Multi-Class SVM을 이용한 다양한 주행상황에서의 CIPV 감지

Abstract

In this paper, we propose a multiple classification method that can detect the closest-in-path-vehicle (CIPV) and determine the various driving situations by using multi-class support vector machine (SVM) algorithm as a method of predicting the driving path of the object vehicle based on the vehicle behavior of the object vehicle. It will be helpful to improve the performance of adaptive cruise control (ACC). The performance of proposed method was validated via experimental results.