Towards a Very Large Scale Traffic Simulator for Multi-Agent Reinforcement Learning Testbeds

Abstract

Listen

Translate article

Smart traffic control and management become an emerging application for Deep Reinforcement Learning (DRL) to solve traffic congestion problems in urban networks. Different traffic control and management policies can be tested on the traffic simulation. Current DRL-based studies are mainly supported by the microscopic simulation software (e.g., SUMO <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup> <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup> https://www.eclipse.org/sumo/), while it is not suitable for city-wide control due to the computational burden and gridlock effect. To the best of our knowledge, there is a lack of studies on the large-scale traffic simulator for D RL testbeds. In view of this, we propose a meso-macro traffic simulator for very large-scale DRL scenarios. The proposed simulator integrates meso scopic and macroscopic traffic simulation models to improve efficiency and eliminate gridlocks. The meso scopic link model simulates flow dynamics on roads, and the macroscopic Bathtub model depicts vehicle movement in regions. Moreover, both types of models can be hybridized to accommodate various DRL tasks. The result shows that the developed simulator only takes 46 seconds to finish a 24-hour simulation in a very large city with 2.2 million vehicles, which is much faster than SUMO. In the future, the developed meso-macro traffic simulator could serve as a new environment for very large-scale DRL problems.