自动驾驶发展现状及趋势
Current Status and Development Trends of Autonomous Driving
当前全球自动驾驶技术正沿着欧美主导的“单车智能 ‒ 车云协同”与中国特色的“车路协同 ‒ 车路云一体化”两条主流路径演进,两者各有特色和优势。本文系统梳理了道路交通自动驾驶技术的发展脉络及产业进程,指出该技术正从基于规则和深度学习的传统方法向世界模型驱动的方向演进,并在端到端模型及视觉 ‒ 语言 ‒ 动作融合推动下,迈向认知智能的新时代。同时,我国凭借系统优势,构建了“车路云一体化”的总体架构,初步形成了适应中国国情的智能交通解决方案。未来,随着大语言模型、生成式人工智能等新兴技术与“车路云一体化”深度融合,自动驾驶将进入自主智能与人机协同新阶段。最后,从科技创新、教育、国家政策、国际视野、产业现状与发展趋势多个维度,对自动驾驶的关键问题与未来发展路径进行分析,并提出了强化核心技术自主创新、加速基础设施智能化建设、完善产业支撑体系、构建面向产业发展的社会机制等发展建议。
Currently, autonomous driving is advancing along two mainstream paths: the "individual-vehicle intelligence, vehicle‒cloud collaboration" path led by Europe and the United States, and the "vehicle-to-infrastructure cooperation, vehicle‒road‒cloud integration" path with Chinese characteristics. Both paths have their district features and advantages. This study outlines the development trajectory and industrial progress of autonomous driving, highlighting its evolution from conventional approaches based on rules and deep learning toward a direction driven by world models. Propelled by end-to-end models and vision‒language‒action integration, autonomous driving is advancing into a new era of cognitive intelligence. Leveraging its system advantages, China has established an overall architecture for vehicle‒road‒cloud integration, preliminarily forming an intelligent transportation solution tailored to China's national conditions. Looking ahead, with the deep integration of the vehicle‒road‒cloud architecture with emerging technologies such as large language models and generative artificial intelligence, autonomous driving will enter a new stage characterized by autonomous intelligence and human‒machine collaboration. Furthermore, the study analyzes the key issues and future development paths for autonomous driving from the dimensions of technological innovation, education, national policies, international perspectives, industrial status, and development trends, and proposes recommendations including strengthening independent innovation in core technologies, accelerating the intelligent development of infrastructure, improving the industrial support system, and establishing social mechanisms for industrial development.
| [1] |
中华人民共和国交通运输部. 关于“人工智能+交通运输”的实施意见[EB/OL].(2025-09-25)[2026-02-07]. https://xxgk.mot.gov.cn/2020/jigou/kjs/202509/t20250925_4177256.html. |
| [2] |
Ministry of Transport of the People’s Republic of China. Implementation opinions on “artificial intelligence + transportation”[EB/OL].(2025-09-25)[2026-02-07]. https://xxgk.mot.gov.cn/2020/jigou/kjs/202509/t20250925_4177256.html. |
| [3] |
李克强. 新一代AI驱动的中国方案:智能网联汽车关键技术创新与应用[J]. 中国经济报告,2024(4):141-142. |
| [4] |
Li K Q. A new generation of AI-driven China scheme:Innovation and application of key technologies of intelligent networked vehicles[J]. China Policy Review,2024(4):141-142. |
| [5] |
冉斌,谭华春,张健, 智能网联交通技术发展现状及趋势[J]. 汽车安全与节能学报,2018,9(2):119-130. |
| [6] |
Ran B,Tan H C,Zhang J,et al. Development status and trend of connected automated vehicle highway system[J]. Journal of Automotive Safety and Energy,2018,9(2):119-130. |
| [7] |
李升波,刘畅,殷玉明, 汽车端到端自动驾驶系统的关键技术与发展趋势[J]. 人工智能,2023(5):1-16. |
| [8] |
Li S B,Liu C,Yin Y M,et al. Key technology and development trend of end-to-end auto-driving system[J]. AI-View,2023(5):1-16. |
| [9] |
李升波,江昆,田野, 汽车智能驾驶技术发展与趋势展望[J]. 前瞻科技,2025,4(2):144-157. |
| [10] |
Li S B,Jiang K,Tian Y,et al. Development and outlook of intelligent driving technology[J]. Science and Technology Foresinght,2025,4(2):144-157. |
| [11] |
李克强,戴一凡,李升波, 智能网联汽车(ICV)技术的发展现状及趋势[J]. 汽车安全与节能学报,2017,8(1):1-14. |
| [12] |
Li K Q,Dai Y F,Li S B,et al. State-of-the-art and technical trends of intelligent and connected vehicles[J]. Journal of Automotive Safety and Energy,2017,8(1):1-14. |
| [13] |
GB/T 40429—2021 汽车驾驶自动化分级[S]. |
| [14] |
GB/T 40429—2021 Automated driving levels[S]. |
| [15] |
中国汽车工程学会,国家智能网联汽车创新中心. 中国智能网联汽车产业发展报告(2022)[M]. 北京:社会科学文献出版社,2023. |
| [16] |
China Society of Automotive Engineers,National Intelligent Connected Vehicle Innovation Center. China’s intelligent connected vehicle industry development report(2022)[M]. Beijing:Social Sciences Academic Press,2023. |
| [17] |
李克强,李家文,常雪阳, 智能网联汽车云控系统原理及其典型应用[J]. 汽车安全与节能学报,2020,11(3):261-275. |
| [18] |
Li K Q,Li J W,Chang X Y,et al. Principles and typical applications of cloud control system for intelligent and connected vehicles[J]. Journal of Automotive Safety and Energy,2020,11(3):261-275. |
| [19] |
Hu Y H,Yang J Z,Chen L,et al. Planning-oriented autonomous driving[C]// 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR),2023:17853-17862. |
| [20] |
Chib P S,Singh P. Recent advancements in end-to-end autonomous driving using deep learning:A survey[J]. IEEE Transactions on Intelligent Vehicles,2024,9(1):103-118. |
| [21] |
Wu J D,Huang C,Huang H L,et al. Recent advances in reinforcement learning-based autonomous driving behavior planning:A survey[J]. Transportation Research Part C:Emerging Technologies,2024,164:104654. |
| [22] |
Cui C,Ma Y S,Cao X,et al. A survey on multimodal large language models for autonomous driving[C]// 2024 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops(WACVW),2024:958-979. |
| [23] |
Wang Y Q,He J W,Fan L,et al. Driving into the future:Multiview visual forecasting and planning with world model for autonomous driving[C]// 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR),2024:14749-14759. |
| [24] |
Wu D M,Han W C,Liu Y F,et al. Language prompt for autonomous driving[J]. Proceedings of the AAAI Conference on Artificial Intelligence,2025,39(8):8359-8367. |
| [25] |
Ganesan M,Kandhasamy S,Chokkalingam B,et al. A comprehensive review on deep learning-based motion planning and end-to-end learning for self-driving vehicle[J]. IEEE Access,2024,12:66031-66067. |
| [26] |
Shao H,Hu Y X,Wang L T,et al. LMDrive:Closed-loop end-to-end driving with large language models[C]// 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR),2024:15120-15130. |
| [27] |
Zheng W Z,Song R Q,Guo X D,et al. GenAD:generative end-to-end autonomous driving[C]// Computer Vision—ECCV 2024,2024:87-104. |
| [28] |
张文涛,孙奥兰,瞿晓阳, 面向具身智能的视觉 ‒ 语言 ‒ 动作模型动作表征和生成策略综述[J/OL]. 计算机应用,2025-09-30. https://kns.cnki.net/KCMS/detail/detail.aspx?filename=JSJY20250929006&dbname=CJFD&dbcode=CJFQ. |
| [29] |
Zhang W T,Sun A L,Qu X Y,et al. Survey on action representation and generation strategies in Vision-Language-Action models for embodied intelligence[J/OL]. Journal of Computer Applications,2025-09-30. https://kns.cnki.net/KCMS/detail/detail.aspx?filename=JSJY20250929006&dbname=CJFD&dbcode=CJFQ. |
| [30] |
Wang X F,Zhu Z,Huang G,et al. DriveDreamer:towards real-world-drive world models for Autonomous driving[C]// Computer Vision—ECCV 2024,2024:55-72. |
| [31] |
Jia X S,Yan Z J,Li Q F,et al. Bench2Drive:Towards multi-ability benchmarking of closed-loop end-to-end autonomous driving[C]// Advances in Neural Information Processing Systems 37,2024:819-844. |
| [32] |
Pan C B,Yaman B,Nesti T,et al. VLP:Vision language planning for autonomous driving[C]// 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR),2024:14760-14769. |
| [33] |
Zhou X C,Liu M Y,Yurtsever E,et al. Vision language models in autonomous driving:A survey and outlook[J]. IEEE Transactions on Intelligent Vehicles,2024(99):1-20. |
| [34] |
Wen Y Q,Zhao Y C,Liu Y F,et al. Panacea:Panoramic and controllable video generation for autonomous driving[C]// 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR),2024:6902-6912. |
| [35] |
Jia X S,Wu P H,Chen L,et al. Think twice before driving:Towards scalable decoders for end-to-end autonomous driving[C]// 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR),2023:21983-21994. |
| [36] |
丁飞,张楠,李升波, 智能网联车路云协同系统架构与关键技术研究综述[J]. 自动化学报,2022,48(12):2863-2885. |
| [37] |
Ding F,Zhang N,Li S B,et al. A survey of architecture and key technologies of intelligent connected vehicle-road-cloud cooperation system[J]. Acta Automatica Sinica,2022,48(12):2863-2885. |
| [38] |
Guerra J,Luís M,Rito P. Performance evaluation of 5G new radio V2X sidelink for coexisting traffic[J]. IEEE Access,2025,13:131400-131410. |
| [39] |
Feng Z Y,Zhang J,Jiang H,et al. Energy consumption prediction strategy for electric vehicle based on LSTM-transformer framework[J]. Energy,2024,302:131780. |
| [40] |
Jiang X,Zhang J,Shi X Y,et al. Learning the policy for mixed electric platoon control of automated and human-driven vehicles at signalized intersection:A random search approach[J]. IEEE Transactions on Intelligent Transportation Systems,2023,24(5):5131-5143. |
| [41] |
中国智能网联汽车产业创新联盟. 车路云一体化系统白皮书[R/OL]. 2023-01:1-2. https://www.cictci.com/uploads/20240925/ce9f98ae83c6e4353ddd2fbceec758fe.pdf. |
| [42] |
China Intelligent and Connected Vehicle Industry Innovation Alliance. White paper on integrated vehicle‒road‒cloud systems[R/OL]. 2023-01:1-2. https://www.cictci.com/uploads/20240925/ce9f98ae83c6e4353ddd2fbceec758fe.pdf. |
| [43] |
Liu F,Lu Z H,Lin X K. Vision-based environmental perception for autonomous driving[PP/OL]. arXiv(2022-12-22)[2025-10-19]. https://doi.org/10.48550/arXiv.2212.11453. |
| [44] |
Ruan S L,Wang R W,Shen X C,et al. A survey of multi-sensor fusion perception for embodied AI:Background,methods,challenges and prospects[PP/OL]. arXiv(2025-06-24)[2025-10-19]. https://doi.org/10.48550/arXiv.2506.19769. |
| [45] |
Zuo Q,Li C Z H,Fan Y L,et al. Advanced multi-modal sensor fusion architectures for robust autonomous driving systems[C]// 2025 IEEE 5th International Conference on Electronic Technology,Communication and Information(ICETCI),2025:1556-1560. |
中国工程院咨询项目“新兴技术治理模式研究”(2025-GDZK-14)
工业和信息化部项目“面向车联网(智能网联汽车)领域的公共服务平台建设”(CEIEC-2020-ZM02-0100)
/
| 〈 |
|
〉 |