High curvature two-clothoid road model estimation

Abstract

Road recognition is of fundamental importance for many image understanding tasks involving cameras mounted on moving vehicles. We consider the well-known approach to road recognition based on a clothoid road model. In the case of high curvature roads this approach shows its limitations. We propose an extended road model that covers explicitly the possibility of having two different clothoid segments in the camera field of view. An algorithm estimating parameters of both clothoids and their transition point is presented and tested in a simulated environment.