Research On Simulation Of Left-Turn Driving Behavior At Unsignalized T-shaped Intersection Based On Cellular Automata Model

T-intersections play a vital role in urban traffic network,which connect roads and expand them as lines to a broader surface.Despite the importance of intersections,they can also serve as a deterrent and decrease the traffic capacity due to high frequencies of traffic accidents.Therefore,in order to enrich the study of traffic flows on T-intersections and get a more intuitive understanding of their characteristics,it is crucial to design models to fit practical traffic flows and explore the methods of controlling and management for T-intersections.This research adopts Cellular Automata Model as it has strong rule adaptability and high operational efficiency with simple theoretical principles to stimulate and model traffic flows on Tintersections.This paper will illustrate the work as followed:1.Processing collected video data on T-intersections.By collecting and preprocessing the video data,this paper concludes the patterns of driving behaviors for left-turning vehicles on Tintersections,and captures the characters of unusual left-turning driving behaviors based on the understanding of normal traffic flows on unsignallized T-intersections.2.Designing rules for vehicles on T-intersections based on different marked driving behaviors through the study of Cellular Automata Model and analysis of data.This paper segments judgement areas for T-intersections for the purpose of more practical policy-designing for real-time vehicles driving and stimulating complicated traffic behaviors.Meanwhile,this paper designs Lane Changing Model based on speed difference to fit normal driving psychology by improving the Cellular Automata Model.3.Stimulating different driving behaviors in the designed model by setting status of vehicles,accidents,delay and traffic flows as output indicators,and comparing different characteristics and trends of indicators to conclude the influences on T-intersections caused by different driving behaviors;exploring the impact of different left-turning behavior trace on Tintersections from two perspectives: operational efficiency and security.