欢迎访问《汽车安全与节能学报》，
汽车安全与节能学报

汽车安全与节能学报 ›› 2025, Vol. 16 ›› Issue (3): 386-395.DOI: 10.3969/j.issn.1674-8484.2025.03.004

• 汽车安全 • 上一篇    下一篇

基于自组织K-means的城市道路VRU事故场景复杂度评价

程瑞1(), 卢春成1(), 袁泉2,*(), 崔涛3, 王涛1   

  1. 1.桂林电子科技大学，广西智慧交通重点实验室，桂林 541004，中国
    2.清华大学 车辆与运载学院，智能绿色车辆与交通全国重点实验室，北京 100084，中国
    3.Mercedes-Benz Group China Ltd.，北京 100102，中国
  • 收稿日期:2024-09-22 修回日期:2025-01-04 出版日期:2025-06-30 发布日期:2025-07-01
  • 通讯作者: 袁泉，教授级高工。E-mail：yuanq@tsinghua.edu.cn。
  • 作者简介:程瑞（1992—），男（汉），山东，副教授。E-mail：ruicheng1992@yeah.net。
    卢春成（2000—），男（汉），广西，硕士研究生。E-mail：458291953@qq.com。
  • 基金资助:
    国家自然科学基金项目(52072214);清华大学-梅赛德斯奔驰可持续交通联合研究院项目;广西自然科学基金项目(2023GXNSFAA026359);桂林市科学研究与技术开发计划资助项目(20230120-7)

Evaluation on the complexity of scenarios for VRU on urban roads based on self-organizing K-means

CHENG Rui1(), LU Chuncheng1(), YUAN Quan2,*(), CUI Tao3, To. Jeremy3, WANG Tao1   

  1. 1. Guangxi Key Laboratory of ITS, Guilin University of Electronic Technology, Guilin 541004, China
    2. School of Vehicle and Mobility, State Key Laboratory of Intelligent Green Vehicle and Mobility, Tsinghua University, Beijing 100084, China
    3. Mercedes-Benz Group China Ltd., Beijing 100102, China
  • Received:2024-09-22 Revised:2025-01-04 Online:2025-06-30 Published:2025-07-01

RichHTML

2

PDF

158

摘要：

为了满足智能汽车避撞系统验证中高风险测试环境的需要，同时丰富面向弱势道路使用者（VRU）的自动驾驶场景评价内容和方法，该文通过对广西桂林市2016—2020年交通事故案例收集整理，筛选得到1 429例汽车与VRU碰撞事故数据；依据事故调查经验选取了13种风险因素，基于自组织K-means聚类分析构建了10类适用于中国城市交通状况的汽车与VRU碰撞的典型场景；利用信息熵理论建立了VRU典型场景复杂度评价模型，通过联合logistic模型与反向神经（BP）网络确定变量状态及各维度权重，计算得到各类场景复杂度；运用Guass混合模型对复杂度进行聚类，最终获得4个场景复杂度等级。 结果表明：在限速30 km/h的道路上，夜间直行汽车与横穿马路的电动自行车在非人行横道区域发生侧面碰撞的场景复杂度最高。该文的研究成果可为智能汽车安全性测试提供具备中国城市道路特征的实验场景，同时为车外VRU避撞方案和决策的制定提供一定的依据。

关键词: 弱势道路使用者（VRU）, 智能汽车, 典型场景, 自组织K-means聚类分析

Abstract:

In order to address the requirements of high-risk testing environments for validating intelligent vehicle collision avoidance systems, while simultaneously to enrich the content and methods for evaluating autonomous driving scenarios involving vulnerable road users (VRU). This study collected and systematically analyzed traffic accident cases in Guilin City, Guangxi Province, from 2016 to 2020. A total of 1 429 vehicle-VRU collision accident data were screened. Based on accident investigation experience, 13 risk factors were identified, and 10 typical vehicle-VRU collision scenarios applicable to urban traffic conditions in China were constructed using self-organizing K-means clustering analysis. An evaluation model for the complexity of VRU scenarios was established utilizing information entropy theory. The state of variables and the weight of each dimension were determined through a combination of logistic regression models and back propagation (BP) neural networks, and the complexity of various scenarios was calculated. Additionally, the Gaussian mixture model was employed to cluster the complexity levels, resulting in four distinct scene complexity categories. The results show that on roads with a speed limit of 30 km/h, the nighttime side collision between a straight-moving car and an electric bicycle crossing the road outside a pedestrian crossing area is the most complex scenario. The findings in this study provide an experimental scenario reflective of urban road characteristics in China for intelligent vehicle safety testing and offer a basis for the formulation of external VRU collision avoidance strategies and decision-making.

Key words: vulnerable road users (VRU), intelligent vehicles, typical hazardous scenarios, self-organizing, k-means clustering analysis

中图分类号: