the information interaction perception is analyzed in the process of vehicles cooperation.Secondly,the data acquiring method of vehicular sensor networks based on compressed sensing(CS)is proposed.By analyzing the signal sparse of vehicle sensor and the random observation of Gauss matrix,the key vehicle acceleration and speed information can be perceived during the process of Vehicle collaborative.It is verified through simulation that the CS-based vehicular data including vehicle speed and acceleration could be reconstructed by means of the proposed method.the simulation results show that the method applied to the vehicle sensor data acquisition has feasibility.Then,according to the collection method through CS for vehicle sensor information,The paper studies the information transmission process and transmission delay from vehicle to vehicle.A fusion method of information interaction among a number of vehicles is put forward.Realize the multiple vehicles information interactive fusion by establishing a joint sparse model about information interactive further.Thereby it reduces the information transmission load during information interactive process.Finally,Selecting one of line in Hangzhou as simulation example,it is established the microscopic simulate model for vehicles collaboration approach in PARAMICS.The results show that the approach can effectively maintain the stabilization of traffic conditions,improve the efficiency of road running and ensure vehicle for safety driving.It can apply to the vehicle-road coordination system in the future.A united simulation platform is built with PARAMICS of traffic system simulation and OMNET++ of network communication simulation.Information interaction performance evaluation of vehicle moving scenes can be achieved when moving scenes of vehicles collaboration are transplanted into network communication simulation.The paper compares of information interaction performance before and after implementing the method of vehicles collaboration and before and after implementing information compressed sensing.The results demonstrate that information interaction course caused by vehicles collaboration increases the burden of information interaction.While information throughput is increased and time delay is decreased effectively in vehicles collaboration with the method of information compressed sensing at the same time.