智能网联汽车数字孪生测试关键场景提取和识别  

Test-based extraction and identification of key scenarios for digital twins of intelligent networked vehicles

作　　者：祖晖 龙洋[3] 韩庆文[3] 王勇[3] 曾令秋[3] 陈旭[2] 张迪思 卓玺 ZU Hui;LONG Yang;HAN Qingwen;WANG Yong;ZENG Lingqiu;CHEN Xu;ZHANG Disi;ZHUO Xi(China Merchants Testing Vehicle Technology Research Institute Co.,Ltd.,Chongqing 401122,China;Chongqing University of Technology,Chongqing 400054,China;Chongqing University,Chongqing 400044,China;Chongqing High-tech Zone Urban Construction Affairs Center,Chongqing 402365,China)

机构地区：[1]招商局检测车辆技术研究院有限公司,重庆401122 [2]重庆理工大学,重庆400054 [3]重庆大学,重庆400044 [4]重庆高新区城市建设事务中心,重庆402365

出　　处：《重庆理工大学学报（自然科学）》2023年第1期75-84,共10页Journal of Chongqing University of Technology：Natural Science

基　　金：国家自然科学基金项目(62172066,U21A20448);重庆市自动驾驶系统与智能网联汽车研究与测试工程技术研究中心资助项目(20AKC17)。

摘　　要：场景生成是智能网联汽车数字孪生(DT)测试面临的关键问题之一,场景的典型性是决定测试有效性的关键。智能网联汽车的测试场景源自真实车辆行驶数据,提出了一种DT测试场景生成方法,基于路侧雷达采集的局部道路车辆行驶数据提取典型测试场景,以FCW、LCW和ICW 3种典型应用为基础,建立基于碰撞风险因素和交通质量因素的场景典型性评价方法,构建LSTM-AE-Attention模型实现典型场景识别。实验结果表明,提出的方法能够有效评价场景典型性,并有效识别典型场景,为测试场景库的构建提供了有效支撑。Scenario generation is one of the key problems in digital twin(DT) technology, and the typicality of scenarios is the key to test effectiveness. The test scenarios of an intelligent networked vehicle are derived from real vehicle driving data. This paper proposes a new DT test scene generation method which extracts typical test scenes based on local road vehicle driving data collected by roadside radar, establishes a typical scene evaluation method of collision risk factors and traffic quality factors on the basis of the three typical applications of FCW, LCW and ICW, and builds an LSTM-AE-Attention model to identify these critical scenarios. The experimental results show that the constructed model can effectively evaluate and identify typical scenes, which provides effective support for the construction of the test scene library.

关 键 词：数字孪生 典型场景 识别 

分 类 号：TN92[电子电信—通信与信息系统]