Road and obstacle detection research based on four-line ladar

In order to achieve the goal of fast and accurate perception of external environment in structured road, this paper proposes an algorithm of curb and obstacle detection for intelligent vehicle in outdoor environment. It first detects curb using the improved algorithm combined Hough Transform with least-squares method in this paper to acquire the objects scanned in the road of interest, which removes the points scanned on the ground. Then based on the remaining data, it clusters taking advantage of the improved method combined DBSCAN with K-Means method, which overcomes DBSCAN's defect that it couldn't divide obstacles with similar density. At the same time, this method can eliminate noise points effectively, playing the role of filter. Finally, after cluster vehicle can acquire obstacle's information, such as angle, distance, size and so on, to complete the task of obstacle detection. The algorithm presented in this paper has been applied to obstacle detection in our vehicle. The test proved that it is consistent and reliable, which meets the needs of autonomous driving of intelligent vehicles.