http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
- 中文 을 입력하시려면 zhongwen을 입력하시고 space를누르시면됩니다.
- 北京 을 입력하시려면 beijing을 입력하시고 space를 누르시면 됩니다.
RISS 인기검색어
- 검색키워드
- 저자 : 윤서빈(Seo-Bin Yun) (검색결과 3 건)
- 무료
- 기관 내 무료
- 유료
-
윤서빈(Seo-Bin Yun),천정호(Jung-Ho Cheon),윤상석(Sang-Seok Yun) 대한인간공학회 2020 대한인간공학회 학술대회논문집 Vol.2020 No.10
Objective: The aim of this study is to allow the mobile robot to react quickly to make face-to-face interaction and to keep a distance between humans and robots. Background: Since the emergence of social robots that enable social behavior such as interacting with people and sharing daily communication and communion, social interaction between humans and robots has emerged as the most active area of research in robot technology. Method: It recognizes the human body joint and estimates the human body pose in space by calculating the relative coordinates of the body joint information. Based on the estimated human location, an omnidirectional robot is used to move to a target point that maintains a certain distance while facing the user. Results: In the user"s pose estimation, the angle error within 5 degrees was confirmed. In the experiment of moving the robot"s destination position, it was confirmed that the error was large at a position of 3m or more. Conclusion: The robot can move to a position where face-to-face interaction is possible by estimating the user"s pose by extracting the user"s joint coordinates based on the RGB-D sensor and calculating the geometric transfer function. Application: The proposed method can be utilized to perform triggering for dialogue between the user and the robot or to control the robot"s posture when the user moves.
-
휴먼 위치인식과 Pan-Tilt모듈을 이용한 객체 응시추적 시스템 구현
윤서빈(Seo-bin Yun),윤상석(Sang-Seok Yun) 한국정보통신학회 2019 한국정보통신학회 종합학술대회 논문집 Vol.23 No.2
본 논문은 실내공간에 장착된 RGB-D 센서를 이용하여 사용자의 위치를 추정한 후 Pan-Tilt모듈이 장착된 모바일 로봇을 이용하여 추적 대상자에게 접근하면서 대상자를 지속적으로 응시할 수 있는 객체 응시추적 시스템을 제안한다. RGB-D 센서로부터 추출된 이미지 정보에 Haar-cascade알고리즘을 사용하여 객체를 검출한 후 Depth 정보로부터 3차원 공간 내에서의 위치를 추정하게 되며, 기준좌표를 통한 3차원 좌표변환을 통하여 로봇은 휴먼객체의 상대적인 공간위치를 인식하게 된다. 4x3(m)의 실내공간에서 추적실험을 수행한 결과 움직이는 로봇에 설치된 Pan-Tilt모듈을 이용하여 휴먼 객체를 지속적인 응시가 가능하며, 안정적으로 추적이 가능함을 확인하였다. This paper uses RGB-D sensors mounted in interior space to estimate the user"s position, and then propose an object Examining Estimation System system that can continuously stare at the target while approaching the target using mobile robots equipped with Pan-Tilt modules. After detecting the object using the Haar-cascade algorithm on the image information extracted from the RGB-D sensor, the position of the object in the 3D space is estimated from the depth information. Through 3D coordinate transformation through reference coordinates, the robot recognizes the relative spatial position of the human object. As a result of experiment in the indoor space of 4x3(m), it was confirmed that the Pan-Tilt module installed in the mobile robot can stably track while continuously staring at the human object.
-
연관 검색어 추천
이 검색어로 많이 본 자료
활용도 높은 자료
