Urban traffic control is a large and complex system. When traffic congestion occurs it generally cannot be resolved by the workings of a single intersection. Using a multi-agent system to set up a traffic control framework, we can simulate different control strategies. We have defined different kinds of traffic-control-agent classes while researching the interactions and dynamic adapting mechanism among agents. Utilizing self-organization and sliding window skills, we have designed an effective congestion control mechanism. In simulating the traffic environment, we obtained a valuable experimental result.
New Methods for Research in Multi-Agent Traffic Signal Control
Ninth International Conference of Chinese Transportation Professionals (ICCTP) ; 2009 ; Harbin, China
ICCTP 2009 ; 1-7
2009-07-23
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
New Methods for Research in Multi-Agent Traffic Signal Control
British Library Conference Proceedings | 2009
|Optimizing multi-agent based urban traffic signal control system
Taylor & Francis Verlag | 2019
|Reinforcement learning-based multi-agent system for network traffic signal control
Tema Archiv | 2010
|