Abstract

Detection of road obstacles is essential for autonomous driving of intelligent vehicles on highways and urban roads. In this paper, we present a robust and accurate road obstacle detection method using 3D lidar data. First, we acquire a stable and accurate description of road curbs through double validations. Then, a RANSAC and least squares road plane fitting procedure is performed. Finally, obstacles including cars, trucks, pedestrians, and bikes are detected based on the road plane. Experimental results show good performances of the proposed method on various road situations.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call