TY - GEN
T1 - Real Time Building and Maintaining Causal Congestion Graph for Intelligent Traffic Management
AU - Kapoor, Viral
AU - Saxena, Divya
AU - Raychoudhury, Vaskar
AU - Kumar, Sandeep
N1 - Funding Information:
ACKNOWLEDGMENT We sincerely thank Dr. Ajay Kshemkalyani from CSE, Univ. of Illinois at Chicago for his valuable inputs and support. This paper is partially funded by the Alexander von Humboldt Foundation through the Post-doctoral Fellow Dr. Vaskar Raychoudhury.
Publisher Copyright:
© 2018 IEEE.
PY - 2018/10/2
Y1 - 2018/10/2
N2 - Traffic congestion is a major problem for commuters and it has several negative impacts on environment and economy alike. Existing congestion detection techniques are mostly centralized in nature. In this paper, we proposed a distributed and localized congestion detection approach which seems more suited to real-time congestion detection across large-scale road network. Our algorithm studies causal relation between congested road intersections, i.e., how the congestion propagates from a point in the road network to all directions and can further predict the possible propagation pattern. We have evaluated our algorithm using GPS traces of more than 2299 taxis in Shanghai collected from Jan. 31 to March 1, 2007. We study the repetitive nature of traffic on weekdays and weekends across 281 physical locations (sites) and 500 road intersection points. The results show that we can predict future congestion patterns by analyzing real time causal congestion relations with accuracy up to 65%.
AB - Traffic congestion is a major problem for commuters and it has several negative impacts on environment and economy alike. Existing congestion detection techniques are mostly centralized in nature. In this paper, we proposed a distributed and localized congestion detection approach which seems more suited to real-time congestion detection across large-scale road network. Our algorithm studies causal relation between congested road intersections, i.e., how the congestion propagates from a point in the road network to all directions and can further predict the possible propagation pattern. We have evaluated our algorithm using GPS traces of more than 2299 taxis in Shanghai collected from Jan. 31 to March 1, 2007. We study the repetitive nature of traffic on weekdays and weekends across 281 physical locations (sites) and 500 road intersection points. The results show that we can predict future congestion patterns by analyzing real time causal congestion relations with accuracy up to 65%.
KW - causal congestion tree
KW - causal graphs
KW - causal relations
KW - intersections
KW - road congestion
KW - road traffic
UR - http://www.scopus.com/inward/record.url?scp=85056462030&partnerID=8YFLogxK
U2 - 10.1109/PERCOMW.2018.8480324
DO - 10.1109/PERCOMW.2018.8480324
M3 - Conference article published in proceeding or book
AN - SCOPUS:85056462030
T3 - 2018 IEEE International Conference on Pervasive Computing and Communications Workshops, PerCom Workshops 2018
SP - 770
EP - 775
BT - 2018 IEEE International Conference on Pervasive Computing and Communications Workshops, PerCom Workshops 2018
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2018 IEEE International Conference on Pervasive Computing and Communications Workshops, PerCom Workshops 2018
Y2 - 19 March 2018 through 23 March 2018
ER -
