Research On The Interaction Between Two Vehicles Under The Car Following Conditions Based On Natural Driving Data

In recent years,the increase in the number of motor vehicles and drivers has had a great impact on people ’s production and life,and people have paid more and more attention to the resulting road congestion,environmental pollution and other issues,especially paying more attention to bringing huge casualties and property Lost frequent traffic accidents.Nowadays,the car following behavior has become one of the most common traffic phenomena on urban roads.Studying the driving laws of adjacent vehicles has aroused the interest of many researchers,studying how the parameters of the two vehicles affect each other under the following conditions And quantifying the mutual influence of the parameters of the two vehicles will provide a reference for improving the safety of following the vehicle and establishing the following model.Natural driving data is comprehensive,true and reliable,and the samples are widely representative.With its unique advantages,it provides an important basis for the study of domestic driving behavior characteristics.This article will rely on the natural driving data collected by the China Automotive Engineering Research Institute Co.,Ltd.to carry out relevant research on the driving conditions.Firstly,perform noise reduction and filtering processing on the collected natural driving data,refer to other people’s literature and combine the data itself to formulate the extraction rules of the following driving conditions,extract a total of 1207 following driving conditions,and use the kernel based on the following driving fragments Density estimation studies the probability density distribution of the driver ’s front vehicle speed,front vehicle acceleration,own vehicle speed,own vehicle acceleration,relative distance between two vehicles,and relative speed of two vehicles when following the vehicle.Probability density distribution under acceleration conditions and probability density distribution under own vehicle speed and acceleration in front of vehicle speed and acceleration.Secondly,in order to further analyze the interaction between the two vehicles under different kinematics of the preceding vehicle in the following conditions,the kinematics of the preceding vehicle is divided,and the threshold for each kinematics is established by K-means clustering algorithm to determine The acceleration of the preceding car in the following car is used as input to obtain the thresholds for the division of acceleration,deceleration,and uniform speed,and the following driving conditions of different kinematics of the preceding car are extracted,and then the kernel density is used to obtain the difference of the preceding car in the following car Probability density distribution of the parameters of the preceding vehicle speed,the preceding vehicle acceleration,the own vehicle speed,the own vehicle acceleration,the relative speed of the two vehicles,and the relative distance between the two vehicles under the motion state,and then separately analyze the front of the vehicle in different motion states with the following vehicle conditions Probability density distribution of vehicle speed and acceleration under vehicle speed and acceleration conditions and probability density distribution of vehicle speed and acceleration under vehicle speed and acceleration conditions.Finally,a quantitative analysis of the interaction between the parameters of the two vehicles under the following conditions was carried out.The parameters of the two vehicles under the following conditions were analyzed first,and then the information flow between the parameters of the two vehicles under the following conditions was analyzed using transfer entropy.The situation,and then the total car following conditions and the effect of the following car parameters under different kinematics of the preceding car under different kinematics.