A robust linear-parabolic model for lane following

A Much Advanced and Efficient Lane Detection Algorithm for Intelligent Highway Safety

“…Then, the method is applied to identify the lane markings and remove other irrelevant objects in the image. To improve the performance, it was proposed using curvature method only in the far end of the image but adopting the straight line pattern in the near end to identify the lane markings in order to reduce the time required for identification [6]. Many approaches have been applied to lane detection, which can be classified as either feature-based or model based [7], [8].…”

An Improved Linear-Parabolic Model for Lane Following and Curve Detection

“…Jung and Kelber [10] In this work, we improve the linear-parabolic model by imposing constraints based on the geometry of the road, and explore the curvature of the parabolic part of the model to detect incoming curves. Next, the original and improved versions of the linear-parabolic model are explained.…”

“…In [10], both lane boundaries (left and right) are approximated by independent linear-parabolic functions, that are composed by linear functions in the near field, and parabolic functions in the far field. The expression for the linearparabolic model f k (x) can be written as 1 :…”

“…We propose an improvement to our linear-parabolic lane model [10], by imposing geometric constraints that relate left and right lane boundaries (in the original work, both lane boundaries are computed independently). We also use the second derivative of the parabolic part of the model to estimate incoming curves in the far field, indicating the driver if the road ahead is approximately straight, right-curved or left-curved.…”

An Integrated System of Driving Environment Recognition based on THASV-II

“…For example, the NavLab series intelligent vehicles by Carnegie Mellon University, USA [1]; the VaMP and VaMoRs intelligent vehicles by UBM, Germany [2]; the ARGO intelligent vehicle and its GOLD vision guidance system by University of Parma, Italy [3]; the THMR series intelligent vehicles of Tsinghua University, China [4], etc. Except for the above platforms in research, some products have been put into practice, including the AutoVue system by DaimlerChrysler Corporation, Germany and Iteris Corporation, USA together [8]; the LDW system by Nissan's north American corporation Infiniti; the STAR system by Toyoto Corporation, Japan, etc. But these systems mainly focus on the single application in lateral or longitudinal active safety systems without integration, which largely confines the application range.…”