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

Autor: Gong T; College of Automotive Engineering, Jilin University, Changchun, 130022, Jilin, China; National key laboratory of automotive chassis integration and bionics, Jilin University, Changchun, 130022, Jilin, China. Electronic address: gongty22@mails.jlu.edu.cn., Yu X; College of Automotive Engineering, Jilin University, Changchun, 130022, Jilin, China; National key laboratory of automotive chassis integration and bionics, Jilin University, Changchun, 130022, Jilin, China. Electronic address: yuxm@jlu.edu.cn., Zhang Q; Huawei Technologies Co Ltd, Huawei 2012 Research Center, Shenzhen, 518000, Guangdong, China. Electronic address: zhangqunli1@huawei.com., Feng Z; Huawei Technologies Co Ltd, IAS BU Technique Development Department, Shenzhen, 518000, Guangdong, China. Electronic address: fengzilin@huawei.com., Yang S; School of Transportation Science and Engineering, Beihang University, BeiJing, 100091, BeiJing, China. Electronic address: yangshichun@buaa.edu.cn., Cao Y; Hangzhou International Innovation Institute, Beihang University, Hangzhou, 310051, Zhejiang, China; State Key Lab of Intelligent Transportation System, Beihang University, BeiJing, 100091, BeiJing, China. Electronic address: caoyaoguang@buaa.edu.cn., Xu J; School of Transportation Science and Engineering, Beihang University, BeiJing, 100091, BeiJing, China. Electronic address: ZF2213124@buaa.edu.cn., Feng X; School of Transportation Science and Engineering, Beihang University, BeiJing, 100091, BeiJing, China. Electronic address: bhfengxinjie@buaa.edu.cn., Pang Z; School of Information and Communication Engineering, Hainan University, Haikou, 570228, Hainan, China. Electronic address: pangzhaowen@hainanu.edu.cn., Wang Y; School of Transportation Science and Engineering, Beihang University, BeiJing, 100091, BeiJing, China. Electronic address: zy2343306@buaa.edu.cn., Wang P; Huawei Technologies Co Ltd, Huawei 2012 Research Center, Shenzhen, 518000, Guangdong, China. Electronic address: wangpeng394@huawei.com.
Jazyk: angličtina
Zdroj: Accident; analysis and prevention [Accid Anal Prev] 2024 Nov 23; Vol. 210, pp. 107842. Date of Electronic Publication: 2024 Nov 23.
DOI: 10.1016/j.aap.2024.107842
Abstrakt: 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.
Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
(Copyright © 2024 Elsevier Ltd. All rights reserved.)
Databáze: MEDLINE
Externí odkaz: