Method for generating guide lane for autonomous driving under atypical environment and system thereof

• Main Authors: MIN KYOUNG WOOK, CHOI JEONG DAN, CHOI DOO SEOP, PARK JAE HYUCK, JO YONG WOO
• Format: Patent
• Language: English; Korean
• Published: 22.06.2021
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE ORDIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; COUNTING; PERFORMING OPERATIONS; PHYSICS; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TOTHE CONTROL OF A PARTICULAR SUB-UNIT; TRANSPORTING; VEHICLES IN GENERAL; VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISEPROVIDED FOR

The present invention provides a method for generating a tracking line for autonomous driving so that a reliable tracking line for autonomous driving can be generated even in an atypical environment such as rain/snow in which lane recognition is impossible, comprising the following steps of: deriving a plurality of tracking line candidates using a deep artificial neural network based on image information and driving state data; analyzing a driving environment of own vehicle based on the image information and the driving environment data; evaluating suitability for each of the plurality of tracking line candidates in consideration of the analyzed driving environment; and tracking the tracking line based on an evaluation value for each evaluated tracking line candidate. 본 발명은 차선 인식이 불가능하거나 강우/강설 등의 비정형 환경에서도 신뢰도 높은 자율주행용 추종선을 생성할 수 있도록, 영상 정보 및 주행상태 데이터에 기초하여 심층인공신경망을 이용해 복수의 추종선 후보를 도출하는 단계와, 영상 정보 및 주행환경 데이터에 기초하여 자차의 주행환경을 분석하는 단계와, 상기 분석된 주행환경을 고려하여 상기 복수의 추종선 후보 각각에 대하여 적합성을 평가하는 단계와, 상기 평가된 각 추종선 후보별 평가값을 토대로 추종선을 추적하는 단계를 포함하는 자율주행용 추종선 생성 방법을 제공한다.