RouteRL: Multi-agent reinforcement learning framework for urban route choice with autonomous vehicles

TL;DR

RouteRL is a framework combining multi-agent reinforcement learning with traffic simulation to optimize autonomous vehicle route choices in urban environments, advancing research in MARL and transportation modeling.

Contribution

It introduces a novel MARL-based framework for urban route choice simulation involving autonomous vehicles and human drivers, integrating behavioral models with reinforcement learning.

Findings

Effective simulation of driver and AV route choices

Potential for improving urban traffic efficiency

Framework supports research in MARL and human-AI interaction

Abstract

RouteRL is a novel framework that integrates multi-agent reinforcement learning (MARL) with a microscopic traffic simulation, facilitating the testing and development of efficient route choice strategies for autonomous vehicles (AVs). The proposed framework simulates the daily route choices of driver agents in a city, including two types: human drivers, emulated using behavioral route choice models, and AVs, modeled as MARL agents optimizing their policies for a predefined objective. RouteRL aims to advance research in MARL, transport modeling, and human-AI interaction for transportation applications. This study presents a technical report on RouteRL, outlines its potential research contributions, and showcases its impact via illustrative examples.