ISSN 1991-2927
 

ACP № 3 (65) 2021

Author: "Dmitrii Sergeevich Demakhin"

Irina Aleksandrovna Sedykh, Lipetsk State Technical University, Candidate of Physics and Mathematics, graduated from the Faculty of Automatization and Information Technologies of Lipetsk State Technical University (LSTU); Associate Professor at the Department of Mathematics of Lipetsk State Technical University; an author of monographs and articles, holds State Registration Certificates of computer programs in the field of neighborhood modelling of dynamic systems. [e-mail: sedykh-irina@yandex.ru]I. Sedykh,

Dmitrii Sergeevich Demakhin, Lipetsk State Technical University, graduated from the Faculty of Physics and Technology of Lipetsk State Technical University; Candidate for the Master’s Degree at LSTU; an author of articles, holds State Registration Certificates of computer programs in the field of neighborhood modelling of dynamic systems. [e-mail: dima-demahin@mail.ru]D. Demakhin

Flexible Control of Traffic Lights System on the Basis of Neural Networks 000_13.pdf

The traditional algorithm of crossroad traffic management with fixed order and switching-on duration was described in the article. Neural networks-based kind of control for crossroad group of traffic lights was offered as an alternative. The appropriate algorithm has been developed and realized in C++. The basic characteristics of designed neural network for crossroad traffic lights control were described. They include the architecture, assignment of neurons for input and output layers, the number of intermediate layer neurons, the activation function of neurons, the learning method. The realized neural network-based algorithm allows to put in practice flexible control of coordination traffic lights in case of the non-fixed order of traffic lanes priorities. Herewith, the possibility of change in green time of traffic light signal within predetermined limits with the aim of increasing capacity of crossroads in most problematic directions of transport movements was foreseen. Also, restrictions for prevention of blocking segregate streams for too long including blocking pedestrians in crosswalks were provided.

Traffic lights systems, mathematical modelling, neural networks, crossroad traffic management.

2017_ 1

Sections: Artificial intelligence

Subjects: Artificial intelligence.


© FRPC JSC 'RPA 'Mars', 2009-2021 The web-site runs on Joomla!