Indirect Driver Monitoring by Online Estimation of Driver Model Parameters
- Delivery
- Available on this site
- Format
- Price
- Non-members (tax incl.):¥1,100 Members (tax incl.):¥880
- Publication code
- 20219067
- Paper/Info type
- Other International Conferences
- Pages
- 1-3(Total 3 p)
- Date of publication
- Sep 2021
- Publisher
- JSAE
- Language
- English
Detailed Information
Author(E) | 1) Hironori Suzuki, 2) Kakeru Fujiwara |
---|---|
Affiliation(E) | 1) Nippon Institute of Technology, 2) Nippon Institute of Technology |
Abstract(E) | Simultaneously with the rapid development of autonomous driving technologies, increasing attention is being paid to monitoring systems aimed at improving interactions between human drivers and autonomous machines. This technical paper proposes a new approach to producing online estimations of driver model parameters that can provide the foundation for an indirect driver monitoring system. Assuming that improper driver behaviors resulting from fatigue, drowsiness, distraction, etc., are all reflected in car-following behaviors, we developed a method of identifying model parameters for analyzing driver behaviors that are based on a dual particle filter. The numerical analyses conducted in conjunction with this study show that our proposed model can precisely identify model parameters and driver behaviors, including those related to car-following stability, and thus can provide the basis for a fundamentally new approach to driver monitoring systems. |