자율주행자동차 검증을 위한 고위험 시나리오 개발 방법론에 관한 연구 (A Study on the Methodology for Developing High-Risk Scenarios for the Validation of Autonomous Vehi cles)

서지정보

· 발행기관 : 한국자동차공학회
· 수록지 정보 : 한국자동차공학회 논문집 / 32권 / 3호 / 301 ~ 308페이지
· 저자명 : 정승환

초록

The evaluation and validation processes of autonomous vehicles are as difficult and challenging as thedevelopment process. Autonomous vehicles need to perceive objects and combine them with road conditions, time, and seasons to create various verification and validation conditions. Additionally, autonomous vehicles cannot be validated by the NCAP(New Car Assessment Program) protocol defined for ADAS(Advanced Driver Assistance System) vehicles. Therefore, in this paper, we are proposing a methodology to generate reasonable high-risk scenarios that can be used inevaluating and validating the safety features of autonomous vehicles. The methodology consists of five steps, starting with acquiring an accident database and including the generation of relevant collision equations and estimated speed profiles. High-risk scenarios only include Vehicle-to-Vehicle driving situations, and can be used for the MILS(Model In the Loop Simulation) and VILS(Vehicle In the Loop Simulation) validation of autonomous vehicles after estimating driving speeds. The results of this study confirmed the generation of economical t est scenarios for autonomous vehicles and the reconstruction of reasonable driving situations.

영어초록

The evaluation and validation processes of autonomous vehicles are as difficult and challenging as thedevelopment process. Autonomous vehicles need to perceive objects and combine them with road conditions, time, and seasons to create various verification and validation conditions. Additionally, autonomous vehicles cannot be validated by the NCAP(New Car Assessment Program) protocol defined for ADAS(Advanced Driver Assistance System) vehicles. Therefore, in this paper, we are proposing a methodology to generate reasonable high-risk scenarios that can be used inevaluating and validating the safety features of autonomous vehicles. The methodology consists of five steps, starting with acquiring an accident database and including the generation of relevant collision equations and estimated speed profiles. High-risk scenarios only include Vehicle-to-Vehicle driving situations, and can be used for the MILS(Model In the Loop Simulation) and VILS(Vehicle In the Loop Simulation) validation of autonomous vehicles after estimating driving speeds. The results of this study confirmed the generation of economical t est scenarios for autonomous vehicles and the reconstruction of reasonable driving situations.

참고자료

· 없음