DeepAD: An integrated decision-making framework for intelligent autonomous driving

Autonomous vehicles have the potential to revolutionize intelligent transportation by improving traffic safety, increasing energy efficiency, and reducing congestion. In this study, a novel framework termed DeepAD was proposed and validated for decision making in intelligent autonomous driving via deep reinforcement learning. This framework incorporates multiple driving objectives such as efficiency, safety, and comfort to make informed decisions regarding autonomous vehicles (AVs). The decision-making process utilizes the origin–destination information for macrolevel routing and determines microlevel car-following and lane-changing behaviors. The lane-changing behavior is discretized and learned through a deep Q-network, and the continuous car-following behavior is learned through a deep deterministic policy gradient. Comprehensive simulation experiments on a real-world network demonstrated that DeepAD outperformed human driving while maintaining a desirable level of efficiency, safety, and comfort. In the real-world road networks experiment, multiple indexes of vehicles in the high AVs penetration rate group significantly outperformed that of the group with lower AVs penetration rate. Overall, the proposed framework provides insights into intelligent autonomous driving to improve urban mobility.

• Record URL:
  https://doi.org/10.1016/j.tra.2024.104069
• Record URL:
  http://www.sciencedirect.com/science/article/pii/S0965856424001174
• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/issn/09658564
• Supplemental Notes:
  • © 2024 Elsevier Ltd. All rights reserved. Abstract reprinted with permission of Elsevier.
• Authors:
  • Shi, Yunyang
  • Liu, Jinghan
  • Liu, Chengqi
  • Gu, Ziyuan
  • 0000-0002-2059-4809
• Publication Date: 2024-5

Language

• English

Media Info

• Media Type: Web
• Features: Appendices; References;
• Pagination: 104069
• Serial:
  • Transportation Research Part A: Policy and Practice
  • Volume: 183
  • Issue Number: 0
  • Publisher: Elsevier
  • ISSN: 0965-8564
  • Serial URL: http://www.sciencedirect.com/science/journal/09658564

Subject/Index Terms

• TRT Terms: Autonomous vehicles; Decision making; Machine learning; Origin and destination; Simulation; Travel behavior
• Subject Areas: Data and Information Technology; Highways; Operations and Traffic Management;

Filing Info

• Accession Number: 01919314
• Record Type: Publication
• Files: TRIS
• Created Date: May 22 2024 10:28AM