An emergency operation strategy and motion planning method for autonomous vehicle in emergency scenarios

Ensuring driving operational safety in emergency scenarios is paramount for autonomous vehicles to prevent accidents, particularly when vehicle motion completely depends on autonomous systems. Numerous factors must be evaluated when designing emergency collision avoidance strategies for critical situations, such as trajectory feasibility, vehicle motion stability, and driver comfort. Therefore, this study proposes a framework for emergency operation that uses collision-free area calculations to inform maneuver decisions and facilitate collision avoidance trajectory planning, preventing vehicle collisions. In case of danger, the emergency maneuver decision module evaluates the safety level and selects safety terminal state by considering a pre-specified cluster of candidate maneuvers before generating trajectories. This process avoids infeasible trajectories and selects maneuvers for greater driver comfort when available. Subsequently, the dynamic trajectory planning module converts the collision-free area into mixed-integer constraints, utilizing time-varying Nonlinear Model Predictive Control (NMPC) for trajectory planning and ensuring vehicle motion stability by integrating dynamic and collision-free constraints throughout the motion planning process. Eventually, simulations and field testing validate the framework’s effectiveness, mitigating collisions in emergency scenarios with prompt and safe operations. The framework is designed to function autonomously, independent of the intelligent driving system, engaging only during risk events and restoring control to the driver or the intelligent system after the event.

• Record URL:
  • https://doi.org/10.1016/j.aap.2024.107842
• Record URL:
  • http://www.sciencedirect.com/science/article/pii/S0001457524003877
• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/issn/00014575
• Supplemental Notes:
  • © 2024 Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies. Abstract reprinted with permission of Elsevier.
• Authors:
  • Gong, Tianyang
  • 0000-0002-1116-7178
  • Yu, Xiumin
  • Zhang, Qunli
  • Feng, Zilin
  • Yang, Shichun
  • Cao, Yaoguang
  • Xu, Jingyun
  • Feng, Xinjie
  • Pang, Zhaowen
  • Wang, Yu
  • Wang, Peng
• Publication Date: 2025-2

Language

• English

Media Info

• Media Type: Web
• Features: References;
• Pagination: 107842
• Serial:
  • Accident Analysis & Prevention
  • Volume: 210
  • Issue Number: 0
  • Publisher: Elsevier
  • ISSN: 0001-4575
  • Serial URL: http://www.sciencedirect.com/science/journal/00014575

Subject/Index Terms

• TRT Terms: Autonomous vehicles; Crash avoidance systems; Emergencies; Trajectory control; Vehicle design; Vehicle trajectories
• Subject Areas: Highways; Safety and Human Factors; Security and Emergencies; Vehicles and Equipment;

Filing Info

• Accession Number: 01939345
• Record Type: Publication
• Files: TRIS
• Created Date: Dec 12 2024 4:58PM