Learning Eco-Driving Strategies at Signalized Intersections

TL;DR

This paper introduces a reinforcement learning-based eco-driving strategy at signalized intersections, significantly reducing fuel consumption and emissions while improving travel speed, with effectiveness demonstrated across various levels of autonomous vehicle penetration.

Contribution

It presents a novel RL approach for eco-driving at intersections, showing its effectiveness and generalizability across different traffic scenarios and levels of autonomous vehicle adoption.

Findings

Up to 18% reduction in fuel consumption.

Up to 25% reduction in CO2 emissions.

20% improvement in travel speed.

Abstract

Signalized intersections in arterial roads result in persistent vehicle idling and excess accelerations, contributing to fuel consumption and CO2 emissions. There has thus been a line of work studying eco-driving control strategies to reduce fuel consumption and emission levels at intersections. However, methods to devise effective control strategies across a variety of traffic settings remain elusive. In this paper, we propose a reinforcement learning (RL) approach to learn effective eco-driving control strategies. We analyze the potential impact of a learned strategy on fuel consumption, CO2 emission, and travel time and compare with naturalistic driving and model-based baselines. We further demonstrate the generalizability of the learned policies under mixed traffic scenarios. Simulation results indicate that scenarios with 100% penetration of connected autonomous vehicles (CAV) may yield as high as 18% reduction in fuel consumption and 25% reduction in CO2 emission levels while even improving travel speed by 20%. Furthermore, results indicate that even 25% CAV penetration can bring at least 50% of the total fuel and emission reduction benefits.