Neural Network Representation of Visual-Motor Coordination at Vehicle Driving
- Delivery
- Available on this site
- Format
- Price
- Non-members (tax incl.):¥1,100 Members (tax incl.):¥880
- Publication code
- 923088
- Paper/Info type
- AVEC
- Pages
- 493-498(Total 6 p)
- Date of publication
- Sep 1992
- Publisher
- Others, Unknown
- Language
- English
- Event
- AVEC '92
Detailed Information
| Author(E) | 1) Jeong Soo Shim, 2) Seung Jin Heo, 3) Yong San Yoon |
|---|---|
| Affiliation(E) | 1) Korea Academy of Industrial Tech., 2) Korea Academy of Industrial Tech., 3) KAIST |
| Abstract(E) | In this paper, a model of the driver-vehicle-road system is suggested taking the visual data processing method. The learning capability of the curved roads is devised using neural network technique. The learned network can extract the road curvature and the sight angle relevant to it from the frontal viewing image. Then the steering wheel angle is calculated by the extracted informations. This method has some merits such as easy implementation to the driver-vehicle simulation and realistic representation of the visual-motor coordination. For the computer simulation of the driver-vehicle-road system, a mathematical model of vehicle dynamics is established, where 3-dimensional motion and suspension kinematics of the vehicle system are considered using the suspension super-element concept. To validate the model of the driver-vehicle-road system, a real driving test is performed, and the results are compared with the ones from the model simulation on the computer. |