Multi-section Cellular Automata Model of Traffic Flow

Liang Jing-Yun,Zhang Li-Li,Luan Xi-Dao,Guo Jin-Lin,Lao Song-Yang,Xie Yu-Xiang

DOI: https://doi.org/10.7498/aps.66.194501

2017-01-01

Abstract:It is more common for drivers to pass through multiple sections to reach destinations instead of single road section. Howerver, most of researches concentrate on improving the effect in an independent section. Based on traditional cellular automata traffic model, a multi-section model is proposed by regarding serverl road sections as a traffic system. In this model, different sections of the road might have different lengths, numbers of lanes or maximal speeds. And vehicles travel from one section to another. The main difficulty lies in dealing with the relationships among the traffic flows of different sections. Besides basic rules in NaSch model, the vehicle inflow rule, crossroad randomization brake rule and crossroad inflow rule is added in this paper to enable vehicles to flow between sections. At the beginning of section, to avoid conflicting at crossroads under open boundary condition, the concept of car pool is introduced when new vehicles enter into sections. Before arriving at the end of section, crossroad randomization brake is used to simulate the influences of crossroads. Speed decreases in probability until lower than a maximal crossroad speed. When leaving the section, vehicles go to the next section with a straight ratio. Also, new vehicles may enter according to traffic condition. Therefore, cellular automata of different sections can be connected in series.Finally, numerical simulation is demonstrated to study the influences of important parameters, including traffic inflow probability, maximal crossroad speed and crossroad randomization brake probability. Compared with traditional models, this model focuses on connecting sections. And improvements of basic models can be implanted easily, thereby increasing the accuracy of the whole model in the future. The experimental result are as follows. 1) According to space-time graphs of different inflow probabilities, there is a new kind of traffic flow called mixed flow. Traffic congestion often starts from crossroads, and spreads to the whole section. And traffic jams in previous section might relieve traffic pressure in latter section. 2) With the increase of traffic inflow probability, crossroads tends to have a greater influence on average speed as well as average traffic density. What is more, the moderate increase of vehicle numbers could cause the road capacity to drop rapidly if it exceeds the threshold value.