장애물의 위험도를 반영한 차량의 장애물 회피 경로생성 알고리즘

Abstract

This paper proposes path planning algorithm for avoiding obstacles. The proposed algorithm classifies risk level of each obstacle and weight of the obstacle is determined based on the risk level. Path is generated using the weights of obstacles and basic principle of support vector machine (SVM). This method is able to enhance not only safety of autonomous vehicle driving but also correspondence of path planning in various scenarios. Proposed algorithm was simulated by MATLAB/Simulink, Carsim and Prescan was used to compose the simulation environment.