Pedestrian Behavior Prediction for Automated Driving: Requirements, Metrics, and Relevant Features

Automated vehicles require a comprehensive understanding of traffic situations to ensure safe and anticipatory driving. In this context, the prediction of pedestrians is particularly challenging as pedestrian behavior can be influenced by multiple factors. In this paper, the authors thoroughly analyze the requirements on pedestrian behavior prediction for automated driving via a system-level approach. To this end they investigate real-world pedestrian-vehicle interactions with human drivers. Based on human driving behavior they then derive appropriate reaction patterns of an automated vehicle and determine requirements for the prediction of pedestrians. This includes a novel metric tailored to measure prediction performance from a system-level perspective. The proposed metric is evaluated on a large-scale dataset comprising thousands of real-world pedestrian-vehicle interactions. The authors furthermore conduct an ablation study to evaluate the importance of different contextual cues and compare these results to ones obtained using established performance metrics for pedestrian prediction. The results highlight the importance of a system-level approach to pedestrian behavior prediction.

• Record URL:
  • https://doi.org/10.1109/TITS.2021.3135136
• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/oclc/41297384
• Supplemental Notes:
  • Copyright © 2022, IEEE.
• Authors:
  • Herman, Michael
  • Wagner, Jörg
  • Prabhakaran, Vishnu
  • Möser, Nicolas
  • Ziesche, Hanna
  • Ahmed, Waleed
  • Bürkle, Lutz
  • Kloppenburg, Ernst
  • Gläser, Claudius
• Publication Date: 2022-9

Language

• English

Media Info

• Media Type: Web
• Pagination: pp 14922-14937
• Serial:
  • IEEE Transactions on Intelligent Transportation Systems
  • Volume: 23
  • Issue Number: 9
  • Publisher: Institute of Electrical and Electronics Engineers (IEEE)
  • ISSN: 1524-9050
  • Serial URL: http://ieeexplore.ieee.org/xpl/RecentIssue.jsp?punumber=6979

Subject/Index Terms

• TRT Terms: Autonomous vehicles; Behavior; Pedestrian vehicle interface; Pedestrians
• Subject Areas: Highways; Operations and Traffic Management; Pedestrians and Bicyclists; Safety and Human Factors;

Filing Info

• Accession Number: 01875979
• Record Type: Publication
• Files: TRIS
• Created Date: Mar 21 2023 9:17AM