Development and evaluation of an agent-based adaptive traffic signal control system

Adaptive traffic control systems aim to smooth traffic flows, increase throughput, and reduce delays, environment emissions and fuel consumption. These systems can help ease congestion and its negative externalities without the cost and environmental impacts of road expansion. A decentralised control approach, allowing individual controllers to make decisions based on information they receive from loop detectors and from each other, is proposed for a new generation of adaptive traffic signal control strategies. There is some evidence in the recent literature to suggest higher benefits for such systems, when compared with existing approaches. The primary objective of this study is to demonstrate the feasibility of using agent technologies to develop a decentralised adaptive traffic control system. By modelling the signal controllers as intelligent agents, it is possible to tune the actions of individual controllers through the agent concept of collaboration. The study also aims to evaluate the performance of the agent-based traffic signal control system by comparing its performance to other adaptive signal control systems. This will be achieved in simulation studies by developing an interface to link the agent-based traffic signal control system with a microscopic traffic simulation tool.


  • English

Media Info

  • Pagination: 13p

Subject/Index Terms

Filing Info

  • Accession Number: 01390327
  • Record Type: Publication
  • Source Agency: ARRB
  • Files: ATRI
  • Created Date: Aug 23 2012 4:44AM