The present invention relates to a device and a method for controlling autonomous driving based on artificial intelligence VILS, and more particularly, to a technology for verifying a road operation of an autonomous vehicle through an operation of a real road by generating a realistic driving environment through a generative adversarial network (GAN) model. The device of the present invention comprises: a driving environment generating unit for generating a virtual environment through status information indicating a modeled road environment required for verification and driving information including information on a surrounding vehicle of an autonomous vehicle, in order to autonomously drive the autonomous vehicle in a virtual environment; a result generating unit for databasing the status information and the driving information acquired by the driving environment generating unit by using layer classification through a three-dimensional occupancy grid map (OGM); and an autonomous driving verification unit for controlling the autonomous vehicle by operating an autonomous driving algorithm calculated from data generated by the result generating unit, and tracking and recording an actual moving trajectory of the autonomous vehicle to input the same to the virtual environment. According to the present invention, the reliability on autonomous driving evaluation can be improved by using an advanced artificial intelligent driver simulation model which learns actual driving of various drivers for VILS verification.

    본 발명은, 인공지능 VILS 기반의 자율주행 제어 장치 및 방법에 관한 것으로, 특히 생산적 적대 학습(GAN: Generative Adversarial Network) 모델을 통해 실제와 같은 주행 환경을 생성하여 실도로의 운행을 통해 자율주행차의 도로 운행을 검증하는 기술에 관한 것이다. 본 발명은, 가상 환경에서 자율주행차를 자율 주행시키기 위해, 검증이 필요한 모델링된 도로 환경을 나타내는 상태 정보와 상기 자율주행차의 주변 차량에 대한 정보를 포함하는 주행 정보를 통해 상기 가상 환경을 생성하는 주행환경 생성부; 상기 주행환경 생성부에서 획득된 상태 정보 및 주행 정보를 3차원 OGM(Occupancy Grid Map)을 통해 레이어 분류를 이용하여 데이터화를 진행하는 결과물 생성부; 및 상기 결과물 생성부에서 생성된 데이터로부터 산출된 자율주행 알고리즘을 작동시켜 상기 자율주행차를 제어하고 상기 자율주행차의 실제의 움직이는 궤적을 추적 및 기록하여 상기 가상 환경에 입력하는 자율주행 검증부를 포함하는 인공지능 VILS 기반의 자율주행 제어 장치를 제공한다. 본 발명에 의해, 실제 다양한 운전자들의 주행을 학습한 고도화된 인공지능 운전자 모사 모델을 VILS 검증에 활용함으로써, 자율주행 평가의 신뢰성을 높일 수 있을 것이다.


    Access

    Download


    Export, share and cite



    Title :

    VILS AUTONOMOUS DRIVING CONTROL APPARATUS AND METHOD BASED ON AI VEHICLE IN THE LOOP SIMULATION


    Additional title:

    인공지능 VILS 기반의 자율주행 제어 장치 및 방법


    Contributors:
    LIM SE JOON (author) / YUN YEONG RO (author)

    Publication date :

    2020-08-12


    Type of media :

    Patent


    Type of material :

    Electronic Resource


    Language :

    Korean


    Classification :

    IPC:    B60W CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION , Gemeinsame Steuerung oder Regelung von Fahrzeug-Unteraggregaten verschiedenen Typs oder verschiedener Funktion / B60K Anordnung oder Einbau von Antriebseinheiten oder von Kraft- bzw. Drehmomentübertragungen in Fahrzeugen , ARRANGEMENT OR MOUNTING OF PROPULSION UNITS OR OF TRANSMISSIONS IN VEHICLES / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen



    RSU VILS METHOD AND SYSTEM FOR EVALUATING AUTONOMOUS DRIVING BASED ON VILS USING RSU

    JUNG JIWON / HONG JUN | European Patent Office | 2025

    Free access

    CONTROL PLATFORM INTERWORKING VILS-BASED TRAFFIC ENVIRONMENT REPRODUCING METHOD AND SYSTEM

    JUNG JIWON / HONG JUN | European Patent Office | 2024

    Free access

    VILS SYSTEM BASED ON REAL ROAD ENVIRONMENT AND METHOD THEREOF

    European Patent Office | 2025

    Free access


    Longitudinal Control for closed loop simulation of Autonomous driving Vehicle

    Kumbhare, Bhuvana / Akant, Kalyani / Khanapurkar, Milind et al. | IEEE | 2023