What Can Traffic Lights Learn from Us?

This article attempts to answer the following question. Could traffic light controlled junctions ever think for themselves? The Transportation Research Group (TRG) at the University of Southampton has been investigating the use of Artificial Intelligence in traffic control. While the analogy with human thought is still a bit of a stretch, Machine Learning techniques have been employed to develop junction controllers that can, in a real sense lean, strategies from changing the traffic lights from a human expert and even learn improved strategies through experience. This article will present a trained neural network based junction controller and show how the control system can be self learning at T-junction intersections..

• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/issn/00410683
• Authors:
  • Box, Simon
• Publication Date: 2011-9

Language

• English

Media Info

• Media Type: Digital/other
• Features: Figures; Tables;
• Pagination: pp 337-339
• Serial:
  • Traffic Engineering & Control
  • Volume: 52
  • Issue Number: 8
  • Publisher: Hemming Group, Limited
  • ISSN: 0041-0683
  • Serial URL: http://www.tecmagazine.com/

Subject/Index Terms

• TRT Terms: Intersections; Neural networks; T intersections; Traffic control devices; Traffic signal controllers; Traffic signals; Wireless communication systems
• Subject Areas: Highways; Operations and Traffic Management; I73: Traffic Control;

Filing Info

• Accession Number: 01356867
• Record Type: Publication
• Files: TRIS
• Created Date: Nov 10 2011 10:39AM