Multi-Agent Based Chaotic Traffic Simulation for Cairo Ring Road

摘要

Transportation simulations are an important part of today's decision making process for transport infrastructure and management. While proposed changes are getting more and more complex, tools supporting the decision making process are struggling to keep up. Traditional, flow-based traffic assignment tools are limited in the number of different person groups that can be distinguished and do in most cases not offer fully time-dynamic results. Newer technologies like agent-based simulations overcome those problems. This paper presents a novel traffic simulation scheme capable of modeling chaotic motorway traffic. Different from other lane-based or following-based approaches, the proposed approach models traffic as a large navigational problem in an agent based simulation context. In addition, the approach is efficiently able to handle hard cases like overtaking, behavior at turning and aggressive driving behavior. The simulation was demonstrated at real-time rates using MATSim applied to Cairo Ring Road. It has been described as well how MATSim simulation was extended to incorporate aggressive and careless drivers' behavior.