Bus Signal Priority Control Under Mixed Traffic Conditions Considering Connected Automated Vehicles

With the rapid development of economy and urbanization,car ownership is growing faster as well.While bringing convenience to travel of people,it also causes urban problems such as traffic congestion,traffic accidents and air pollution,which seriously affects the sustainable development of cities.Public transport has the advantages of larger passenger capacity,less road occupation,and hence the priority for public transport can alleviate traffic congestion.With the development of connected vehicles technology,connected automated vehicles are gradually applied to the road environment,which can obtain a variety of basic traffic data.Therefore,this thesis adopts Kalman filter,car following theory and shockwave theory to predict bus speed,queue length in front of the buses under the mixed traffic condition and bus arrival time,then a bus signal priority method is proposed,which contribute to enrich the theory of bus signal priority and promote the application of bus signal priority to realistic roads.Firstly,a Kalman filter model for multiple moving scenarios is designed based on the kinematics of connected automated vehicles.Then the car following characteristic of buses and connected automated vehicles is analyzed and a bus speed prediction model is presented based on the car following theory.The calculated results indicate that the absolute error of predicting bus speed is less than 1m/s,the relative error is less than 10%,and the maximum relative error is less than 25%.Secondly,the shockwave theory is utilized to develop the queue length estimation model.The maximum likelihood function is then constructed by using the trajectory of connected automated vehicles to estimate the arrival rate at a signal cycle.Considering the predicted bus speed,average arrival density and critical density of vehicles are calculated.Following that,the queue length under under-saturated and over-saturated conditions is estimated.Then this study constructs a simulation platform based on VISSIM to evaluate the accuracy of the estimated value.The results show that under the under-saturated condition,the mean absolute percentage error of the maximum queue length is less than 8%when connected automated vehicles penetration reaches 40%.Under the over-saturation condition,the mean absolute percentage error of the maximum queue length is less than 9%and the mean absolute percentage error of the minimum queue length is less than 6%when connected automated vehicles penetration increases to 40%.Finally,combined with the trajectory of connected automated vehicles,the prediction model of bus arrival time at intersections under mixed traffic conditions is established based on the predicted bus speed and queue length.Then bus signal priority method of mixed traffic state is proposed by minimizing average person delay and green extension and red truncation are the priority strategies.After that,the intersection of Kexue Road and Shifocun Road is selected as the case to validate the performance of the proposed method.Results demonstrate that compared with the method with no bus signal priority,the proposed method can reduce average bus passenger delay and average person delay for the whole intersection.Under the under-saturated condition,the reduction ratio of average bus passenger delay decreases from 59.96%to 22.05%when traffic saturation increases from 0.6 to 0.9.This change turns from 55.43%to 16.96%for average person delay.Under the over-saturated condition,the reduction ratio of average bus passenger delay decreases from 13.17%to 3.74%when traffic saturation increases from 1 to 1.2.This change turns from 9.72%to 1.84%in terms of average person delay.