• Journal of Semiconductor Engineering
• /
• 제2권2호
• /
• pp.130-135
• /
• 2021

Recently, deep neural networks (DNNs) are actively used for action control so that an autonomous system, such as the robot, can perform human-like behaviors and operations. Unlike recognition tasks, the real-time operation is essential in action control, and it is too slow to use remote learning on a server communicating through a network. New learning techniques, such as reinforcement learning (RL), are needed to determine and select the correct robot behavior locally. In this paper, we propose an energy-efficient DNN processor with a LUT-based processing engine and near-zero skipper. A CNN-based facial emotion recognition and an RNN-based emotional dialogue generation model is integrated for natural HRI system and tested with the proposed processor. It supports 1b to 16b variable weight bit precision with and 57.6% and 28.5% lower energy consumption than conventional MAC arithmetic units for 1b and 16b weight precision. Also, the near-zero skipper reduces 36% of MAC operation and consumes 28% lower energy consumption for facial emotion recognition tasks. Implemented in 65nm CMOS process, the proposed processor occupies 1784×1784 um2 areas and dissipates 0.28 mW and 34.4 mW at 1fps and 30fps facial emotion recognition tasks.

• ETRI Journal
• /
• 제27권5호
• /
• pp.563-568
• /
• 2005

A motion-control chip contains major functions that are necessary to control the position of each motor, such as generating velocity command profiles, reading motor positions, producing control signals, driving several types of servo amplifiers, and interfacing host processors. Existing motion-control chips can only generate velocity profiles of fixed characteristics, typically linear and s-shape smooth symmetric curves. But velocity profiles of these two characteristics are not optimal for all tasks in industrial robots and automation systems. Velocity profiles of other characteristics are preferred for some tasks. This paper proposes a motion-control chip to generate velocity profiles of desired acceleration and deceleration characteristics. The proposed motion-control chip is implemented with a field-programmable gate array by using the Very High-Speed Integrated Circuit Hardware Description Language and Handel-C. Experiments using velocity profiles of four different characteristics will be performed.

• 한국공작기계학회:학술대회논문집
• /
• 한국공작기계학회 2002년도 춘계학술대회 논문집
• /
• pp.381-386
• /
• 2002

In this paper, we have proposed a new approach to the design of robot vision system to develop the technology for the automatic test and assembling of precision mechanical and electronic parts for the factory automation. In order to perform real time implementation of the automatic assembling tasks in the complex processes, we have developed an intelligent control algorithm based-on neural networks control theory to enhance the precise motion control. Implementing of the automatic test tasks has been performed by the real-time vision algorithm based-on TMS320C31 DSPs. It distinguishes correctly the difference between the acceptable and unacceptable defective item through pattern recognition of parts by the developed vision algorithm. Finally, the performance of proposed robot vision system has been illustrated by experiment for the similar model of fifth cell among the twelve cell for automatic test and assembling in S company.

• 한국공작기계학회:학술대회논문집
• /
• 한국공작기계학회 2001년도 춘계학술대회 논문집(한국공작기계학회)
• /
• pp.217-222
• /
• 2001

In this paper, we have proposed a new approach to the design of robot vision system to develop the technology for the automatic test and assembling of precision mechanical and electronic parts for the factory automation. In order to perform real time implementation of the automatic assembling tasks in the complex processes, we have developed an intelligent control algorithm based-on neural networks control theory to enhance the precise motion control. Implementing of the automatic test tasks has been performed by the real-time vision algorithm based-on TMS320C31 DSPs. It distinguishes correctly the difference between the acceptable and unacceptable defective item through pattern recognition of parts by the developed vision algorithm. Finally, the performance of proposed robot vision system has been illustrated by experiment for the similar model of fifth cell among the twelve cell for automatic test and assembling in S company.

• 디지털산업정보학회논문지
• /
• 제8권1호
• /
• pp.1-7
• /
• 2012

In this paper Vehicle driving simulator have been used in the development and modification of models. A real-time simulation system and washout algorithm for an excavator have been developed for a driving simulator with six degrees of freedom. An interesting question, "how the 6 DOF Driving Simulator can be controlled optimally for the various tasks?" is not easy to be answered. This paper presents the hardware and software developed for a driving simulator of construction vehicle. A simulator can reduce cost and time a variety of driving simulations in the laboratory. Using its 6 DOF Simulator can move in various modes, and perform dexterous tasks. Driving simulators have begun to proliferate in the automotive industry and the associated research community. This effort involves the real-time dynamic of wheel-type excavator the design and manufacturing of the Stewart platform an integrated control system of the platform and three-dimensional graphic modeling of the driving environments.

• 대한지구과학교육학회지
• /
• 제3권3호
• /
• pp.209-218
• /
• 2010

In this study, we investigate the effects of instruction using science magic program on the scientific attitude and scientific inquiry ability in elementary students. For this study, it was chosen two classes of the forth grades J elementary school in Seoul. Instruction using science magic program was applied to the experimental group for 8 weeks during the school hours. The results of this study were as follows : 1. Science tasks applied science magic had influence on elementary learners' scientific attitude in positive way. 2. Science tasks applied science magic had valuable significance to observation, classification, data intepretation ability. However it had no valuable significance to scientific integrated inquiry.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.60.1-60
• /
• 2001

In recent years, distributed control systems(DCS) using fieldbus such as CAN have been being applied to process systems but it is very difficult to design the DCS in order to guarantee the given end-to-end constraints such as precedence constraints, time constraints, and periods and priorities of tasks and messages. This paper presents a scheduling method to guarantee the given end-to-end constraints. The presented scheduling method is the integrated one considering both tasks executed in each node and messages transmitted via the network and is designed to be applied to a general DCS that has multiple loops with several types of constraints, where each loop consists of ...

• 교육녹색환경연구
• /
• 제8권1호
• /
• pp.1-10
• /
• 2009

This study is about the present condition of facilities for the disabled at elementary schools in Ulsan and improvement plan of facilities for the disabled. The important thing of tasks of societies and nations is rights to education for the disabled. If an education is natural rights that give people identification, Societies and nations do their best to adapting anyone no matter who has any circumstance to civilization. The more people are poor, the more Societies and nations support and give them chance to education. Among people who really need the chance to education, there was the disabled. Therefore, societies and nations support them carefully more than any others. Above all things, considering the fact that separation to integration of a education for the disabled, facilities for the disabled that can base on an integrated education must be improved as soon as possible. An integrated education contribute to growing sociality and security of human rights for the disabled. It is natural that basic facilities are necessary for the disabled. A systematic program development that make ordinary students look fairly disabled students is really demanded for that ordinary students give disabled students love and help.

• 대한수학교육학회지:수학교육학연구
• /
• 제27권1호
• /
• pp.137-156
• /
• 2017

본 연구는 테크놀로지 활용 지원체계의 확대, 교사의 지식과 신념, 교사교육 프로그램의 분석 틀의 문헌연구를 기초로 국내 수학교육과의 관련 교육의 현황과 현안, 발전 방향을 탐색하였다. 강좌개설 및 운영 자료는 설문조사와 강의계획서로 수집한 결과 강의는 수업의 향상을 목적으로 하며 전체 학과의 1/3에 수학교육관련 ICT 강의가 개설되지 않았다. 테크놀로지 통합 예비 중등수학교사교육을 위하여 국가적으로 교육과정의 방향과 세부내용의 부조화, 소프트웨어와 교과서 등 지원 부족 문제 해결을 통한 교육현장 개선과 교사교육의 포괄적, 세부적 방향설정이 중요하며 무엇보다도 테크놀로지 활용 관련 비전의 공유가 필요하다.

• ETRI Journal
• /
• 제45권1호
• /
• pp.18-27
• /
• 2023

We present an English-Korean speech translation corpus, named EnKoST-C. End-to-end model training for speech translation tasks often suffers from a lack of parallel data, such as speech data in the source language and equivalent text data in the target language. Most available public speech translation corpora were developed for European languages, and there is currently no public corpus for English-Korean end-to-end speech translation. Thus, we created an EnKoST-C centered on TED Talks. In this process, we enhance the sentence alignment approach using the subtitle time information and bilingual sentence embedding information. As a result, we built a 559-h English-Korean speech translation corpus. The proposed sentence alignment approach showed excellent performance of 0.96 f-measure score. We also show the baseline performance of an English-Korean speech translation model trained with EnKoST-C. The EnKoST-C is freely available on a Korean government open data hub site.