MODEL BASED DETECTION OF ROAD BOUNDARIES WITH A LASER SCANNER

In this paper, the authors describe a road boundaries detection system that is based on rage data. A laser scanner mounted on a vehicle is used to measure distances and detect road boundaries. The range data is used to estimate the parameters of a model of the road boundaries. Kalman filtering is used to process successive scans. Test results indicate that the road boundaries can be reliably detected as long as guardrails or posts are available.