Share Email Print

Proceedings Paper

Real-time 3D reconstruction of road curvature in far look-ahead distance from analysis of image sequences
Author(s): Reinhold Behringer
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

A system for visual road recognition in far look-ahead distance, implemented in the autonomous road vehicle VaMP (a passenger car), is described. Visual cues of a road in a video image are the bright lane markings and the edges formed at the road borders. In a distance of more than 100 m, the most relevant road cue is the homogeneous road area, limited by the two border edges. These cues can be detected by the image processing module KRONOS applying edge detection techniques and areal 2D segmentation based on resolution triangles (analogous to a resolution pyramid). An estimation process performs an update of a state vector, which describes spatial road shape and vehicle orientation relative to the road. This state vector is estimated every 40 ms by exploiting knowledge about the vehicle movement (spatio-temporal model of vehicle dynamics) and the road design rules (clothoidal segments). Kalman filter techniques are applied to obtain an optimal estimate of the state vector by evaluating the measurements of the road border positions in the image sequence taken by a set of CCD cameras. The road consists of segments with piecewise constant curvature parameters. The borders between these segments can be detected by applying methods which have been developed for detection of discontinuities during time-discrete measurements. The road recognition system has been tested in autonomous rides with VaMP on public Autobahnen in real traffic at speeds up to 130 km/h.

Paper Details

Date Published: 1 December 1995
PDF: 8 pages
Proc. SPIE 2646, Digital Photogrammetry and Remote Sensing '95, (1 December 1995); doi: 10.1117/12.227865
Show Author Affiliations
Reinhold Behringer, Univ. der Bundeswehr Muenchen (Germany)

Published in SPIE Proceedings Vol. 2646:
Digital Photogrammetry and Remote Sensing '95
Eugeny A. Fedosov, Editor(s)

© SPIE. Terms of Use
Back to Top