Digital twin-enabled reinforcement learning for end-to-end autonomous driving

Jingda Wu, Zhiyu Huang, Peng Hang, Chao Huang, Niels De Boer, Chen Lv

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review

16 Citations (Scopus)

Abstract

Digital twin maps the physical plant to a real-time digital representation and facilities product design and decision-making processes. In this paper, we propose a novel digital twin-enabled reinforcement learning approach and apply it to an autonomous driving scenario. To further improve the data efficiency of reinforcement learning, which often requires a large amount of agent-environment interactions during the training process, we propose a digital-twin environment model that can predict the transition dynamics of the physical driving scene. Moreover, we propose a rollout prediction-compatible reinforcement learning framework, which is able to further improve the training efficiency. The proposed framework is validated in an autonomous driving task with a focus on lateral motion control. The simulation results illustrate that our method could significantly speed up the learning process and the resulting driving policy could achieve better performance, compared to the conventional reinforcement learning approach, which demonstrates the feasibility and effectiveness of the proposed digital-twin-enabled reinforcement learning method.

Original languageEnglish
Title of host publicationProceedings 2021 IEEE 1st International Conference on Digital Twins and Parallel Intelligence, DTPI 2021
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages62-65
Number of pages4
ISBN (Electronic)9781665433372
DOIs
Publication statusPublished - 15 Jul 2021
Externally publishedYes
Event1st IEEE International Conference on Digital Twins and Parallel Intelligence, DTPI 2021 - Beijing, China
Duration: 15 Jul 202115 Aug 2021

Publication series

NameProceedings 2021 IEEE 1st International Conference on Digital Twins and Parallel Intelligence, DTPI 2021

Conference

Conference1st IEEE International Conference on Digital Twins and Parallel Intelligence, DTPI 2021
Country/TerritoryChina
CityBeijing
Period15/07/2115/08/21

Keywords

  • Autonomous driving
  • End-to-end control
  • Environment digital twin model
  • Reinforcement learning

ASJC Scopus subject areas

  • Artificial Intelligence
  • Information Systems and Management
  • Safety, Risk, Reliability and Quality
  • Media Technology
  • Control and Optimization
  • Modelling and Simulation

Fingerprint

Dive into the research topics of 'Digital twin-enabled reinforcement learning for end-to-end autonomous driving'. Together they form a unique fingerprint.

Cite this