Vehicle dynamics를 이용한 lane centering 제어기 개발

Abstract

The current LKAS(Lane keeping assist system) and LCC(Lane centering control) systems use low sampled vision sensor information for lateral motion control of vehicle. In this case, vehicle lateral motion can occur oscillation because sampling time of vision information (70㎳) is slower than sampling time of lateral motion controller (10㎳). Especially, LCC system need more smooth motion than LKAS because main agent of driving is not human but LCC system. In this study, we propose lane estimation method using acquired lane information which are lateral offset, heading angle, curvature, and curvature derivative and vehicle dynamics sensors which are longitudinal acceleration sensor, lateral acceleration sensor and wheel speed sensor. By lane estimation, we can compensate steering control input and expect more smoothness. The performance of the proposed system is verified with the commercial software.