Developing a Merge Lane Change Decision Policy for Autonomous Vehicles by Deep Reinforcement Learning

Bingyi Fan, Yuhan Zhou, Hani Mahmassani

Research output: Chapter in Book/Report/Conference proceedingConference contribution

6 Scopus citations

Abstract

With autonomous vehicles (AVs) being actively developed, it becomes possible to optimize vehicle control policies and traffic management tools in a mixed vehicular environment. For individual AV control, acceleration and lane change are the two elementary driving behaviors that need to be coordinated to minimize disturbance to the entire traffic dynamics. In this paper, a joint decision policy of acceleration and lane change actions for AVs on a merging ramp is proposed and trained in a mixed autonomy traffic, using the technique of deep reinforcement learning. Our method is able to train policies that have limited impact on highway traffic while maintaining a relatively high merge throughput. We experimented with two reward functions, designed for the AV's selfish benefits and for the network traffic's social benefits. This paper then examines the emergent behaviors exhibited by the trained policies and their impacts on the main highway traffic at different density levels.

Original languageEnglish (US)
Title of host publication2021 IEEE International Intelligent Transportation Systems Conference, ITSC 2021
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages963-968
Number of pages6
ISBN (Electronic)9781728191423
DOIs
StatePublished - Sep 19 2021
Event2021 IEEE International Intelligent Transportation Systems Conference, ITSC 2021 - Indianapolis, United States
Duration: Sep 19 2021Sep 22 2021

Publication series

NameIEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC
Volume2021-September

Conference

Conference2021 IEEE International Intelligent Transportation Systems Conference, ITSC 2021
Country/TerritoryUnited States
CityIndianapolis
Period9/19/219/22/21

ASJC Scopus subject areas

  • Automotive Engineering
  • Mechanical Engineering
  • Computer Science Applications

Fingerprint

Dive into the research topics of 'Developing a Merge Lane Change Decision Policy for Autonomous Vehicles by Deep Reinforcement Learning'. Together they form a unique fingerprint.

Cite this