A Multi-Agent Approach to Microscopic Urban Traffic Flow Simulation

Abstract

Urban traffic flow simulation has been important for making new control strategies that can reduce traffic jams. In this paper, we propose a method to microscopically simulate urban traffic flow based on multi-agent systems. Our proposed model is composed of four kinds of agent sets to be cooperated one another. This paper especially describe vehicle agents, each of which emerges traffic flow directly. Each vehicle agent uses two types of knowledge bases to determine its own actions to realize more realistic simulation, in which one includes basic traffic rules, e.g., stop if the signal is red, and the other includes drivers' individualities, e.g., speed maniac. In determining a behavior, each agent first recognizes neighboring environments, then an action is achieved by repeating one of three unit driving operations, i.e., acceleration, braking, or steering operations. We have developed a 3D animation system to visualize traffic flow, demonstrating that traffic jams are emerged due to interaction among vehicle agents.