The present invention provides an apparatus and a method for integrally determining the condition of a driver, reducing complexity of calculation and required time using a hybrid dynamic Bayesian network method, integrally determining drowsiness, inattention and overload of a driver by considering the image and voice of the driver, bio-information on the driver, information on the vehicle, states of sensors and data reliability, and reducing a false alarm rate regarding determination of the condition of the driver, thereby improving the safety of the driver′s performance. The apparatus for integrally determining the condition of a driver comprises a driver condition determination part and a driver model result processing part. The driver condition determination part determines the condition of the driver, including wakefulness, inattention and overload, based on sensing information input by a sensor part and a driver′s inclination. The driver electronic model result process part classifies the determined driver condition information according to each type of sensing information to be congregated and estimates an integral condition of the driver by calculating the probability of the occurrence of drowsiness, inattention and overload (reason) using the hybrid dynamic Bayesian network defined in such a manner that a result (maintaining a speed and including the degree of shutting of eyes) by a reason (including drowsiness, inattention and overload) is arranged with a conditional probability table by defining a causal relationship between respective nodes of a Bayesian network layer.

    본 발명은 하이브리드 동적 베이지안 네트워크 방법을 사용하여 계산의 복잡성과 소요시간을 줄이고, 운전자의 영상, 음성, 생체 정보와 차량 정보 외에 센서들의 상태 및 데이터 신뢰도를 고려하여, 운전자의 졸음, 부주의 및 과부하를 통합 판단하고, 운전자 상태 판단에 대한 오경보율을 감소시킴으로써, 운전자의 주행 안전에 도움을 줄 수 있도록 하는 운전자 상태 통합 판단 장치 및 방법을 제공하기 위한 것으로서, 센서부로부터 입력되는 센싱 정보들 및 운전 성향을 기반으로 운전자의 각성도, 부주의, 과부하를 포함하는 운전자 상태를 판단하는 운전자 상태 판단부와, 상기 판단된 운전자 상태 정보를 이용하여 센싱 정보별로 분류하여 군집화하고, 베이지안 네트워크 층의 각 노드 간 인과관계를 정의하여 원인(졸음, 부주의, 과부하를 포함)에 의한 결과(속도 유지, 눈 감김 정도를 포함)를 조건부 확률표로 정리하여 정의된 하이브리드 동적 베이지안 네트워크를 이용하여 졸음, 부주의 및 과부하(원인)가 발생될 확률을 계산하여 운전자 통합 상태를 추정하는 운전자 모델결과 처리부를 포함하여 구성되는데 있다.


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    INTEGRATED ASSESSMENT APPARATUS AND METHOD OF DRIVERS' DROWSINESS, INATTENTION AND WORKLOAD


    Weitere Titelangaben:

    운전자 졸음, 부주의 및 과부하 통합 판단 장치 및 방법


    Beteiligte:
    YANG JI HYUN (Autor:in) / RYU DONG WOON (Autor:in)

    Erscheinungsdatum :

    2016-04-05


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Koreanisch


    Klassifikation :

    IPC:    B60K Anordnung oder Einbau von Antriebseinheiten oder von Kraft- bzw. Drehmomentübertragungen in Fahrzeugen , ARRANGEMENT OR MOUNTING OF PROPULSION UNITS OR OF TRANSMISSIONS IN VEHICLES



    On the concept and measurement of driver drowsiness, fatigue and inattention: implications for contermeasures

    Gimeno,P.T. / Cerezuela,G.P. / Montanes,M.C. et al. | Kraftfahrwesen | 2006


    On the concept and measurement of driver drowsiness, fatigue and inattention: implications for countermeasures

    Gimeno, Pilar Tejero / Cerezuela, Gemma Pastor / Montanes, Mariano Choliz | Tema Archiv | 2006



    INATTENTION DETERMINATION DEVICE, INATTENTION DETERMINATION SYSTEM, INATTENTION DETERMINATION METHOD, AND STORAGE MEDIUM

    SAKUMA NANA / TSUKAHARA HIDENORI / SHIMIZU YUTA | Europäisches Patentamt | 2019

    Freier Zugriff

    Drivers drowsiness detection in embedded system

    Tianyi Hong, / Huabiao Qin, | IEEE | 2007