Robust real-time lane marking detection for intelligent vehicles in urban environment

In this paper, we propose a framework of robust and real-time lane marking detection and tracking for the autonomous driving of intelligent vehicle under urban road environments. Our framework use hyperbola model as the lane marking model and several adaptive techniques to robustly extract lane marking pairs for the current lane. The framework provides a complete solution from environment perception of intelligent vehicle to vehicle control. The experimental results achieved on urban road scenarios show that our system performs well under various situation. © 2011 Springer-Verlag.