Improving driver's visual field using estimation of curvature

Abstract

When driving at night, it is difficult to detect obstacles such as pedestrian and wild animals on the opposite lane which is out of sight. Adaptive front lighting system (AFS) can be used to avoid accidents with obstacles that are difficult to detect with a conventional lighting system. The basic concept of AFS is to collect the information such as vehicle speed, yaw rate, steering angle, suspension height and driving mode from vehicle sensors. This information is applied to calculate an optimal angle of irradiation and used to change beam pattern and position according to the current radius of curvature. However, the driver may have to turn quickly on a sharp curve, in which the optimal angle of irradiation cannot be achieved before entering the opposite lane because swivel mode on AFS reacts based on the current radius of curvature. This paper presents a method to improve driver's visual field by using the pre-calculated curvature information on the road ahead. The effectiveness of the proposed method is evaluated with a set of experiments in the real environment.