Mixed Virtual-Real Testing of Comfort-Based Autonomous Driving Testing and Validation
- Delivery
- Available on this site
- Format
- Price
- Non-members (tax incl.):¥1,100 Members (tax incl.):¥880
- Publication code
- 20215225
- Paper/Info type
- Proceedings (Spring)
No.51-21
- Pages
- 1-6(Total 6 p)
- Date of publication
- May 2021
- Publisher
- JSAE
- Language
- English
- Event
- 2021 JSAE Annual Congress (Spring)[Online Meeting]
Detailed Information
Author(E) | 1) Son Tong, 2) Flavia Sofia Acerbo, 3) Ludovico Ruga, 4) Anoosh Hegde, 5) Theo Geluk, 6) Herman van der Auweraer |
---|---|
Affiliation(E) | 1) Siemens Digital Industries Software, 2) Siemens Digital Industries Software, 3) Siemens Digital Industries Software, 4) Siemens Digital Industries Software, 5) Siemens Digital Industries Software, 6) Siemens Digital Industries Software |
Abstract(E) | Currently, the focus of autonomous driving design and validation is safety. However, to increase occupant acceptance, ADAS should also provide a comfortable and human-like behavior. This shifts the goal of ADAS performance engineering to the correct balance of safety and comfort, such that these systems can provide both a risk-free and naturalistic driving style. Our goal is to obtain such behavior by learning from humans driving data. We first learn from real life collected traffic data. The designed system is then validated through a digital twin, made of vehicle model and virtual reconstruction of a real driving scenario. |