Global Information
회사소개 | 문의

ADAS 및 자율주행차 관련 산업 분석 2018년(I) : 컴퓨팅 플랫폼과 시스템 아키텍처

ADAS and Autonomous Driving Industry Chain Report 2018 (I) - Computing Platform and System Architecture

리서치사 ResearchInChina
발행일 2018년 07월 상품 코드 666530
페이지 정보 영문 152 Pages
가격
US $ 3,400 ₩ 3,863,000 Unprintable PDF by E-mail (Single User License) help
1명만 이용할 수 있는 라이선스입니다. PDF 파일은 Copy & Paste 가 되지 않으며, 인쇄 또한 불가능합니다.
US $ 3,600 ₩ 4,091,000 Hard Copy help
동일 기업 내에서 공유하여 이용할 수 있는 라이선스입니다. 이용 인원수 제한은 없습니다.
US $ 5,000 ₩ 5,682,000 Printable & Editable PDF by E-mail (Enterprise-wide License) help
구매기업과 모회사, 자회사, 관련 회사 등 회사 전체에서 인원수에 제한없이 이용할 수 있는 라이선스입니다. PDF파일은 Copy & Paste 가 가능합니다. 인쇄횟수에 제한은 없으나, 인쇄물의 이용 범위는 PDF 이용 범위에 준합니다.


ADAS 및 자율주행차 관련 산업 분석 2018년(I) : 컴퓨팅 플랫폼과 시스템 아키텍처 ADAS and Autonomous Driving Industry Chain Report 2018 (I) - Computing Platform and System Architecture
발행일 : 2018년 07월 페이지 정보 : 영문 152 Pages

중국의 ADAS 및 자율주행차 시장은 2017년에 59억 위안, 2021년에는 426억 위안으로 매년 67%의 AAGR(연간 성장률)로 성장할 전망입니다.

세계와 중국의 ADAS(첨단운전자보조시스템)/자율주행 관련 플랫폼 및 시스템 아키텍처 정비 상황에 대해 분석했으며, 전체적인 시장 구조 및 동향 전망, 주요 자동차 제조업체의 개발 전략, 현재의 소프트웨어/하드웨어 구조, 각종 안전기준의 정비 상황, 주요 기업에서의 개발·정비 상황과 선행 사례 등의 정보를 정리하여 전해드립니다.

제1장 ADAS 및 자율주행의 개요

  • ADAS(첨단운전자보조시스템)의 분류와 정의
  • 자율주행차의 주요 기술 정의
    • 환경 인식 기술 : 센서 인식에서 데이터 융합으로의 이동
    • 포지셔닝 기술
    • 경로 계획 기술
    • 자동 주차 기술
  • 자율주행의 등급(미국 SAE 기준)
  • 자율주행의 등급(중국내 기준)
  • ADAS/자율주행의 규제와 규격
    • 빈조약(도로교통에 관한 조약, 1968년)의 변경에 의한 자율주행의 실현
    • 자율주행 시험에 관한 규제
    • EU : 11 항목의 자동차 안전 시스템 의무화(2021년부터)
  • 자율주행의 전형적인 프레임워크
    • 스텝 1 : 측위(포지셔닝)
    • 스텝 2 : 인식
    • 스텝 3 : 교통량 시나리오 예측
    • 스텝 4 : 의사결정
    • 스텝 5 : 행동 계획
    • 스텝 6 : 시행

제2장 시장 규모와 예측

  • 세계의 자율주행차 판매 대수 예측(2015-2050년)
  • 세계 ADAS 시장의 연간 평균 성장률(AAGR) 예측(2017-2025년)
  • Veoneer : 액티브 세이프티의 시장 규모가 300억 달러에 도달(2025년)
  • 중국 ADAS 및 자율주행차의 시장 규모(2016-2021년)
  • 자국내 ADAS 설치 승용차의 누계 대수(2017년) : ACC·FCW·LKS가 가장 급속히 확대

제3장 자동차 제조업체의 ADAS/자율주행용 전략

  • Geely
  • GM Intelligent Driving
  • Nissan·BMW·Xpeng : Mobileye Route
  • BMW : L3 CO-PILOT의 대량생산 계획(2021년)
  • Bosch : Chang'an·FAW·NIO·SAIC와의 제휴 계획
  • Bosch : 자율주행 솔루션
    • Bosch의 도메인 컨트롤러
    • TJP 솔루션
    • 센서 솔루션
    • HD 지도 솔루션
  • Aptiv와 Great Wall Motor의 제휴
  • Denso : GAC의 흡수 통합
  • Hyundai : 레벨 4 무인주행차용 센서의 레이아웃
  • Ford : 하이빔 LiDAR를 핵심 센서로 이용
  • BYTON와 Aurora의 협업

제4장 ADAS 및 자율주행차의 소프트웨어 아키텍처

  • ADAS/자율주행 시스템의 핵심 요소
  • Autosar의 개요
    • 로드맵
    • 주요 멤버
    • 클래식 버전과 어댑티브 버전
    • 클래식 버전의 구조
    • 어댑티브 버전의 소프트웨어 층별화 : 클래식/어댑티브 버전의 비교
    • 어댑티브 버전의 로드맵
  • ROS : 자율주행용 OS
    • 일부 자동차 제조업체가 인증하고 있는 ROS
    • ROS(로봇 오퍼레이팅 시스템)의 개요
    • ROS2.0 : 곧 실현화
    • ROS의 전환
    • ROS의 보안
  • QNX ADAS 2.0 : 최고의 ASIL D 레벨을 획득
    • QNX ADAS 2.0 지원 스코프

제5장 ADAS 및 자율주행의 하드웨어 아키텍처

  • 전형적인 자동차용 네트워크 아키텍처
  • 센트럴 게이트웨이로부터 도메인 컨트롤러 구조(NXP)로 이동
  • 미래의 자동차용 전기·전자기기 아키텍처(Bosch)
  • 왜 도메인 컨트롤러를 이용하는가?
    • 현재/향후 자동차용 전자기기 아키텍처
    • 도메인 컨트롤러에 의한 하드웨어 리소스의 공유 : OS·기본 소프트웨어와의 공유 실현화
    • I/O 아키텍처와 도메인 컨트롤러
    • 도메인 컨트롤러의 토대 : 자동차용 이더넷, 자동차용 버스와의 비교
  • 자동차용 이더넷
    • 자동차용 이더넷의 프로토 타입 : EAVB
    • EAVB의 다음 단계 : TSN
    • TSN 네트워크
    • TSN 이더넷 스위치 : 미래 자율주행용 컴퓨팅 시스템의 핵심
  • Waymo가 이용하고 있는 컴퓨터 시스템 아키텍처
  • NVIDIA PX2 : 아키텍처
  • NXP S32G : 게이트웨이
  • Renesas : L4 컴퓨팅 플랫폼의 구조

제6장 ADAS 및 자율주행의 안전 인증 기준

  • 정부의 자동차 규격에 적합한 칩의 인증제도
  • AEC의 인증
  • ISO 26262 : 기능적 안전성과 ASIL
  • ISO 26262 프로세스
  • 각종 안전 레벨에 따라 다른 판단 기준의 필요성
  • 자율주행용 ECU의 전형적 구조 : 모델 파트는 B레벨, 계획 파트는 D레벨에 도달

제7장 ADAS용 프로세서 벤더

  • ADAS/자율주행용 프로세서 업계
  • ARM
  • NXP
  • Renesas
  • Nvidia
  • Ambarella
  • Mobileye
  • TDA Series of Texas Instruments
  • Infineon
KSA 18.08.10

ADAS and Autonomous Driving Industry Chain Report 2018 (I) - Computing Platform and System Architecture underscores the followings:

  • Introduction to ADAS and autonomous driving;
  • ADAS and autonomous driving market forecast;
  • ADAS and autonomous driving strategy of carmakers including Geely, GM, SAIC, Dongfeng, Great Wall, GAC, Chang'an, NIO, Xpeng and BYTON;
  • Software architecture of ADAS and autonomous driving, including AUTOSAR Classic and Adaptive, ROS 2.0 and QNX;
  • Hardware architecture of ADAS and autonomous driving, including automotive Ethernet, TSN, Ethernet switch and gateway, and domain controller;
  • Safety certification of ADAS and autonomous driving, including ISO26262 and AEC-Q100;
  • Study into processor firms, including NXP, Renesas, Texas Instruments, Mobileye, Nvidia, Ambarella, Infineon and ARM.

According to ResearchInChina, the Chinese ADAS and autonomous driving market was worth about RMB5.9 billion in 2017 and is expected to reach RMB42.6 billion in 2021 at an AAGR of 67% or so.

Automotive vision, MMW radar and ADAS are the market segments that develop first with the MMW radar market enjoying an impressive growth rate, closely followed by low-speed autonomous driving. While LiDAR, commercial-vehicle autonomous driving and passenger-car autonomous driving markets lag behind.

As the automobile enters an era of ADAS and autonomous driving, product iteration races up and lifecycle of products is shortened. The automotive market is far smaller than consumer electronics market but sees bigger difficulty in design and higher design and production costs than that in consumer electronics market. Thus automotive ADAS and autonomous driving processor is confronted with higher risks. Hence adequate financial and human resources are required to support the development of automotive ADAS and autonomous driving processors. Globally, only very a few enterprises like NXP and Renesas are capable of developing whole series of ADAS and autonomous driving processors.

With regard to safety certification, autonomous driving chips must attain ASIL B at least, a level only Renesas R-CAR H3 has reached for now. As GPU is a universal design and not car-dedicated design, it is hard to reach the certified safety level of ISO26262 from the point of design. The certification cycle of ASIL is up to two to four years.

Reliability, precision and functionality of stereo camera are well above that of mono camera, but as the stereo camera must use FPGA, it costs much. High costs restraint the application of the stereo camera only on luxury cars. However, with emergence of Renesas and NXP hardcore stereo processors, the stereo camera will be vastly used in ADAS and autonomous driving field, expanding from luxury models to mid-range ones.

With an explosive growth in data transmission, automotive Ethernet will become a standard configuration of the automobile, and Ethernet gateway or Ethernet switch is indispensable to autonomous driving.

Autosar will act as a standard configuration in ADAS and autonomous driving field.

CNN/DNN graphics machine leaning: GPU is most suitable when data is irrelevant to sequence. Nvidia GPU can be used in multiple fields except for automobile and finds shipments far higher than that of automotive ASIC, enjoying superiority in cost performance. TPU lifts speed and reduces power consumption (only 10% of that of GPU) at the expense of the precision of computation.

RNN/LSTM/reinforcement learning sequence-related machine learning: FPGA has distinct advantages, particularly in power consumption, consuming less than one-fifth of GPU under same performance. However, high-performance FPGA is incredibly costly. FPGA can also process graphics machine leaning and improve performance by reducing precision.

ASIC stands out by performance-to-power consumption ratio but has shortcomings of long development cycle, the highest development cost and the poorest flexibility. The unit price will be very high or firms will make losses if the shipments are small (at least annual shipments of 120 million units if 7-nanometer process is employed). Most ASICs for deep-learning graphics machine learning are similar to TPU.

Power consumption and cost performance are crucial in in-vehicle field. GPU is no doubt a winner in graphic machine learning. However, as algorithms are constantly improved, the ever low requirements on the precision of computation, and low power consumption will ensure a place of FPGA in graphics machine learning. FPGA has overwhelming advantages in sequence machine learning.

Autonomous driving can be divided into two types, one represented by Waymo, which has solved most of the problems concerning environmental perception and concentrates on behavior decision-making with computing architecture of CPU+FPGA (usually Intel Xeon 12-core and above CPU plus Altera or Xilinx's FPGA; the other represented by Mobileye which has not solved all problems involving environmental perception and concentrates on it with computing architecture of CPU+GPU/ASIC.

CPU+GPU will be the mainstream in the short run, but CPU+FPGA/ASIC may dominate in the long term, largely due to continuous decline in the precision of computation of graphics because of improvement in algorithms and performance of sensors (LiDAR in particular), which is conducive to FPGA, while it is hardly for the power consumption of GPU to fall. It is easier for FPGA to meet car-grade requirements.

In chip contract manufacturing field, TSMC has won all 7-nanometer chip orders, including A12 exclusively provided for Apple, marking for the first time TSMC overtook Intel to become the vendor with the most advanced semiconductor manufacturing process, a must in the production of digital logic chip whose computing capability is underlined in AI autonomous driving.

Table of Contents

1. Introduction to ADAS and Autonomous Driving

  • 1.1. Definition and Classification of ADAS
  • Main Functions of ADAS
  • 1.2. Definition and Key Technologies of Autonomous Vehicle
    • 1.2.1. Environmental Perception Technology: from Sensor Perception to Data Fusion
  • Environmental Perception Technology: Different Sensors Have Different Advantages
    • 1.2.2. Positioning Technology
    • 1.2.3. Path Planning Technology
    • 1.2.4. Automatic Parking Technology
  • 1.3. Grading of Autonomous Driving (SAE)
  • 1.4. Grading of Autonomous Driving (China)
  • 1.5. Regulations on and Standards for ADAS and Autonomous Driving
    • 1.5.1. Amendment to the 1968 Vienna Convention on Road Traffic Allows Autonomous Driving
    • 1.5.2. Regulations on Autonomous Driving Tests
    • 1.5.3. EU Lists 11 Automotive Safety Systems to Become Mandatory from 2021
  • 1.6. Typical Framework of Autonomous Driving
    • 1.6.1. First Step, Positioning
  • HD Map and V2X
    • 1.6.2. Step 2, Perception
  • 3D Bounding with Route Fusion
    • 1.6.3. Step 3: Traffic Scenario Forecast
  • Forecast Includes Scenario Understanding
    • 1.6.4. Step 4: Decision-making
  • Lane Overall Planning
  • Shorter Routes May Be Not Better.
  • Behavior Planning Is the Most Difficult
  • There Are Many Behavior Planning Algorithms, Mostly Immature
    • 1.6.5. Step 5: Action Planning
    • 1.6.6. Step 6: Execution

2. Market Size and Forecast

  • 2.1. Global Sales Volume of Autonomous Vehicles, 2015-2050E
  • 2.2. AAGR of Global ADAS Market, 2017-2025E
  • 2.3. Veoneer: Active Safety Market Is Expected to Reach USD30 Billion by 2025
  • 2.4. Chinese ADAS and Autonomous Driving System Market Size, 2016-2021E
  • 2.5. Concurrent Comparison of Domestic Passenger Car ADAS Cumulative Installations in 2017: ACC, FCW and LKS Saw the Fastest Growth Rate

3. Carmakers' ADAS and Autonomous Driving Strategies

  • 3.1. Geely
  • 3.2. GM Intelligent Driving
  • 3.3. Mobileye Route of Nissan, BMW and Xpeng
  • 3.4. BMW Plans to Mass-produce L3 CO-PILOT in 2021.
  • Intel's Driverless Cars Use 32-beam LiDAR
  • 3.5. Bosch Route of Chang'an, FAW, NIO and SAIC
  • 3.6. Bosch's Autonomous Driving Solutions
    • 3.6.1. Bosch's Domain Controllers
  • Comparison between Various Domain Controllers
    • 3.6.2. TJP Solutions
    • 3.6.3. Sensor Solutions
    • 3.6.4. HD Map Solutions
    • 3.6.5. Planning for Commercial Vehicle Autonomous Driving
  • 3.7. Aptiv Route of Great Wall
  • Aptiv's Road Model Relies on LiDAR
  • 3.8. Denso Route of GAC
  • 3.9. Layout of Hyundai L4 Driverless Car Sensors
  • 3.10. Ford Uses High-beam LiDAR as the Core Sensor
  • 3.11. BYTON Collaborates with Aurora

4. Software Architecture of ADAS and Autonomous Driving

  • 4.1. Core Elements of ADAS and Autonomous Driving System
  • 4.2. Introduction to Autosar
    • 4.2.1. Roadmap
    • 4.2.2. Main Members
    • 4.2.3. Classic Version and Adaptive Version
    • 4.2.4. Architecture of Classic Version
    • 4.2.5. Software Stratification of Adaptive Version; Comparison between Classic Version and Adaptive Version
    • 4.2.6. Roadmap of Adaptive Version
  • 4.3. ROS: an Autonomous Driving Operating System
    • 4.3.1. ROS Recognized by Some Carmakers
    • 4.3.2. Introduction to ROS
    • 4.3.3. ROS2.0 Is Close to Real Time
    • 4.3.3. Transformation of ROS
    • 4.3.4. Security of ROS
  • 4.4. QNX ADAS 2.0 Achieves the Highest ASIL D Level
    • 4.4.1. Scope Supported by QNX ADAS 2.0

5. Hardware Architecture of ADAS and Autonomous Driving

  • 5.1. Typical Automotive Network Architecture
  • 5.2. From the Central Gateway to the Domain Controller Structure (NXP)
  • 5.3. Future Automotive Electronic and Electrical Architecture (Bosch)
  • 5.4. Why Use A Domain Controller
    • 5.4.1. Current and Future Automotive Electronic Architecture
    • 5.4.2. Domain Controllers Share Hardware Resources, so that Operating System and Basic Software Realize Sharing
    • 5.4.3. I/O Architecture and Domain Controller
    • 5.4.4. Basis of Domain Controller: Automotive Ethernet, Automotive Bus Comparison
  • Automotive Bus Comparison
  • 5.5. Automotive Ethernet
    • 5.5.1. Prototype of Automotive Ethernet: EAVB
    • 5.5.2. The Next Step of EAVB: TSN
    • 5.5.3. TSN Network
    • 5.5.4. TSN Ethernet Switch Is the Core of the Future Autonomous Driving Computing System
  • 5.6. The Computing System Architecture Used by Waymo
  • 5.7. NVIDIA PX2: Architecture
  • 5.8. NXP S32G: Gateway
    • 5.8.1. Architecture of NXP Autonomous Driving Blue Box
    • 5.8.2. Gateway and Ethernet Switch
  • 5.9. Architecture of Renesas L4 Computing Platform
    • 5.9.1. Renesas' Vision of the Future Automotive Electronic Architecture

6. Safety Certification of ADAS and Autonomous Driving

  • 6.1. Chip Certification in Line with National Automotive Standards
  • 6.2. AEC Certification
  • 6.3. ISO26262, Functional Safety and ASIL
  • 6.4. ISO26262 Process
  • 6.5. Different Safety Levels Require Different Judgmental Independence
  • 6.6. Typical Structure of Autonomous Driving ECU; the Model Part Reaches the B Level; the Planning Part Reaches the D Level

7. ADAS Processor Vendors

  • 7.1. ADAS and Autonomous Driving Processor Industry
    • 7.1.1. FPGA/GPU/ASIC/CPU/TPU and Machine Learning
    • 7.1.2. Soft/Solid/Hard Core
    • 7.1.3. Solid Core Is the Mainstream
    • 7.1.4. Architecture of Typical L4 Computing System
  • 7.2. ARM
    • 7.2.1. Application Structure of ARM Autonomous Vehicles
    • 7.2.2. Autonomous Driving SoC Design Recommended by ARM
    • 7.2.3. ARM A Series
    • 7.2.4. ARM R Series and M Series
  • 7.3. NXP
    • 7.3.1. NXP Autonomous Driving CPU Roadmap
    • 7.3.2. Roadmap of NXP's ADAS and Autonomous Driving Vision Processing Chip
    • 7.3.3. Introduction to NXP S32V3
    • 7.3.4. NXP S32V3 Vision Processing System
    • 7.3.5. Framework Diagram of NXP ADAS Chassis Control MCU MPC5746R
    • 7.3.6. NXP Autonomous Driving Chassis Control MCU: S32D/S Series
  • 7.4. Renesas
    • 7.4.1. Renesas R-CAR H3
    • 7.4.2. Renesas R-CAR V3H
    • 7.4.3. Renesas RH850/P1H-C
  • MCU with the Highest Safety Level Designed for Chassis Control
    • 7.4.4. Renesas Cooperates with Dibotics to Develop LiDAR Applications
    • 7.4.5. Renesas Partners with USHR in HD Map
    • 7.4.6. Renesas Teams up with QNX and University of Waterloo in Operating System
    • 7.4.7. Renesas Collaborates with Leddartech on LiDAR
    • 7.4.8. Renesas' Cooperation in Autonomous Driving
  • 7.5. Nvidia
    • 7.5.1. Parameters of Nvidia DRIVE Series Products
    • 7.5.2. Circuit Schematic Diagram of PX2
    • 7.5.3. Nvidia DRIVE Xavier
    • 7.5.4. Nvidia DRIVE Pegasus
  • 7.6. Ambarella
    • 7.6.1. Technology Distribution and Roadmap
    • 7.6.2. Core Technology CVflov and Stereo-camera Data Processing Hard Core
    • 7.6.3. Ambarella CV2AQ
    • 7.6.4. Ambarella CV2AQ
  • 7.7. Mobileye
    • 7.7.1. Internal Framework Diagram of Mobileye Eyeq4/5
    • 7.7.2. Dual-EYEQ4 L3 Solutions (HiRain Technologies)
  • 7.8. TDA Series of Texas Instruments
    • 7.8.1. Introduction to TDA2 Series
    • 7.8.2. TDA4 and TIDL
    • 7.8.3. Single-chip MMW Radar Solutions
  • 7.9. Infineon
    • 7.9.1. MEMS LiDAR Solutions
    • 7.9.2. MMW Radar Transceivers
Back to Top
전화 문의
이용안내