November 14, 2012
Dr. Nebiyou Tilahun, UPP, presents a seminar entitled "An agent based model of origin destination estimation (ADOBE)" Wednesday, November 14th at 4:00 pm in Rm 1127 SEO
November 7, 2012
Mr. Thomas Murtha, CMAP, will address the CTS-IGERT community at 4:00 p.m. in Room 1127 SEO.
October 24, 2012
Please join us in welcoming Dr. Bo Zou, CME, on Wednesday, October 24th, Room 1127 SEO, 4:00 p.m.
September 25, 2012
Award Received by Joshua Auld, CTS-IGERT alumnus.
April 20, 2012
Congratulations to James Biagioni, CTS Fellow and CS PhD candidate, winner of the Dean's Scholar award.
January 2, 2012
James Biagioni, CTS Fellow, receives "Best Presentation Award" at SenSys2011
July 30, 2010
Dr. Ouri Wolfson, Dr. Phillip Yu, and Leon Stenneth, CS student and CTS Associate, recently had a paper accepted to the 6th IEEE International Conference on Wireless and Mobile Computing, Networking and Communications (WiMob 2010).
March 4, 2011
Piotr Szczurek, LEAP Fellow and CS Phd candidate, will present a seminar entitled "Emergency Electronic Brake Light: When are Warnings Relevant to Drivers?" at 2:30 p.m. in Room 1000 SEO.
In 2005, the National Highway Traffic Safety Administration (NHTSA) identified eight potential safety applications that utilize Dedicated Short Range Communication (DSRC) technology. The applications were selected based on potential safety benefits they provide. Among them, the Emergency Electronic Brake Light (EEBL) application was determined to be one of three applications to possess high benefit potential. EEBL was defined as an application that warns drivers of any hard braking done by vehicles in front of them. The idea was to extend drivers visibility through the emergency brake notifications. This was described as most helpful in situations where visibility is limited, such as in adverse weather conditions.
While the EEBL application provides a high safety benefit, it is dependent on drivers reactions to the warnings it provides. When too many false warnings are shown, drivers can become desensitized, which over time will make them ignore the system. Knowing when to show and when not to show a warning is the problem of determining the relevance of the information.
This talk will present our method for estimating the relevance of information using machine learning techniques. The idea is to train a model by learning from examples of how drivers normally react, at the time when an EEBL warning is generated. It will be shown that this technique can maintain the safety benefit of the EEBL application, while drastically reduce the number of false warnings shown to the drivers.
Piotr Szczurek is a Ph.D. candidate at the University of Illinois at Chicago. He received his B.S. degree in Computer Science at the same university in 2005. From 2006, he has been a fellow of the Integrative Graduate Education and Research Traineeship (IGERT) program in Computational Transportation Science. He is advised by Professor Ouri Wolfson and Professor Jie Lin. His research interests include mobile databases and applications of machine learning.