Common Methodology for Data-Driven Scenario-Based Safety Assurance in the HEADSTART Project
Contributors
Description
One objective of introducing Connected and Automated Driving (CAD) functions on the roads is to
reduce the number of accidents due to human errors by reducing the tasks of the driver (partial
automation) or removing it completely from the driving system. Building safe and reliable automated
vehicles require specific testing methods that are adapted to higher levels of automation. Harmonised
European Solutions for Testing Automated Road Transport (HEADSTART) is a research project
funded by the European Union that aims to define testing and validation procedures for CAD
functions. HEADSTART brings a methodology for testing and validating these functions with
Data-Driven Scenario-Based Safety Assurance. The goal of this paper is first to present the overall
HEADSTART methodology for validating CAD safety, and then further explain three aspects of it,
such as: The database mechanics to extract and parametrize logical scenarios, the relevance metrics for
the selection of scenarios, and the allocation of scenarios for test execution.
Files
2020_ITS-America_Headstart.pdf
Files
(1.1 MB)
Name | Size | Download all |
---|---|---|
md5:0e8be76e869ca37f415e0f89f2fa138f
|
1.1 MB | Preview Download |