基于通行规则模型的自动驾驶仿真场景构建研究

doi:10.19562/j.chinasae.qcgc.2025.08.001

摘要/Abstract

摘要：

本文提出一种面向自然语言道路通行规则的普适性和可扩展的描述模型，通过本体论方法将通行规则语言描述转化成逻辑命题，建立统一的基于设计运行范围、动态驾驶任务的道路通行规则要素层，构建匹配系统设计运行范围、动态驾驶任务的道路通行规则测试边界识别模型，提出基于道路通行规则描述模型的场景设计方法。选取两种自动驾驶算法进行测试验证，表明本文提出的道路通行规则场景设计方法，能够基于有限的测试用例实现高覆盖度的道路通行规则符合性测试，减少因被测系统与测试场景不匹配造成的无效测试，有效识别不符合道路通行规则的情形。

关键词: 自动驾驶, 测试场景, 仿真测试, 道路通行规则

Abstract:

In this paper， a universal and extensible description model for natural language road traffic rules is proposed， which transforms the language description of traffic rules into logical propositions through ontology methods. Then a unified road traffic rule element layer based on design operation domain and dynamic driving task is established. A road traffic rule test boundary recognition model that matches the system design operation domain and dynamic driving task is constructed， and a scenario design method based on the road traffic rule description model is proposed. The test verification of two automated driving systems shows that the road traffic rule scenario design method proposed in this paper can achieve high-coverage of road traffic rule compliance testing based on limited test cases， reduce invalid testing caused by mismatch between the system under test and the test scenario， and effectively identify situation that does not comply with road traffic rules.

Key words: automated driving, test scenarios, simulation test, road traffic rules

马明月,苗泽霖,王韦清,赵光明,王长君. 基于通行规则模型的自动驾驶仿真场景构建研究[J]. 汽车工程, 2025, 47(8): 1437-1447.

Mingyue Ma,Zelin Miao,Weiqing Wang,Guangming Zhao,Changjun Wang. Research on Construction of Autonomous Driving Simulation Scenario Based on the Traffic Rule Model[J]. Automotive Engineering, 2025, 47(8): 1437-1447.

图/表 9

图1

表1

算子定义及句法示例"

名称	符号	句法示例
否定	?	?φ_n
合取	?	φ_n ?φ_n_＋1
析取	?	φ_n ?φ_n_＋1
蕴含	→	P→C
下一时刻	Χ	Χω_n
直到	U	ω_nUω_n_＋1
总是	G	$G$ ω_n
最终	F	Fω_n
t₁~t₂内总是发生	G（t₁，t₂）	G（t₁，t₂）ω_n
t₁~t₂内会发生	F（t₁，t₂）	F（t₁，t₂）ω_n

表1

图2

图3

图4

图5

表2

MTL_TrafficRegulation81.01测试用例赋值"

要素

稳定行驶区

触发区

v_{Ego_0}/

（km·h^-1）

v_{Obj_0}/（km·h^-1）

d_{s_0}/m

$V i s i b i l i t y$ /

D_v i s i b i l i t y

$V i s i b i l i t y$ /

上限

120

120

100

200

下限

200

100

表2

表3

表4

B算法道路通行规则测试违规结果"

序号	违规项目	测试用例数量	违规率	违规情形
1	MTL_Trafficlaw42.02	10	100%	夜间道路行驶，未减速。
2	MTL_Trafficlaw42.04	5	100%	遇有沙尘、冰雹、雨、雪、雾、结冰等气象条件行驶时，未减速。
3	MTL_TrafficRegulation59.09	5	100%	通过事故发生路段，未减速慢行。
4	MTL_TrafficRegulation84.01	5	100%	通过施工作业路段，未减速行驶。
5	MTL_Trafficlaw42.03	5	100%	车辆平均行驶速度超过最高限速标线表明的速度。
6	MTL_TrafficRegulation81.01	5	100%	雾天在高速公路上行驶且能见度低于50 m，未按照规定速度行驶，且未能发出接管请求。
7	MTL_TrafficRegulation81.02	5	100%	雨天在高速公路上行驶且能见度低于50 m，未按照规定速度行驶，且未能发出接管请求。
8	MTL_TrafficRegulation81.03	5	100%	雾天在高速公路上行驶且能见度低于100 m，未按照规定速度行驶，且未能发出接管请求。
9	MTL_TrafficRegulation81.04	5	100%	雨天在高速公路上行驶且能见度低于100 m，未按照规定速度行驶，且未能发出接管请求。
10	MTL_Trafficlaw53.04	10	100%	后方有正在执行紧急任务的警车加速靠近，长时间未能让行，且未能发出接管请求。
11	MTL_Trafficlaw53.05	10	100%	后方有正在执行紧急任务的消防车加速靠近，长时间未能让行，且未能发出接管请求。
12	MTL_Trafficlaw53.06	10	100%	后方有正在执行紧急任务的救护车加速靠近，长时间未能让行，且未能发出接管请求。
13	MTL_Trafficlaw53.07	10	100%	后方有正在执行紧急任务的工程救险车加速靠近，长时间未能让行，且未能发出接管请求。
14	MTL_TrafficRegulation44.03	10	50%	车辆向左变道，未按照规定让行左侧相临车道后方车辆。
15	MTL_TrafficRegulation44.09	15	50%	车辆向右变道，未按照规定让行右侧相临车道后方车辆。
16	MTL_TrafficRegulation38.05	10	100%	车辆遇黄灯且已过停止线，停在路口中间，未按照规定继续行驶。
17	MTL_Trafficlaw38.11	10	100%	遇交通警察发出要求-靠边停车信号，未能停车，且未发出接管请求。
总计	17	135

表4

参考文献 26

[1]	北京市自动驾驶车辆道路测试报告（2023年）［R/OL］.北京智能车联产业创新中心，2024. http：//www.mzone.site/index.php/index/index/cid/2/sid/21.html.
	Road test report of self-driving vehicles in Beijing （2023）［R/OL］. Beijing Intelligent Vehicle Association Industry Innovation Center，2024. http：//www.mzone.site/index.php/index/index/cid/2/sid/21.html.
[2]	上海市智能网联汽车发展报告（2023年度）［R/OL］.上海市交通委员会，2024. https：//jtw.sh.gov.cn/zxzfxx/20240205/7aa0c16cc00e42cb9b3311b564dc8ffe.html.
	Report on the development of intelligent connected vehicles in Shanghai（2023）［R/OL］. Shanghai Municipal Transportation Commission， 2024. https：//jtw.sh.gov.cn/zxzfxx/20240205/7aa0c16cc00e42cb9b3311b564dc8ffe.html.
[3]	BUECHEL M， HINZ G， RUEHL F， et al. Ontology-based traffic scene modeling， traffic regulations dependent situational awareness and decision-making for automated vehicles［C］. 2017 IEEE Intelligent Vehicles Symposium （IV）. IEEE， 2017： 1471-1476.
[4]	工业和信息化部公安部住房和城乡建设部交通运输部关于开展智能网联汽车准入和上路通行试点工作的通知［EB/OL］. （2023-11-17）［2024-10-10］. https：//ythxxfb.miit.gov.cn/ythzxfwpt/hlwmh/tzgg/xzxk/clsczr/art/2023/art_ae314a1df30e4ba1a49bd4ffa10186f6.html.
	Notice of MIIT， MPS， MOHURD and MOT on carrying out pilot work on access and road traffic of intelligent connected vehicles［EB/OL］. （2023-11-17）［2024-10-10］. https：//ythxxfb.miit.gov.cn/ythzxfwpt/hlwmh/tzgg/xzxk/clsczr/art/2023/art_ae314a1df30e4ba1a49bd4ffa10186f6.html.
[5]	BENCH-CAPON T J M， ROBINSON G O， ROUTEN T W， et al. Logic programming for large scale applications in law： a formalisation of supplementary benefit legislation［C］. Proceedings of the lst International Conference on Artificial Intelligence and Law， 1987： 190-198.
[6]	RIZALDI A， ALTHOFF M. Formalising traffic rules for accountability of autonomous vehicles ［C］. 2015 IEEE 18th International Conference on Intelligent Transportation Systems. IEEE， 2015：1658-1665.
[7]	MAIERHOFER S， RETTINGER A K， MAYER E C， et al. Formalization of intersection traffic rules in temporal logic［C］. 2020 IEEE Intelligent Vehicles Symposium （IV）. IEEE， 2020： 752-759.
[8]	YU W， ZHAO C， WANG H， et al. Online legal driving behavior monitoring for self-driving vehicles［J］. Nature Communications， 2024， 15（1）.
[9]	BECKER C， BREWER J C， YOUNT L. Safety of the intended functionality of lane-centering and lane-changing maneuvers of a generic level 3 highway chauffeur system［R］. National Highway Traffic Safety Administration， 2020.
[10]	JACOBO A， NOBUYUKI U， KUNIO Y， et al. Development of a safety assurance process for autonomous vehicles in Japan［C］. Proceedings of ESV Conference， 2019.
[11]	ZHAO D， HUANG X， PENG H， et al. Accelerated evaluation of automated vehicles in car-following maneuvers［J］. IEEE Transactions on Intelligent Transportation Systems， 2017， 19（3）：733-744.
[12]	FENG S， FENG Y， YU C， et al. Testing scenario library generation for connected and automated vehicles， part I： methodology［J］. IEEE Transactions on Intelligent Transportation Systems， 2020， 22（3）：1573-1582.
[13]	SUN J， ZHOU H， XI H， et al. Adaptive design of experiments for safety evaluation of automated vehicles［J］. IEEE Transactions on Intelligent Transportation Systems， 2021， 23（9）：14497-14508.
[14]	DING W， CHEN B， LI B， et al. Multimodal safety-critical scenarios generation for decision-making algorithms evaluation［J］. IEEE Robotics and Automation Letters， 2021，6（2）：1551-1558.
[15]	XIAO Y， ZHANG X， XU X， et al. Deep neural networks with Koopman operators for modeling and control of autonomous vehicles［J］. IEEE Transactions on Intelligent Vehicles， 2022， 8（1）： 135-146.
[16]	ISO. Road vehicles-scenario attributes and categorization： ISO 34504 ［S］. 2018.
[17]	European Commission. Commission Implementing Regulation （EU） 2022/1426 of 5 August 2022 laying down rules for the application of Regulation （EU） 2019/2144 of the European Parliament and of the Council as regards uniform procedures and technical specifications for the type-approval of the automated driving system （ADS） of fully automated vehicles［Z］. Off. J. Eur. Union， 2022， 65（L221）：1-64.
[18]	United Nations Economic Commission for Europe Geneva CH Regulation Addendum. New assessment/test method for automated driving （NATM） guidelines for validating automated driving system （ADS）［Z］. 2022.
[19]	United Nations Economic Commission for Europe Geneva CH Regulation Addendum. UN regulation No 157-Uniform provisions concerning the approval of vehicles with regards to automated lane keeping systems ［Z］. 2022.
[20]	LI Y， WU S， et al. Adaptive mining of failure scenarios for autonomous driving systems based on multi-population genetic algorithm［C］. 2024 IEEE Intelligent Vehicles Symposium （IV）， Jeju Island， Korea， Republic of， 2024： 2458-2464.
[21]	WU S， WANG H， et al. A new SOTIF scenario hierarchy and its critical test case generation based on potential risk assessment［C］. 2021 IEEE 1st International Conference on Digital Twins and Parallel Intelligence （DTPI）， Beijing， China， 2021：399-409.
[22]	CHANG C， et al. MetaScenario： a framework for driving scenario data description， storage and indexing［J］. IEEE Transactions on Intelligent Vehicles， 2023，8（2）： 1156-1175.
[23]	ISO. Road vehicles-engineering framework and process of scenario-based safety evaluation： ISO 34502 ［S］. 2018.
[24]	GRUBER T R. A translation approach to portable ontology specifications［J］. Knowledge Acquisition，1993，5：199-220.
[25]	WAN L， WANG C， et al. Semantic consistency and correctness verification of digital traffic rules［J］. Engineering， 2024， 33：47-62.
[26]	全国汽车标准化技术委员会. 汽车驾驶自动化分级： GB/T 40429—2021［S/OL］．［2024-10-10］. https：//openstd.samr.gov.cn/bzgk/gb/newGbInfo？hcno=4754CB1B7AD798F288C52D