没有合适的资源?快使用搜索试试~ 我知道了~
首页可靠和安全的自动驾驶地图--百度apollo.pdf
资源详情
资源评论
资源推荐
Reliable and safe maps
for automated driving
baidu.com
elektrobit.com
Reliable and safe maps for automated driving
Contents
Contents ................................................................................................................................ 2
1 Preface ............................................................................................................................4
2 Abstract .....................................................................................................................................4
3 Highly reliable maps: use cases ........................................................................................4
3.1 Sensor range extension ..................................................................................................................................5
3.2
Support when sensors work inadequately .................................................................................................. 6
3.3
Location-based information unavailable from sensors ............................................................................6
3.4
Vehicle localization ...........................................................................................................................................6
4 Interrelationships between safety areas .......................................................................... 7
5 Derivation of safety requirements for maps ..................................................................... 7
5.1 Iteration 1: Main function SOTIF ....................................................................................................................8
5.2
Iteration 2: Main function safety in use .......................................................................................................9
5.3
Iteration 3: Technical safety requirements SOTIF .................................................................................... 10
5.4
Iteration 4: Availability of the main function due to safety requirements ......................................... 10
6 Differences between the map and reality ........................................................................11
6.1 Minimization, quantication and mitigation of map inaccuracies ....................................................... 12
6.1.1
Quantication and mitigation ................................................................................................................ 12
6.1.2
Minimization .............................................................................................................................................. 13
6.2
Map data error prevention and detection ................................................................................................. 13
6.2.1
Data collection .......................................................................................................................................... 14
6.2.2
Map production ........................................................................................................................................ 15
6.2.2.1
Human errors ....................................................................................................................................... 15
6.2.2.2
Software/tool errors .......................................................................................................................... 15
6.2.2.3
Process errors ...................................................................................................................................... 16
6.2.2.4
Cybersecurity ....................................................................................................................................... 16
6.2.3
Map transmission ..................................................................................................................................... 16
6.2.4
In-vehicle errors ....................................................................................................................................... 16
3
Reliable and safe maps for automated driving
6.3 Anticipation and detection of reality changes ...........................................................................................17
6.3.1
Anticipation of reality changes by means of information from road building authorities ........17
6.3.2
Anticipation of reality changes by the use of crowdsourcing to detect preparation ................ 18
6.3.3
Detection of reality changes by the use of measurement vehicles .............................................. 18
6.3.4
Detecting reality changes through crowdsourcing ........................................................................... 19
6.4
Detection and mitigation of map update interruptions ......................................................................... 19
6.4.1
Interruption in map transmission ......................................................................................................... 19
6.4.2
Interruption in map production ............................................................................................................20
6.4.3
Interruption in map data collection .....................................................................................................20
6.5
Detection and mitigation of deviations ......................................................................................................20
6.5.1
Detection ....................................................................................................................................................20
6.5.2
Mitigation ...................................................................................................................................................20
7 Remaining challenges ....................................................................................................20
8 Appendix ........................................................................................................................21
8.1 SAE automation levels ................................................................................................................................... 21
8.2
Abbreviations ..................................................................................................................................................22
8.3
References .......................................................................................................................................................22
Legal disclaimer
Company-confidential and proprietary information
ALL RIGHTS RESERVED. No part of this publication may be copied in any form, by photocopy, microfilm,
retrieval system, or by any other means now known or hereafter invented without the prior written
permission of Elektrobit Automotive GmbH and Baidu, Inc.
Reliable and safe maps for automated driving
1 Preface
Using map data in automated driving applications raises the questions: Which applications require it?
Can it be used for safety-related functions? In what ways can highly reliable maps be made?
This document examines these questions, describes some technical implementation approaches, and indicates
the unsolved problems.
2 Abstract
· Automated driving functions can utilize highly reliable map data
· Map suppliers’ tools and processes need reviewing and improvement to achieve the required data integrity
· Map quality metadata is required for HAD applications
· Map suppliers and road building authorities must cooperate to keep map data current
· Some safety aspects require further discussion
(e.g., backend system protection methods, map quality metadata)
· Important aspects need standardization
(e.g., SOTIF acceptable risk level, backend system safety requirements)
3 Highly reliable maps: use cases
At the moment, different application scenarios can be projected for highly reliable maps:
3
© Elektrobit (EB) 2019 | Confidential
This leads to multiple map safety requirements
Highly Reliable Maps
Appli-
cation
Example
Sensor range extension
Support for insufficient
sensor performance
Localization
Location based
information that can not
be derived from sensors
Takeover request in time
before end of automated
driving area
Road clearance for automated
driving by car manufacturer,
country specific rules
Bad weather (snow, ice, fog),
country specific signs, physical
hacking
GPS map matching,
landmarks
Figure 2: Highly reliable map applications
3
© Elektrobit (EB) 2019 | Confidential
This leads to multiple map safety requirements
Highly Reliable Maps
Appli-
cation
Example
Sensor range extension
Support for insufficient
sensor performance
Localization
Location based
information that can not
be derived from sensors
Takeover request in time
before end of automated
driving area
Road clearance for automated
driving by car manufacturer,
country specific rules
Bad weather (snow, ice, fog),
country specific signs, physical
hacking
GPS map matching,
landmarks
5
Reliable and safe maps for automated driving
Maps do not replace sensors, which supply data about dynamic processes in the vehicle’s environment, e.g.,
the actions of pedestrians, cyclists, and other vehicles.
Reliance on a map for specic information means only that sensors cannot provide the information with the
required quality; the vehicle still needs sensors. For example: a camera cannot read an obscured sign, but the
data can be retrieved from a map. Conversely, a map cannot deliver information about the distance to other cars.
According to ISO 26262, map systems cannot be used to decompose safety requirements in conjunction with
sensor systems if only one of the systems provides the necessary information with the required quality, excluding
hardware or software failure.
A map system can thus only be used here either to rectify physical/algorithmic sensor system inadequacies
or for decomposition.
Safety related applications usually only require a subset of the map attributes to be highly reliable rather than
the entire map. These attributes are called reliable map attributes (RMA). Section 5 of this white paper gives an
example of how to derive the detailed RMA list for a given application.
3.1 Sensor range extension
Predicting situations
Map information can extend sensor range through advance identication of potential hazard situations. The
automated driving function can match or improve on an average driver’s skills, e.g., by adjusting speed when
speed limit signs are obscured, or braking before reaching an obstacle
Requesting driver takeover well before HAD route ends
HAD systems are currently only good enough for use in specic areas (e.g., highways).
In HAD mode, drivers generally do not monitor vehicle function and environmental conditions and are unaware
of functional restrictions. They are not therefore expected to identify in advance where HAD mode should end, or
to take immediate control of the vehicle when prompted (time is needed for reaction and situational awareness).
When sensors detect the HAD route end point, insufcient time may remain to transfer control to the driver.
Hazards and discomfort may result if emergency braking is needed, which is even more likely if sensors cannot
detect the end of the HAD route (e.g., signs obscured).
The system can compute the end of the HAD route from map data, and alert the driver in good time.
Steering to a safe stop
If the driver fails to take over, emergency in-lane braking is avoidable if, e.g., a hard shoulder is available.
A map can indicate if and where there is one.
The vehicle safety concept can also include this as the response to a safety-related component failure
or driver medical emergency.
剩余23页未读,继续阅读
Kmiracler
- 粉丝: 24
- 资源: 1
上传资源 快速赚钱
- 我的内容管理 收起
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
会员权益专享
最新资源
- zigbee-cluster-library-specification
- JSBSim Reference Manual
- c++校园超市商品信息管理系统课程设计说明书(含源代码) (2).pdf
- 建筑供配电系统相关课件.pptx
- 企业管理规章制度及管理模式.doc
- vb打开摄像头.doc
- 云计算-可信计算中认证协议改进方案.pdf
- [详细完整版]单片机编程4.ppt
- c语言常用算法.pdf
- c++经典程序代码大全.pdf
- 单片机数字时钟资料.doc
- 11项目管理前沿1.0.pptx
- 基于ssm的“魅力”繁峙宣传网站的设计与实现论文.doc
- 智慧交通综合解决方案.pptx
- 建筑防潮设计-PowerPointPresentati.pptx
- SPC统计过程控制程序.pptx
资源上传下载、课程学习等过程中有任何疑问或建议,欢迎提出宝贵意见哦~我们会及时处理!
点击此处反馈
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功
评论5