Machine Learning
and DNN applications
Eleven industry leaders across the automotive and autonomous driving technology spectrum announced the publication of, “Safety First for Automated Driving,” or SaFAD, a comprehensive, organized framework for the development, testing and validation of safe autonomous passenger vehicles.
These 11 leaders — Aptiv, Audi, Baidu, BMW, Continental, Daimler, Fiat Chrysler Automobiles, HERE, Infineon, Intel and Volkswagen — comprise the broadest representation across the industry and have published, to date, the largest report on how to build, test, and operate safe autonomous vehicles.
The purpose of the SaFAD white paper is to emphasize the importance of safety by design, along with verification and validation, as the industry works toward creating standards for automated driving. For the first time, SaFAD offers the autonomous vehicle (AV) developers and operators a system for clear traceability that proves AVs to be “safer than the average driver,” through components such as cameras or steering systems.
Learn More<Automatic driving safety first>
download
Deriving Capabilities from Dependabilities
(FuSa/SoTIF/Cybersecurity)
Elements for Implementing Capabilities
General Architecture
Key Challenges for V&V
V&V Approach for ADS
Quantity and Quality of Testing
V&V of Elements
Simulation and Field Operation
Road Vehicles:
Safety of the intended
functionality (SOTIF)
Road Vehicles :
Functional safety
Road Vehicles:
Cybersecurity engineering
Geographic information:
Data quality
Geographic information:
Quality assurance of data supply
Quality management systems:
Particular requirements for the
application of ISO 9001:2008 for
automotive production and relevant
service part organizations
Information technology :
Vocabulary - Part 1: Fundamental
terms
Systems and software engineering:
System life cycle processes
Human Factors
Machine Learning
and DNN applications
HD Map as a
non-traditional component
Technological capabilities
of the sensory devices
<Reliable and Safe map for automated driving> is a technical white paper created by Baidu and Elektrobit.
The white paper integrate different safety methodologies and map related quality standards, aimed at provide technical guidance as well as the basis for the future safety standard for maps.
Learn More<Reliable and Safe map for automated driving>
download
Reliable and safe maps | |||
---|---|---|---|
Use cases | Interrelationships between safety areas | Derivation of safety requirements | May/reality deviations as a new fault type |
4 types,17 soures of map/reality deviations: | |||
Map inaccuracies | Map data errors (9 sources) | Reality changes (4 sources) | Map update interruptions (3 sources) |
Remaining challenges |