TY  - JOUR
PY  - 2023//
TI  - ADSCENE scenarios data base: focus on accident data support for validation of automated driving functions
JO  - Transportation research procedia
A1  - Guyonvarch, L.
A1  - Hermitte, T.
A1  - Kroger, R.
A1  - Chauvel, C.
A1  - Arnoux, E.
A1  - Geronimi, S.
SP  - 9
EP  - 16
VL  - 72
IS  - 
N2  - Automated systems such as SAE AD level 2 and higher, face the difficult challenge of proving safe in the field, in actual traffic conditions. Scenario-based simulation approaches are, therefore, necessary complements to the traditional approach, allowing computation of a controlled diversity of key variables in many iterations in a safe, fast, and documented way. ADSCENE capitalizes scenarios to be considered during the design & validation phases of Advanced Driving Assistance Systems as well as Automated Driving systems into a common data base. Several data sources are used to fill in ADSCENE: near crash data, normal driving data, scenarios defined by experts in AD/S ADAS, scenarios from the regulations. This paper describes how accident data complements the scope of the other sources and the methodology used to extract and code relevant accidents into ADSCENE.<p />  <p>Language: en</p>
LA  - en
SN  - 2352-1465
UR  - http://dx.doi.org/10.1016/j.trpro.2023.11.316
ID  - ref1
ER  -