Evaluations of intelligent traffic signal control algorithms under realistic landmark-based traffic patterns over the NCTUns network simulator

Shie-Yuan Wang*, Yu Wei Li

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Scopus citations

Abstract

In this paper, we use the NCTUns network simulator to study intelligent traffic signal control algorithms. NCTUns is both a microscopic traffic simulator and a network simulator. We imported the real-life map of a district of the Taipei city into NCTUns to let simulated vehicles move on the roads of the imported map. Then, we used a new and unique feature of NCTUns - each simulated vehicle moves towards its assigned landmark on the map as its destination, to create more realistic traffic patterns in the studied district. With these realistic settings, we studied an intelligent traffic signal control algorithm that has been proposed in the literature. We found that this algorithm has unfairness problems and thus we proposed two mechanisms to improve it. Our simulation results show that, when compared with the original algorithm, the improved algorithm can further reduce the average time needed for a vehicle to reach its destination. In addition, the improved algorithm can mitigate the unfairness problem.

Original languageEnglish
Title of host publication2012 15th International IEEE Conference on Intelligent Transportation Systems, ITSC 2012
Pages379-384
Number of pages6
DOIs
StatePublished - 21 Dec 2012
Event2012 15th International IEEE Conference on Intelligent Transportation Systems, ITSC 2012 - Anchorage, AK, United States
Duration: 16 Sep 201219 Sep 2012

Publication series

NameIEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC

Conference

Conference2012 15th International IEEE Conference on Intelligent Transportation Systems, ITSC 2012
CountryUnited States
CityAnchorage, AK
Period16/09/1219/09/12

Fingerprint Dive into the research topics of 'Evaluations of intelligent traffic signal control algorithms under realistic landmark-based traffic patterns over the NCTUns network simulator'. Together they form a unique fingerprint.

Cite this