종·횡방향 통합제어를 위한 Laser scanner를 이용한 주변 차량 인식 및 추적

Abstract

The automotive industry has been concentrating efforts in the development of active safety systems such as CAS(Collision Avoidance System), LKAS(Lane Keeping Assistance System) etc. The integrated longitudinal-lateral control system is one of those systems that is combined ACC(Adaptive Cruise Control), CAS and lateral control system. In this system, the distance and relative velocity of surrounding vehicles are the key factors. This paper presents a method of detection and tracking of moving objects using a laserscanner. This method is able to classify objects as vehicles or other obstacles (like a guardrail, road signs, etc.) and can get information of the movement of target vehicles. It is composed by three steps: detection of objects; object classification using a feature extraction and target selection algorithm; and object tracking using a Kalman filter. The performance of this method is investigated via simulation using real data on driving.