Automatic steering control algorithm has been proposed, which uses the motion of objects in visual image (optical flow), obtained from an ITV camera looking ahead in the situation without forward vehicle. This algorithm is improved to be applicable to the situation in which the forward tracking course is invisible owing to the interference of the forward vehicles. The algorithm is based on image processing and requires little computation. Visual information is defined as the image of the road as projected on the front windshield (image) and its motion (dynamic image). Therefore, information is excluded such as actual positions on the road as measured with respect to an external reference frame. As a result, the motion of the course in the forward field of view is perceived in terms of the dynamic image projected on the windshield, rather its physical values. The basic principle for the algorithm is to eliminate the sum of the shift in lane partition lines and lateral deviation. When there is a vehicle ahead obscuring the partition lines, one can use the optical flow of the preceding vehicle to compensate for the loss of partition lines. The adjustment of the control parameters against the change in vehicle speed was confirmed by computer simulation experiments.
Course tracking control algorithm using visual information
Fahrbahnverfolgungsregelung unter Verwendung visueller Information
Vehicle System Dynamics ; 28 , 6 ; 385-398
1997
14 Seiten, 14 Bilder, 2 Tabellen, 3 Quellen
Article (Journal)
English
Course Tracking Control Algorithm using Visual Information
British Library Online Contents | 1996
|Course Tracking Control Algorithm Using Visual Information
Online Contents | 1997
|Course tracking control algorithm using visual information
Automotive engineering | 1997
|Course Tracking Control Algorithm Using Visual Information
Taylor & Francis Verlag | 1997
|A course tracking control algorithm using visual information
Automotive engineering | 1996
|