MATES : Multi-Agent based Traffic and Environment Simulator -- Theory, Implementation and Practical Application -
S. Yoshimura

doi:10.3970/cmes.2006.011.017
Source CMES: Computer Modeling in Engineering & Sciences, Vol. 11, No. 1, pp. 17-26, 2006
Download Full length paper in PDF format. Size = 365,745 bytes
Keywords Intelligent agent, Multi-agent model, Complex system, Traffic simulator, Layered road network, Virtual social experiment.
Abstract This paper describes a development of an advanced traffic simulator based on a multi-agent approach which is named MATES (\relax$\@@underline {\hbox {M}}\mathsurround \z@$\relax ulti-\relax$\@@underline {\hbox {A}}\mathsurround \z@$\relax gent based \relax$\@@underline {\hbox {T}}\mathsurround \z@$\relax raffic and \relax$\@@underline {\hbox {E}}\mathsurround \z@$\relax nvironment \relax$\@@underline {\hbox {S}}\mathsurround \z@$\relax imulator). City traffic phenomena are essentially regarded as complex systems consisting of a number of human beings. Each element creating traffic phenomena such as car (driver), traffic signal, pedestrian and others is modeled as an intelligent agent that possesses its own logic of behavior and preference. The environment surrounding each agent consists of other cars, road network, traffic signals, pedestrian and others. Interaction among numerous agents simulates nonlinear behaviors of city traffic phenomena. First we describe details of the theory and implementation of MATES, and then its fundamental features are examined. Finally MATES is applied to simulate city traffic in Kashiwa city in Japan, employing various real world data as input. Through the simulations, it is clearly demonstrated that MATES is a powerful tool to study complex city traffic problems precisely.
PDF download PDF