• 아동학회지
• /
• 제32권4호
• /
• pp.1-14
• /
• 2011

The purpose of this study was to investigate young children's perceptions of a teacher assistant robot, IrobiQ. in a kindergarten classroom. The subjects of this study were 23 6-year-olds attending to G kindergarten located in E city, Korea, where the teacher assistant robot had been in operation since Oct. 2008. Each child responded to questions assessing the child's perceptions of IrobiQ's identity regarding four domains : it's biological, intellectual, emotional and social identity. Some questions asked the child to affirm or deny some characteristics pertaining to the robot and the other questions asked the reasons for the answer given. The results indicated that while majority of children considered an IrobiQ not as a biological entity, but as a machine, they thought it could have an emotion and be their playmate. The implications of these results are two folds : firstly, they force us to reconsider the traditional ontological categories regarding intelligent service robots to understand human-robot interaction and secondly, they open up an ecological perspective on the design of teacher assistant robots for use with young children in early childhood education settings.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제1권1호
• /
• pp.19-31
• /
• 2007

RFID tag is detected by an RFID antenna and information is read from the tag detected, by an RFID reader. RFID tag detection by an RFID reader is very important at the deployment stage. Tag detection is influenced by factors such as tag direction on a target object, speed of a conveyer moving the object, and the contents of an object. The water content of the object absorbs radio waves at high frequencies, typically approximately 900 MHz, resulting in unstable tag signal power. Currently, finding the best conditions for factors influencing the tag detection requires very time consuming work at deployment. Thus, a quick and simple RFID tag detection scheme is needed to improve the current time consuming trial-and-error experimental method. This paper proposes a back-propagation learning-based RFID tag detection prediction scheme, which is intelligent and has the advantages of ease of use and time/cost savings. The results of simulation with the proposed scheme demonstrate a high prediction accuracy for tag detection on a water content, which is comparable with the current method in terms of time/cost savings.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제7권11호
• /
• pp.2824-2838
• /
• 2013

There has been a lot of attention paid recently to analyze dynamic human gestures that vary over time. Most attention to dynamic gestures concerns with spatio-temporal features, as compared to analyzing each frame of gestures separately. For accurate dynamic gesture recognition, motion feature extraction algorithms need to find representative features that uniquely identify time-varying gestures. This paper proposes a new feature-extraction algorithm using temporal self-similarity based on a hierarchical human model. Because a conventional temporal self-similarity method computes a whole movement among the continuous frames, the conventional temporal self-similarity method cannot recognize different gestures with the same amount of movement. The proposed model-based temporal self-similarity method groups body parts of a hierarchical model into several sets and calculates movements for each set. While recognition results can depend on how the sets are made, the best way to find optimal sets is to separate frequently used body parts from less-used body parts. Then, we apply a multiclass support vector machine whose optimization algorithm is based on structural support vector machines. In this paper, the effectiveness of the proposed feature extraction algorithm is demonstrated in an application for taebo gesture recognition. We show that the model-based temporal self-similarity method can overcome the shortcomings of the conventional temporal self-similarity method and the recognition results of the model-based method are superior to that of the conventional method.

• 지능정보연구
• /
• 제5권1호
• /
• pp.95-101
• /
• 1999

In this paper, a comparison between different methods to combine predictions from neural networks will be given. These methods are bagging, bumping, and balancing. Those are based on the analysis of the ensemble generalization error into an ambiguity term and a term incorporating generalization performances of individual networks. Neural Networks and AI machine learning models are prone to overfitting. A strategy to prevent a neural network from overfitting, is to stop training in early stage of the learning process. The complete data set is spilt up into a training set and a validation set. Training is stopped when the error on the validation set starts increasing. The stability of the networks is highly dependent on the division in training and validation set, and also on the random initial weights and the chosen minimization procedure. This causes early stopped networks to be rather unstable: a small change in the data or different initial conditions can produce large changes in the prediction. Therefore, it is advisable to apply the same procedure several times starting from different initial weights. This technique is often referred to as training ensembles of neural networks. In this paper, we presented a comparison of three statistical methods to prevent overfitting of neural network.

• 한국정보과학회논문지:컴퓨팅의 실제 및 레터
• /
• 제10권1호
• /
• pp.1-13
• /
• 2004

시맨틱 웹은 지능적인 서비스를 제공하기 위한 새로운 웹 환경이다. 시맨틱 웹을 구현하기 위해 많은 새로운 노력들이 진행되어 왔다. 이 중에서 가장 기본적이고 중요한 것은 의미 정보를 표현, 저장, 질의, 관리해 주는 온톨로지 관리 시스템이다. 우리는 이와 같은 기능을 제공하기 위한 새로운 시스템으로서, 토픽맵 모델을 기반으로 한 효율적인 온톨로지 관리 시스템인 K-Box를 구현하였다. K-Box는 온톨로지 관리를 위한 기본적인 기능들을 제공하며, 이질적인 저장소들을 일관된 인터페이스로 접근할 수 있도록 함으로써 저장 장치 독립성을 제공하였다. 또한, 저장 관리되는 모든 온톨로지들의 무결성을 보장하기 위한 새로운 기법과 사용자 관심을 중심으로 한 온톨로지 검색 지원을 위한 방법을 제안하였다. 마지막으로, 우리는 여러 온톨로지들을 적용해 봄으로써 K-Box 시스템이 효율적으로 사용 가능함을 확인하였다.

• 한국기계가공학회지
• /
• 제17권3호
• /
• pp.134-140
• /
• 2018

This paper describes the design and fabrication of an automatic alignment and loading system for workpieces. To move a workpiece to a chucking position of a machine tool using a gantry robot, an automatic aligning device is required to load the workpiece before machining and automatically align them. The automatic alignment system was conceptually designed, and the structural analysis was performed for the main parts such as the top plate, center support, front and back support, and support shaft. Based on the structural analysis results, the size of these structures was determined. The automatic alignment system was manufactured, and the vertical movement characteristics of the workpiece up-and-down movement device and the rotation characteristics of the workpiece rotation device were experimentally examined. The result has confirmed that they operate normally.

• 한국공작기계학회논문집
• /
• 제13권4호
• /
• pp.44-55
• /
• 2004

In order to meet the requirement for higher thickness accuracy in tandem cold rolling processes, it is strongly necessary to have good performance in control units. To meet this requirement, this paper suggested an output regulating control system with a roll-eccentricity estimator for each rolling stand of tandem cold mills. Considering entry thickness variation and roll eccentricity simultaneously as the major disturbances, a synthesis of multivariable control systems was presented based on H$\infty$ control theory, which could reflect the knowledge of input direction and spectrum of disturbance signals on design. Then, to effectively reject roll eccentricity, a weight function having some poles on the imaginary axis was introduced. This lead to a non-standard H$\infty$ control problem, and the design procedures for solving this problem were analytically presented. The effectiveness of the proposed control method was evaluated through computer simulations and compared to that of the conventional linear quardratic control and feedforward control methods for roll eccentricity.

• 한국정밀공학회지
• /
• 제29권4호
• /
• pp.373-379
• /
• 2012

Disruptions in manufacturing systems caused by system changes and disturbances such as the tool wear, machine breakdown, malfunction of transporter, and so on, reduce the productivity and the increase of downtime and manufacturing cost. In order to cope with these challenges, a new method to build an intelligent manufacturing system with biological principles, namely an ant colony inspired manufacturing system, is presented. In the developed system, the manufacturing system is considered as a swarm of cognitive agents where work-pieces, machines and transporters are controlled by the corresponding cognitive agent. The system reacts to disturbances autonomously based on the algorithm of each autonomous entity or the cooperation with them. To develop the ant colony inspired manufacturing system, the disturbances happened in the machining shop of a transmission case were analyzed to classify them and to find out the corresponding management methods. The system architecture with the autonomous characteristics was generated with the cognitive agent and the ant colony technology. A test bed was implemented to prove the functionality of the developed system.

• 한국조명전기설비학회지:조명전기설비
• /
• 제9권6호
• /
• pp.41-48
• /
• 1995

본 연구에서는 Extended Kalman Filter(EKF)를 이용한 속도센서없는 우도전동기의 벡터제어의 구현을 제안하였다. 또한 회전자 저항의 변동을 보상 할 수 있도록 회전자 저항도 추정한다. 이산화된 유도전동기의 모델을 통해 유도 전동기의 속도와 회전자 저항을 포함한 상태 변수를 정의하고 벡터 제어에 필요한 자속각을 추정하여 노이즈 환경에 놓인 시스템의 동작 특성을 안정되게 하였다. EKF알고리즘의 연산을 위하여 DSP를 이용하고, 전류제어 장치로 공간 전압벡터 변조 방식의 적용이 용이한 마이크로 콘트롤러를 체용하고, 인버터는 IPM(Intelligent Power Module)으로 실험 장치를 구성하였다. 시뮬레이션과 실험을 통하여 속도 추정 특성과 회전자 저항 특성을 살펴본 결과, 본 논문의 EFK 알고리즘이 속도 센서없는 유도전동기 벡터제어에 적합함을 입증할 수 있었다.

• 한국산업융합학회 논문집
• /
• 제21권3호
• /
• pp.117-123
• /
• 2018

Continued R&D investment by the government and corporate support played a major role as the background of the rapid growth of the Republic of Korea. In 2017 of the Republic of Korea, the R&D support size of the government accounted for 19.7 trillion won, accounting for 4.7% of the government budget. Government R&D budgets are increasing by 2.5% each year. In this paper, we analyzed the outcomes of the Busan regional company support project conducted in the 2 years. For the time series analysis, we gathered company support amount by year, sales after company support, employment. We used IBM SPSS(Statistical Package for the Social Sciences) statistics 18 for correlation analysis.