A Lane Recognition Method Using the Tangent Vectors of White Lane Markers

摘要

This paper describes a method for improving the robustness of lane recognition by image processing. A Genetic Algorithm has been constructed that makes use of the tangent vectors of lane markers projected in images of the road at a distance from the host vehicle. When this method was used in a simulation it was confirmed that it provides robust lane recognition even in the case of a road surface having other white lines, such as broken lines, running parallel to the lane markers.