中国安全科学学报 ›› 2022, Vol. 32 ›› Issue (8): 140-145.doi: 10.16265/j.cnki.issn1003-3033.2022.08.1768

• 安全工程技术 • 上一篇    下一篇

面向智能汽车预期功能安全的驾驶场景评价*

罗崎瑞1(), 张道文1,2, 周华1,2, 庞劭荣1, 李哓艳1, 王朝健1   

  1. 1 西华大学 汽车与交通学院,四川 成都 610039
    2 汽车测控与安全四川省重点实验室,四川 成都 610039
  • 收稿日期:2022-02-12 修回日期:2022-05-26 出版日期:2022-09-05 发布日期:2023-02-28
  • 作者简介:

    罗崎瑞 (1997—),男,四川广元人,硕士研究生,研究方向为智能交通安全,道路交通安全。E-mail:

    张道文,教授。

    周华,副研究员。

  • 基金资助:
    国家市场监督管理总局项目(202248); 四川省科技厅项目(212587); 四川省重点实验室课题(QCCK2021-011); 西华大学研究生创新基金资助(YCJJ2021152)

Evaluation on driving scenarios for safety of intended functionality of intelligent vehicles

LUO Qirui1(), ZHANG Daowen1,2, ZHOU Hua1,2, PANG Shaorong1, LI Xiaoyan1, WANG Chaojian1   

  1. 1 School of Automobile and Transportation, Xihua University, Chengdu Sichuan 610039, China
    2 Key Laboratory of Vehicle Measurement, Control and Safety of Sichuan Province, Chengdu Sichuan 610039, China
  • Received:2022-02-12 Revised:2022-05-26 Online:2022-09-05 Published:2023-02-28

摘要:

智能汽车对驾驶场景的感知能力直接影响到车辆的预期功能安全。为准确可靠地从感知层面进行场景评价,提出一种新的驾驶场景评价方法,该方法运用模糊推理,在分析场景和场景要素的基础上,首先,从驾驶人感知层面进行场景的模糊感知判断,建立场景模糊等级,利用模糊关系,建立场景要素间的联系;然后,按照决策函数计算场景的评价值;最后,验证该评价方法的合理性。结果表明:该方法可综合考虑驾驶人对场景的感知,更全面地认识场景,在场景评价应用方面具有可行性。

关键词: 智能汽车, 预期功能安全, 场景评价, 模糊推理, 自动驾驶

Abstract:

Safety of intended functionality of intelligent vehicles was directly affected by their sensing ability of driving scenarios. In order to evaluate scenarios accurately and reliably from a perceptual level, a new evaluation method was proposed. By using fuzzy reasoning and based on analysis of scenarios and elements, fuzzy judgment of scenarios was made from a perspective of drivers' perception, and their fuzzy level was established before correlation between scenario factors was developed by utilizing fuzzy relationship. Then, evaluation value was calculated according to decision function. In the end, rationality of the evaluation method was verified. The results show that this method could achieve a comprehensive understanding of scenarios by integrating driver's perception of them, thus proving to be feasible in scenario evaluation.

Key words: intelligent vehicle, safety of intended functionality, scenario evaluation, fuzzy reasoning, autonomous driving