自动驾驶汽车连续测试场景复杂度评估方法  

Continuous test scenario complexity evaluation method for automated driving vehicles

作　　者：朱冰[1] 范天昕 赵文博 李伟男 张培兴 ZHU Bing;FAN Tian-xin;ZHAO Wen-bo;LI Wei-nan;ZHANG Pei-xing(National Key Laboratory of Automotive Chassis Integration and Bionics,Jilin University,Changchun 130022,China;China Intelligent and Connected Vehicles(Beijing)Research Institute Co.,Ltd.,Beijing 100176,China;China First Automobile Corporation R&D Center,Changchun 13001l,China)

机构地区：[1]吉林大学汽车底盘集成与仿生全国重点实验室,长春130022 [2]国汽(北京)智能网联汽车研究院有限公司,北京100176 [3]中国第一汽车集团股份有限公司研发总院,长春130011

出　　处：《吉林大学学报(工学版)》2025年第2期456-467,共12页Journal of Jilin University:Engineering and Technology Edition

基　　金：国家自然科学基金项目(U22A20247,5240120450);中国博士后科学基金项目(2023M741354);国家资助博士后研究人员计划项目(GZC20230945)。

摘　　要：连续测试场景是自动驾驶汽车测试体系的重要内容,其测试公平性一直受到较多的争议。为此,本文提出一种自动驾驶汽车连续测试场景复杂度评估方法,以期解决测试场景公平性难题。基于六层场景模型建立场景要素重要性评估体系;分析场景要素与感知、决策、执行系统间的复杂度映射关系,建立自动驾驶系统层级的场景复杂度量化评价方法;建立系统影响传递权重系数,实现场景复杂度综合计算。搭建仿真交通环境对两种自动驾驶系统行驶过程复杂度进行计算,两车场景复杂度对比结果分别为1和0.765,与测试过程被测算法遭遇的场景难度趋势一致,证明了本文方法的有效性。Continuous test scenario is an important content of automated driving vehicle scenario-basedtesting system,but its fairness has been disputed.For this reason,this paper proposes a real-timeevaluation method for the continuous test scenario complexity of automated driving vehicle,with a view tosolving the problem of test scenario fairness.Based on the six-layer scenario model,a scenario elementimportance assessment system is established;through analyzing the mapping relationship between scenarioelements and perception,decision-making,and execution systems,a quantitative evaluation method ofscenario complexity at the system layer is constructed;the impact transfer weighting coefficients ofdifferent system are calculated,which can be used to realize the comprehensive calculation of scenariocomplexity in real time.The traffic circle continuous test scenario is selected to analyze the scenariocomplexity of two black-box automated driving systems,the results of the scenario complexity of the twovehicles are 1 and 0.765,which are consistent with the scenario complexity encountered by the two testedsystems in the test process.The results can prove the effectiveness of the method proposed in this paper.

关 键 词：车辆工程 自动驾驶汽车 连续测试场景 场景复杂度 

分 类 号：U461[机械工程—车辆工程]