• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2012.05a
• /
• pp.319-320
• /
• 2012

• The Journal of the Korea institute of electronic communication sciences
• /
• v.8 no.4
• /
• pp.583-588
• /
• 2013

In this paper we propose implement technique of vector current control in order to verify performance of an AC servo driver that is able to easy control of motion with multi-axis in the robot. In doing do, we have developed the AC servo driver to driving PMSM, and then we confirm that this driver whether operating or not normally by controlling of vector current. The vector current control was performed at the no load condition in PMSM. Then we compare command control and tracking control. As a result of verification, we recognize we get a satisfactory result.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2006.05a
• /
• pp.709-712
• /
• 2006

다양한 종류의 컴퓨터가 사람, 사물, 환경 속에 내재되어 있고, 이들이 서로 연결되어, 필요한 곳에서 활용할 수 있는 유비쿼터스 환경에서는 홈 네트워크를 통해 이 기종 기기간 다양한 데이터 교환을 요구한다. 더욱이 원활한 영상 데이터의 처리, 전송, 모니터링 기술은 핵심적 요소가 아닐 수 없다. 공간 및 시간적인 해상도, 컬러의 표현 그리고 화질의 측정방법 등 고전적 영상 처리 연구 분야뿐만 아니라 국한된 대역폭을 갖는 홈네트워크의 전송체계에서 전송률 문제에 대한 심도 있는 연구가 필요하다. 본 논문에서는 홈네트워크 상황에서 콘텐츠의 중심이 되는 영상 데이터의 전송과 처리 그리고 제어를 위하여 새로운 움직임 추정 알고리즘을 제안한다. 각도, 거리등 다양한 환경에서 전송되어지는 스테레오 카메라의 영상데이터들은 축소, 확대, 이동, 보정 등 전처리 후 제안된 변형계층 모션벡터 추정 알고리즘을 이용하여 압축 처리, 전송된다. 기존 모션벡터 추정 알고리즘의 장점을 계승하고 단점을 보완한 변형계층 알고리즘은 비정형, 소형 매크로 블록을 이용하여 휘도의 편차가 큰 영상의 효율적 움직임 추정에 이용된다. 본 논문에서 제안한 변형계층 알고리즘과 이를 이용해 구현된 영상시스템은 유비쿼터스 환경에서 다양하게 활용될 수 있다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2008.06d
• /
• pp.444-448
• /
• 2008

The ubiquitous smart home is the home of the future that takes advantage of context information from the human and the home environment and provides an automatic home service for the human. Human location and motion are the most important contexts in the ubiquitous smart home. We present a real-time human tracker that predicts human location and motion for the ubiquitous smart home. We used four network cameras for real-time human tracking. This paper explains the real-time human tracker's architecture, and presents an algorithm with the details of two functions (prediction of human location and motion) in the real-time human tracker. The human location uses three kinds of background images (IMAGE1: empty room image, IMAGE2:image with furniture and home appliances in the home, IMAGE3: image with IMAGE2 and the human). The real-time human tracker decides whether the human is included with which furniture (or home appliance) through an analysis of three images, and predicts human motion using a support vector machine. A performance experiment of the human's location, which uses three images, took an average of 0.037 seconds. The SVM's feature of human's motion recognition is decided from pixel number by array line of the moving object. We evaluated each motion 1000 times. The average accuracy of all the motions was found to be 86.5%.

• The KIPS Transactions:PartA
• /
• v.16A no.3
• /
• pp.209-216
• /
• 2009

The ubiquitous smart home is the home of the future that takes advantage of context information from the human and the home environment and provides an automatic home service for the human. Human location and motion are the most important contexts in the ubiquitous smart home. We present a real-time human tracker that predicts human location and motion for the ubiquitous smart home. We used four network cameras for real-time human tracking. This paper explains the real-time human tracker's architecture, and presents an algorithm with the details of two functions (prediction of human location and motion) in the real-time human tracker. The human location uses three kinds of background images (IMAGE1: empty room image, IMAGE2: image with furniture and home appliances in the home, IMAGE3: image with IMAGE2 and the human). The real-time human tracker decides whether the human is included with which furniture (or home appliance) through an analysis of three images, and predicts human motion using a support vector machine. A performance experiment of the human's location, which uses three images, took an average of 0.037 seconds. The SVM's feature of human's motion recognition is decided from pixel number by array line of the moving object. We evaluated each motion 1000 times. The average accuracy of all the motions was found to be 86.5%.

• Journal of the Korea Computer Graphics Society
• /
• v.26 no.2
• /
• pp.11-19
• /
• 2020

When a character's pose changes, its center of mass(COM) also changes. The change of COM has distinctive patterns corresponding to various motion types like walking, running or sitting. Thus the motion type can be predicted by using COM movement. We propose a motion generator that uses character's center of mass information. This generator can generate various motions without annotated action type labels. Thus dataset for training and running can be generated full-automatically. Our neural network model takes the motion history of the character and its center of mass information as inputs and generates a full-body pose for the current frame, and is trained using simple Convolutional Neural Network(CNN) that performs 1D convolution to deal with time-series motion data.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.7 no.2
• /
• pp.409-415
• /
• 2012

Recently in industry communication, the efforts that use robot and automation system increased by cooperation with optical communication and industrial Ethernet. In this paper, in order to solve the problem that total network blocking when network fault happens and wiring problem of optical cable, which were big serious disadvantage in industrial optical network systems, we propose bi-directional optical EtherCAT communication technique based on single optical core, which applying WDM method. We describe the content for implementation of WDM bi-directional optical EtherCAT communication module and performance evaluation to verify the performance of related technology as a whole.

• Science of Emotion and Sensibility
• /
• v.8 no.3
• /
• pp.265-276
• /
• 2005

현재의 유비쿼터스 환경의 홈 네트워크 제품 사용자는 단일 사용자가 아닌 다수의 사용자가 사용하는 네트워크 행태를 취하고 있다. 변화하는 사용환경과 시스템들은 현재와는 다른 요구사항을 가지고 있으며, 이에 따른 사용자 중심의 디자인과 제품 인터페이스 체계의 연구활동은 국내외에서 활발하게 이루어지고 있다. 다양한 모바일 디바이스 및 홈 네트워크 제품의 보급화가 빠르게 성장하면서 이를 쉽게 제어하기 위한 다양한 제어방식이 연구되고 있다. 이중 음성인식기술을 비롯한 표정은 안면표정인식기술의 개발이 활발히 진행되고 있다. 모션감지 센서를 활용한 사용자 제스처 콘트롤 체계는 아직까지는 초보적인 단계에 있으나, 제품 제어에 있어서 향후 근미래에는 자연스러운 인터랙티브 인터페이스의 활용도가 높아질 전망이다. 이에 본 연구에서는 효과적인 디바이스 제어를 위한 제스처 유형의 자연스러운 사용언어체계 개발 방법 및 결과 그리고 사용자 맨탈모델와 메타포 실험을 통한 연구내용을 정리하였다. 기존 사용자의 제스처 유형의 자연스러운 사용언어를 분석하면서 디바이스 제어방식으로서 활용 가능성을 검토할 수 있었으며, 동작 감지 카메라 및 센서를 활용한 새로운 디바이스 제어방식 개발과정의 연구를 통하여 제스처 유형의 자연스러운 언어 체계 개발 및 과정을 정립하였다.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.38 no.2
• /
• pp.211-218
• /
• 2014

Recently, software-based programmable logic controller (PLC) systems, which are implemented in standard PLC languages on general hardware, are gaining popularity because they overcome the limitations of classical hardware PLC systems. Another noticeable trend is that the use of integrated development environment (IDE) is becoming important. IDEs can help developers to easily manage the growing complexity of modern control systems. Furthermore, industrial Ethernet, e.g. EtherCAT, is becoming widely accepted as a replacement for conventional fieldbuses in the distributed control domain because it offers favorable features such as short transmission delay, high bandwidth, and low cost. In this paper, we implemented the extension of open source IDE, called Beremiz, for developing EtherCAT-based real-time, synchronized motion control applications. We validated the EtherCAT system management features and the real-time responsiveness of the control function by using commercial EtherCAT drives and evaluation boards.

• Journal of IKEEE
• /
• v.23 no.4
• /
• pp.1295-1301
• /
• 2019

In this paper, we propose a novel method of denoising human motion using a bidirectional recurrent neural network (BRNN) with an attention mechanism. The corrupted motion captured from a single 3D depth sensor camera is automatically fixed in the well-established smooth motion manifold. Incorporating an attention mechanism into BRNN achieves better optimization results and higher accuracy than other deep learning frameworks because a higher weight value is selectively given to a more important input pose at a specific frame for encoding the input motion. Experimental results show that our approach effectively handles various types of motion and noise, and we believe that our method can sufficiently be used in motion capture applications as a post-processing step after capturing human motion.