聯網自動駕駛汽車高速公路 (CAVH)：大規模自動駕駛系統 (ADS) 部署的愿景和發展報告 (2019)（英文版）（42頁）.pdf

《聯網自動駕駛汽車高速公路 (CAVH)：大規模自動駕駛系統 (ADS) 部署的愿景和發展報告 (2019)（英文版）（42頁）.pdf》由會員分享，可在線閱讀，更多相關《聯網自動駕駛汽車高速公路 (CAVH)：大規模自動駕駛系統 (ADS) 部署的愿景和發展報告 (2019)（英文版）（42頁）.pdf（42頁珍藏版）》請在三個皮匠報告上搜索。

1、 Connected Automated Vehicle Highway (CAVH): A Vision and Development Report for Large-Scale Automated Driving System (ADS) Deployment Version 1.0 Working Committee of Automated Driving China Highway and Transportation Society June 2019 I Copyright Statement The copyright of this report belongs to t

2、he Working Committee of Automated Driving of China Highway and Transportation Association and is protected by law. Where the texts or viewpoints of this report are reproduced, extracted or used in other ways, the source shall be indicated as Source: Working Committee of Automated Driving of China Hi

3、ghway and Transportation Association. Those who violate the above statement will be investigated for legal responsibility. III Contributors Publication Agency: Working Committee on Automated Driving, China Highway and Transportation Society Contributing Institutes: 1. Institute on Internet of Mobili

4、ty, Southeast University and University of Wisconsin-Madison 2. Southeast University 3. Tsinghua University 4. Beihang University 5. Beijing Jiaotong University 6. Beijing University of Technology 7. Traffic Management Research Institute, Ministry of Public Security Contributing Authors: 1. Yang CHE

5、NG 2. Yuan ZHENG 3. Huachun TAN 4. Jianbo YI 5. Xu QU 6. Shen LI 7. Shuyan HE 8. Qiang TU 9. Haijian LI 10. Zhenlong LI 11. Yongming HE 12. Zhihong YAO 13. Yangxin LIN 14. Yanjin LI 15. Tengfei YUAN 16. Haoran WU 17. Fan DING 18. Danya YAO 19. Shanglu HE 20. Haiyang GU 21. Zhijun CHEN 22. Kegang ZHA

6、O 23. Yikang RUI 24. Fan YANG 25. Jian ZHANG 26. Dazhi JIANG 27. Jun BI IV 28. Yuanli GU 29. Steven PARKER 30. Shawn LEIGHT 31. Peter JIN 32. David NOYCE 33. Soyoung (Sue) AHN 34. Jianqiang WANG 35. Yugong LUO 36. Shengbo LI 37. Mingyuan BIAN 38. Guizhen YU 39. Guangquan LU 40. Daxin TIAN 41. Haiyan

7、g YU 42. Li LI 43. Jianming HU 44. Wenjie LU 45. Min WANG 46. Wei SUN 47. Jianhua YUAN 48. Zhigang XU 49. Siyuan GONG Advisory Committee of Experts: 1. Mengyong WENG 2. Song PANG 3. Bin RAN 4. Keqiang LI 5. Yunpeng WANG 6. Yi ZHANG 7. Changjun WANG 8. Xiaohui SHI 9. Xiangmo ZHAO 10. Shanzhi CHEN 11.

8、 Jinquan ZHANG 12. Ling JIN 13. Guangsu SHANG 14. Xiaojing WANG 15. Yanqing CEN 16. Gang WANG 17. Bo NIU 18. Yongjian ZHONG V Introduction Intelligent Transportation Systems (ITS) has been considered instrumental methods to alleviate traffic congestions and improve traffic safety when the constructi

9、on of new infrastructure or capacity increase are not viable. With the rapid development of artificial intelligence, mobile internet, big data, and other innovative information technologies, the next generation of ITS will feature the Automated Driving System (ADS) technologies and become the essent

10、ial approach to ultimately address traffic problems. The vehicle-based ADS solutions are lead by Google, Tesla, Uber, and Baidu. Vehicle-based ADS systems use onboard high-resolution sensors to detect real-time driving conditions, and use on-board edge computing and Artificial Intelligence (AI) algo

11、rithms for driving decision making and vehicle control to achieve automated driving. The main limitations of the vehicle-based technologies are the high concentration of onboard technologies and sensors, which make the large-scale deployment of technologies financially and societally difficult. The

12、industry has come to realize that the large-scale deployment of ADS needs to take advantage of intelligent road infrastructure. Vehicle-based ADS sensing and computing systems can also be deployed along intelligent infrastructures to achieve ADS for all vehicles with less intensive resource deployme

13、nt on the vehicle end compared with the vehicle-based approach. The interaction and coupling between the intelligent road infrastructures and intelligent vehicles have the potentials of facilitating or even replacing vehicle-based ADS technologies in large-scale deployment. Under such vehicle-infras

14、tructure integration paradigm, the development of the Connected Automated Vehicle Highway (CAVH) technologies will take the advantage of the rapid development and commercialization of ADS technologies, next-generation wireless communication, artificial intelligence(AI), Internet of Things (IoT), Clo

15、ud and Edge computing, Electric Vehicles and highway electrification, smart sensor and infrastructure, smart city, and other technologies to achieve ADS in large scales. The CAVH approach can also enable deep integration across different industrial sectors, including information technologies, intell

16、igent manufacturing, transportation and logistics, and automobile industries to establish cross-industry ecosystems and supply chains for new scientific, technological, and industrial revolutions. CAVH extends the existing applications of Connected Automated Vehicle (CAV) technologies into system-wi

17、de integration between vehicles and infrastructure. The development and deployment of the CAVH-based ADS technologies could be classified as four different stages: Stage I, information exchange and interaction: establish the Vehicle to Infrastructure & Infrastructure to Vehicle (V2I & I2V) connectiv

18、ity. Stage II, collaborative sensing, prediction, and decision-making: share and integrate the sensing and prediction results between vehicles and infrastructure and execute coordinated VI decision-making. Stage III, coordinated control: coordinate the real-time vehicle and infrastructure control wi

19、th collaborative sensing, prediction, and decision-making results. Stage IV, vehicle-infrastructure integration: based on stage I, II and III, vehicles and infrastructure could achieve overall coordination and complete system functions to achieve global planning, control, and optimization of vehicle

20、-infrastructure operations for ADS. The CAVH-based ADS uses the advanced sensor, network, computing, and control technologies, to achieve the comprehensive sensing for the road and traffic environment, and the wide range and large capacity data sharing between multiple systems facilitating different

21、 vehicles and different traffic automation system at various integration levels. Based on CAVH, automated driving systems can be constructed from three dimensions: vehicle automation, network interconnection, and system integration. Such systems can efficiently execute the essential automated drivin

22、g functions of sensing, prediction, decision-making, and control, and eventually forms intelligent systems that can integrate, coordinate, control, manage, and optimize all vehicles, information services, facilities, equipment, intelligent traffic management, and control. The CAVH is composed of fou

23、r key subsystems, including traffic management subsystem, smart roadside infrastructure subsystem, intelligent vehicle subsystem, and intelligent communication subsystem, and four key functional modules: sensing module, prediction module, decision-making module, and control module. From a generalize

24、d perspective, the CAVH system covers the CAV system and the intelligent road infrastructure system. That is, the intelligent network vehicles, the Internet of vehicles, the active traffic management systems, the automated highway systems, and transportation entities are all included. The advanced C

25、AVH is a more advanced stage of CAVH, which further enhances the intelligence of road infrastructure, which therefore accelerates the commercialization of automated driving, and eventually, achieves the integrated development of vehicles and roads for the automated driving. Traditional technologies

26、and industries are facing reconstruction and reengineering due to continuous innovations. As an emerging system, CAVH is bound to generate new technologies, industries, and businesses. The CAVH system covers automated vehicles, transportation environment, communication facilities, traffic management

27、, and control system, and other entities, related to technologies of computer vision, communication, network security, vehicle collaboration with road, active control, human-vehicle-road-center collaborative service management, highway automation system, and system integration. These technologies ar

28、e key to vehicle road collaborative automated driving toward commercialization. CAVH involves many industries, which are of multiple roles, complementing advantages, and prominent characteristics. Developing CAVH systems and promoting CAVH technology applications can advance the chip, software, info

29、rmation communication, data service, and other industry development and transformation, who are the suppliers of vehicle and infrastructure industries. The CAVH based automated driving ecosystem can enable the deployment of intelligent VII transportation and smart city solutions, a system of systems

30、 integration, and shared digital economy, to foster new economic growth. This report presents the strategic plans and development directions of key technologies in CAVH, analyzes the development trends and CAVH industry positioning, and puts forward suggestions for the development of CAVH that is su

31、itable for China based on an in-depth understanding of the concepts and connotation of CAVH. IX Contents Copyright Statement . I Introduction . V Contents . IX 1 Automated Driving . 1 1.1 Connotation of Automated Driving . 1 1.1.1 Principle of Automated Driving . 1 1.1.2 Automated Driving Classifica

32、tion . 1 1.2 Concept of Vehicle-Road Cooperative Automated Driving . 3 1.2.1 Vehicle Automation . 4 1.2.2 Connectivity . 5 1.2.3 System Integration . 5 1.3 Development Level of Vehicle-Infrastructure Cooperative Automated Driving . 6 1.4 Key Sub-systems and Modules . 7 1.4.1 Key Sub-systems . 7 1.4.

33、2 Key Modules . 8 2 Key Technology and Development Trend of CAVH. 11 2.1 Key Technology Analysis of CAVH . 11 2.1.1 Environment Sensing Technology . 11 2.1.2 Data Fusion and Prediction Technology . 11 2.1.3 Intelligent Decision Technology . 11 2.1.4 Control Execution Technology . 12 2.1.5 I2X and V2

34、X Communication Technology . 12 2.1.6 Network Security Technology . 12 2.1.7 Collaborative Optimization Technology . 13 2.1.8 Integrated Optimization Technology for Transportation System . 13 2.2 Future Direction of CAVH . 13 2.2.1 Integrating High-Resolution and High-Reliability Positioning with Ro

35、adSide Unit based CAVH Applications . 13 2.2.2 Visual Recognition and LiDAR Becoming the Core of Sensing Technologies . 14 2.2.3 Cloud Platform Technologies . 14 2.2.4 Connected Automated Vehicle Technologies . 15 2.2.5 CAVH Traffic System Optimization Technologies . 15 3 Development Trend and Roles

36、 of CAVH . 17 3.1 Development Trend of CAVH . 17 3.1.1 Key Technologies and Infrastructures (Upstream) . 18 3.1.2 Intelligent Manufacturing and System Integration (Middle) . 21 3.1.3 Application Services and Value-Added Services (Downstream) . 22 3.1.4 CAVH Standardization . 24 3.2 Roles of Organiza

37、tions, Agencies, and Enterprises in the Development of CAVH . 25 3.2.1 Government Agencies . 25 X 3.2.2 Industry . 25 3.2.3 Universities and Research Institutes . 26 3.2.4 Associations . 26 3.2.5 Financial Capital and Investment Institutions. 27 4 Policies and Suggestions . 29 4.1 Government . 29 4.

38、1.1 Policies . 29 4.1.2 Standards. 29 4.1.3 Laws and Regulations . 29 4.2 Enterprises . 30 4.3 Universities and Research Institutes . 30 5 Terminology . 31 6 References . 33 Working Committee of Automated Driving, CHTS Development Report of CAVH (2019) 1 1 Automated Driving 1.1 Connotation of Automa

39、ted Driving 1.1.1 Principle of Automated Driving Automated driving means that the vehicle senses the surrounding environment through the sensors onboard, makes control decisions based on the information collected and fused, and makes vehicle control decisions, including both longitudinal control and

40、 lateral control. The process of automated driving mainly includes three stages: information collection, information processing, and instruction execution. In the information collection stage, the automated vehicles detect the surrounding environment through the radar and camera mounted on the vehic

41、le and collects information such as the position, speed, acceleration as well as pedestrians and vehicles nearby. Information processing stage: the autonomous vehicle transmits the collected information to the automotive electronic control unit (ECU) for analysis, calculation, and control decision.

42、Instruction execution phase: the autonomous vehicle transmits the control decisions provided by the vehicles electronic control unit to the engine/motor management system and the electric power steering system (EPS) to achieve vehicle acceleration, deceleration, and steering operations. 1.1.2 Automa

43、ted Driving Classification The SAE level of classification for automated driving is adopted in this report 1, as is shown in Figure 1 and Table 1. Figure 1 Automation Levels 1 Working Committee of Automated Driving, CHTS Development Report of CAVH (2019) 2 Table 1 Classification of Self-Driving Vehi

44、cles 1 NHTSA SAE The degree of automation Specific definition Driving performance Monitoring Take over Application scenes 0 0 No Automation Zero autonomy; the driver performs all driving tasks. Human driver Human driver Human driver None 1 1 Driver Assistance Vehicle is controlled by the driver, but

45、 some driving assist features may be included in the vehicle design. Human driver and vehicle Human driver Human driver Qualified scenes 2 2 Partial Automation Vehicle has combined automated functions, like acceleration and steering, but the driver must remain engaged with the driving task and monit

46、or the environment at all times. Vehicle Human driver Human driver 3 3 Conditional Automation Driver is a necessity but is not required to monitor the environment. The driver must be ready to take control of the vehicle at all times with notice. Vehicle Vehicle Human driver 4 4 Highly Automation The

47、 vehicle is capable of performing all driving functions under certain conditions. The driver may have the option to control the vehicle. Vehicle Vehicle Vehicle Working Committee of Automated Driving, CHTS Development Report of CAVH (2019) 3 5 Fully Automation The vehicle is capable of performing al

48、l driving functions under all conditions. The driver may have the option to control the vehicle. Vehicle Vehicle Vehicle All scenes Level 0: The human driver does all the driving. Level 1: An Advanced Driver Assistance System (ADAS) on the vehicle can sometimes assist the human driver with either st

49、eering or braking/accelerating, but not both simultaneously. Level 2: An Advanced Driver Assistance System (ADAS) on the vehicle can itself actually control both steering and braking/accelerating simultaneously under some circumstances. The human driver must continue to pay full attention (“monitor

50、the driving environment”) at all times and perform the rest of the driving task. Level 3: An Automated Driving System (ADS) on the vehicle can itself perform all aspects of the driving task under some circumstances. In those circumstances, the human driver must be ready to take back control at any t