충돌 회피 시스템을 위한 상대 차량 주행 경로 예측

Abstract

Prediction of target vehicle’s path is important for collision avoidance system. Path prediction methods can be separated by physics based model prediction and maneuver based model prediction. Physics based model prediction is appropriate for short term prediction since the sensitive to sensor noise. On the other hand, maneuver based model prediction considers the driver’s intention. In this paper, integration of physics based model and maneuver based model is proposed for target vehicle path prediction. The purpose of the paper is to more robust to noise and it guarantees prediction accuracy in long term. The algorithm can be used the measurable data by exteroceptive sensors. The proposed algorithm is validated by using the computer simulation software, CarSim and MATLAB/Simulink.