Predicting the Traffic Capacity of an Intersection Using Fuzzy Logic and Computer Vision

Mathematics

Zlata Almetova¹  Vladimir Shepelev¹  Tatyana Gluchshenko²  Tatyana Bedych³  Alexandr Glushkov⁴ 

[1] Department of Automobile Transportation, South Ural State University (National Research University), 454080 Chelyabinsk, Russia;Department of Electric Power, Kostanay Regional University named after A. Baitursynov, Kostanay 110000, Kazakhstan;Department of Energy and Mechanical Engineering, M. Dulatov Kostanay Engineering and Economic University, Kostanay 110000, Kazakhstan;Department of Mathematical and Computer Modelling, South Ural State University (National Research University), 454080 Chelyabinsk, Russia;

关键词: traffic capacity of an intersection;    pedestrian flow;    traffic simulation;    fuzzy logic method;    predictive visualization of a vehicle flow;   

DOI  :  10.3390/math9202631

来源: DOAJ

【 摘 要 】

This paper presents the application of simulation to assess and predict the influence of random factors of pedestrian flow and its continuity on the traffic capacity of a signal-controlled intersection during a right turn. The data were collected from the surveillance cameras of 25 signal-controlled intersections in the city of Chelyabinsk, Russia, and interpreted by a neural network. We considered the influence of both the intensity of the pedestrian flow and its continuity on the traffic capacity of a signal-controlled intersection in the stochastic approach, provided that the flow of vehicles is redundant. We used a reasonably minimized regression model as the basis for our intersection models. At the first stage, we obtained and tested a simulated continuous-stochastic intersection model that accounts for the dynamics of traffic flow. The second approach, due to the unpredictability of pedestrian flow, used a relevant method for analysing traffic flows based on the fuzzy logic theory. The second was also used as the foundation to build and graphically demonstrate a computer model in the fuzzy TECH suite for predictive visualization of the values of a traffic flow crossing a signal-controlled intersection. The results of this study can contribute to understanding the real conditions at a signal-controlled intersection and making grounded decisions on its focused control.

【 授权许可】

Unknown