欢迎访问《汽车安全与节能学报》,

汽车安全与节能学报 ›› 2023, Vol. 14 ›› Issue (3): 249-273.DOI: 10.3969/j.issn.1674-8484.2023.03.001

• 综述与展望 •    下一篇

面向自动驾驶的车路云一体化框架

张亚勤1(), 李震宇2, 尚国斌2, 周谷越1, 高果荣2, 袁基睿1   

  1. 1.清华大学智能产业研究院,北京 100084,中国
    2.阿波罗智联(北京)科技有限公司,北京 100084,中国
  • 收稿日期:2023-04-10 修回日期:2023-06-26 出版日期:2023-06-30 发布日期:2023-06-29
  • 作者简介:张亚勤(1966—),男(汉),山西,教授。E-mail: zhangyaqin@air.tsinghua.edu.cn
    张亚勤 博士
    中国工程院外籍院士,美国艺术与科学院院士,澳大利亚国家工程院院士(外籍),中国人工智能学会会士。现任清华大学智能科学讲席教授,清华大学智能产业研究院院长。他31岁时被授予电气与电子工程师协会会士(Institute of Electrical and Electronics Engineers Fellow,IEEE Fellow),成为历史上获得这一荣誉最年轻的科学家,并于2004年获得IEEE技术先锋奖。他曾在微软公司工作16年,历任全球资深副总裁兼微软亚太研发集团主席、微软亚洲研究院院长兼首席科学家、微软全球副总裁和微软中国董事长;2014年9月―2019年10月担任百度公司总裁。
    他是数字视频和人工智能领域的世界级科学家和企业家,拥有60多项美国专利,发表500多篇学术论文,并出版11本专著。他发明的多项图像视频压缩和传输技术被国际标准采用,广泛应用于高清电视、互联网视频、多媒体检索、移动视频和图像数据库领域。
    他担任世界经济论坛达沃斯“人工智能委员会”委员、“未来交通指导委员会”委员,以及全球最大自动驾驶技术开放平台Apollo联盟理事长和联合国计划发展总署(United Nation Development Program, UNDP)企业董事会董事。他还在十余所世界顶尖高校担任校董、荣誉或客座教授,并在4家高科技公司担任董事。
    Dr. YA-QIN ZHANG
    Dr. Ya-Qin Zhang is Chair Professor of AI Science at Tsinghua University, and Dean of Institute for AI Industry Research of Tsinghua University (AIR). He was the President of Baidu Inc. from 2014 to 2019. Prior to Baidu, Dr. Zhang was a Microsoft executive for 16 years with different key positions, including Managing Director of Microsoft Research Asia, Chairman of Microsoft China, and Corporate Vice President and Chairman of Microsoft Asia R&D.
    Dr. Zhang was elected to the Chinese Academy of Engineering (CAE), the American Academy of Arts and Sciences (AAA&S), the Australian Academy of Technology and Engineering (ATSE). He is a Fellow of Institute of Electrical and Electronics Engineers (IEEE) and Chinese Association for Artificial Intelligence (CAAI). He is one of the top scientists and technologists in digital video and AI, with over 500 papers, 60 granted US patents, and 11 books. His original research has become the basis for start-up ventures, new products, and international standards in digital video, cloud computing, and autonomous driving.
    He serves on the board of directors of four public companies. He is on the industry board of United Nation Development Program (UNDP), and AI global council of the Davos World Economic Forum. He is the Chairman of world’s largest open autonomous driving platform “Apolloalliance with over 200 global partners. He has been an active speaker in global forums including Asia-Pacific Economic Cooperation (APEC), Davos, United Nations, and Bo’Ao Asia Forum.

A unified framework for vehicle-infrastructure-cloud autonomous driving

ZHANG Yaqin1(), LI Zhenyu2, SHANG Guobin2, ZHOU Guyue1, GAO Guorong2, YUAN Jirui1   

  1. 1. Institute for AI Industry Research, Tsinghua University, Beijing 100084, China
    2. Apollo Intelligent Connectivity (Beijing) Technology Co., Ltd, Beijing 100084, China
  • Received:2023-04-10 Revised:2023-06-26 Online:2023-06-30 Published:2023-06-29

摘要:

近年来,尽管自动驾驶(AD)技术在研发和商业化方面已取得显著进展,但自动驾驶规模化商业落地仍面临巨大挑战:一方面单车智能自动驾驶存在一定的安全问题;另一方面由开放道路场景引发的感知长尾、混行博弈等问题造成自动驾驶车辆的可运行设计域(ODD)受限。车辆亟需融合车端、路端、云端多源多维信息,进行一体化协同感知、协同决策、协同控制,拓展单车智能自动驾驶的能力边界。该文提出面向自动驾驶的车路云一体化(VICAD)系统框架,将不同的车路云协同部署策略与自动驾驶算法进行统一建模;在此基础上开展模拟仿真及系统评价,利用评价结果反馈对VICAD系统进行持续迭代优化,从而提升自动驾驶能力。此外,还进一步结合场景案例及产业落地实践,阐述了车路云一体化协同对自动驾驶大规模商业化落地的意义,并为VICAD的后续发展提出建议。

关键词: 自动驾驶(AD), 车路协同, 预期功能安全, 车路云一体化自动驾驶(VICAD), 可运行设计域(ODD)

Abstract:

While major advances have been made in the R&D and commercialization of autonomous driving (AD) in the past decade, there still exists significant challenges in the large-scale commercial deployment of AD in complex open-road scenarios, such as longtail perception problem and limited operational design domain (ODD). Information from vehicles, traffic and the underlying infrastructure (V2X) can be used to enhance the overall system safety and accelerate the deployment, with integration of multi-scaled, multi-dimensional diverse sources. This integration would enable cooperated perception, decision-making, and control, expanding single-vehicle intelligence’s capability boundaries. By using this combined knowledge, some of the obstacles encountered in the commerciliazation of autonomous driving can be addressed. This paper introduces a unified framework for autonomous driving known as Vehicle-Infrastructure-Cloud Autonomous Driving (VICAD). VICAD combines the diverse collaborative deployment strategies related to vehicles, infrastructure, and the cloud with autonomous driving algorithms via an integrated framework. Simulations and evaluations are conducted to evaluate the performance of VICAD system, and evaluation results are then feedback as input of the VICAD system. This iterative process enables the continuous optimization of collaborative deployment strategies and autonomous driving algorithms, thereby enhancing the capabilities of autonomous driving. Moreover, this paper describes the key role of VICAD in fostering the large-scale commercial deployment of autonomous driving with practical cases and industrial applications, and concludes with suggestions for further VICAD development.

Key words: autonomous driving(AD), vehicle-infrastructure cooperation, safety of the intended functionality(SOTIF), Vehilcle-Infrastructure-Cloud Autonomous Driving(VICAD), operational design domain(ODD)

中图分类号: