TY - GEN

T1 - Genetic Based Approach for Novosibirsk Traffic Light Scheduling

AU - Davydov, Ivan

AU - Tolstykh, Daniil

AU - Kononova, Polina

AU - Legkih, Irina

N1 - Publisher Copyright: © 2019 IEEE. Copyright: Copyright 2020 Elsevier B.V., All rights reserved.

PY - 2019/8

Y1 - 2019/8

N2 - Congestion, derived from the permanent increase in road traffic, is a pressing problem in the big cities all around the world nowadays. Thus, the methods of the intelligent control of vehicles traffic have to answer the growing demand. Optimization of traffic signal plans is an important step in this direction. Well-tuned traffic lights schedule augments the efficiency of vehicles flows processing. The research in intelligent traffic signal control helps to significantly improve a traffic situation, reduce the average vehicles waiting time and increase the average speed in the road network. In this study, we propose different configurations of a genetic algorithm to find an effective traffic lights schedule on the real road network. This heuristic evolutionary algorithm is known to cope well with different kinds of optimization problems. For application part of our research, we consider a complex segment of the street network in the city of Novosibirsk, Russia. Using a microscopic traffic simulator, SUMO, we model a corresponding road map fragment. The obtained model serves to evaluate the solutions of the traffic scheduling problem. We analyze the performance of the proposed genetic algorithm with different parameters and discuss the results of numerical experiments considering three different objectives functions which reflect traffic congestion. We show that the proposed approach can be applied to increase the quality of the traffic lights schedule, reducing traffic jams.

AB - Congestion, derived from the permanent increase in road traffic, is a pressing problem in the big cities all around the world nowadays. Thus, the methods of the intelligent control of vehicles traffic have to answer the growing demand. Optimization of traffic signal plans is an important step in this direction. Well-tuned traffic lights schedule augments the efficiency of vehicles flows processing. The research in intelligent traffic signal control helps to significantly improve a traffic situation, reduce the average vehicles waiting time and increase the average speed in the road network. In this study, we propose different configurations of a genetic algorithm to find an effective traffic lights schedule on the real road network. This heuristic evolutionary algorithm is known to cope well with different kinds of optimization problems. For application part of our research, we consider a complex segment of the street network in the city of Novosibirsk, Russia. Using a microscopic traffic simulator, SUMO, we model a corresponding road map fragment. The obtained model serves to evaluate the solutions of the traffic scheduling problem. We analyze the performance of the proposed genetic algorithm with different parameters and discuss the results of numerical experiments considering three different objectives functions which reflect traffic congestion. We show that the proposed approach can be applied to increase the quality of the traffic lights schedule, reducing traffic jams.

KW - genetic algorithm

KW - SUMO

KW - traffic lights

UR - http://www.scopus.com/inward/record.url?scp=85077980952&partnerID=8YFLogxK

U2 - 10.1109/OPCS.2019.8880158

DO - 10.1109/OPCS.2019.8880158

M3 - Conference contribution

T3 - 2019 15th International Asian School-Seminar Optimization Problems of Complex Systems, OPCS 2019

SP - 31

EP - 36

BT - 2019 15th International Asian School-Seminar Optimization Problems of Complex Systems, OPCS 2019

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 15th International Asian School-Seminar Optimization Problems of Complex Systems, OPCS 2019

Y2 - 26 August 2019 through 30 August 2019

ER -