Research On Vehicle Eco-driving Control Strategy Based On Prediction Of Vehicle Speed At Signalized Intersections

In 2020,China will build a well-off society in an all-round way,and people’s living standards will continue to improve.With the rapid development of automobile industry,as a luxury consumer goods,automobile has entered thousands of households.As a common means of transportation in our daily life,cars greatly improve the convenience of people’s travel.However,with the rapid increase of car ownership,the problems of energy consumption,environmental pollution,traffic congestion,noise pollution and so on caused by the use of cars have attracted more and more attention.In the longterm research,it is found that the driving behavior of vehicles is directly related to the fuel consumption and emission of vehicles,especially in the urban signalized intersections with dense traffic flow.Due to the control of signal lights,high fuel consumption and high emission driving behaviors,such as parking at idle speed,severe acceleration and deceleration,often occur at signalized intersections,resulting in increased fuel consumption and emissions.Therefore,it is of great significance to improve the driving behavior of vehicles at signalized intersections.This paper studies the vehicle ecological driving control strategy at signalized intersection based on the speed prediction of the front vehicle in the vehicle road collaborative environment.In order to improve the driving behavior of the target vehicle passing through the signalized intersection,the speed guidance trajectory of the target vehicle passing through the signalized intersection is solved under the condition of vehicle driving ahead.The specific work is as follows:(1)The fuel consumption and emission model of target vehicle is establishedCombined with the application scenario,the micro fuel consumption model is selected,and the fuel consumption data is collected through the test to establish the target vehicle fuel consumption model.According to the specific power parameters,the emission rate,speed,acceleration and other data of each pollutant are collected through the test,and the emission factors of each VSP bin are obtained by processing,and the emission model of the target vehicle based on VSP is established.(2)Prediction of vehicle speed trajectory based on RBF neural networkAt the signalized intersection,the speed track of vehicles passing through the signalized intersection is collected under seven signal light information schemes,which is regarded as the speed track of vehicles in front.Combined with the three initial speeds of vehicles,21 scenes are constructed.RBF neural network is used to predict the speed trajectory of the front car in 21 scenarios.(3)Vehicle eco-driving control strategy at signalized intersectionAccording to the target driving speed of each scenario,the optimal driving path is determined by the target driving speed of each scenario.This paper studies the trajectory of target vehicle passing through signalized intersection in single lane environment.Aiming at the deficiency of vehicle ecological driving control strategy in some scenarios in single lane environment,it designs vehicle ecological driving control strategy at signalized intersection in multi Lane environment,and solves the optimal speed guidance trajectory,lane changing longitudinal displacement and lateral trajectory of target vehicle passing through signalized intersection To induce the target vehicle to change lanes and overtake,so as to further reduce the fuel consumption and emissions of the target vehicle passing through the signalized intersection.(4)Real vehicle testThe fuel consumption and emissions of the target vehicle guided by the optimal speed and the driver driving the vehicle through the signalized intersection are measured by using the drum test bench and the portable emission measurement system(PEMS).Through the comparison of test results,the energy saving and emission reduction effect of the ecological driving control strategy at signalized intersection based on the speed prediction of vehicles in front is verified.