• Journal of Digital Contents Society
• /
• v.10 no.2
• /
• pp.249-258
• /
• 2009

As wireless internet markets become larger than before, many mobile applications are also being developed actively. In this circumstances mobile devices such as cell phones, PDAs are playing an important role to satisfy the user's need of ubiquitous computing. Due to the hardware limitations, however, the mobile devices like PDA can not run large-scale softwares by itself. The main goal of this paper is to make large-scale applications runnable on PDA. To accomplish this, we used the PDA-JPVM cluster computing engine which has been already developed by us. We found out by running the applications and the performance evaluation that large-scale Java softwares can easily run on the hardware-limited PDA. And the performance evaluation results are also presented.

• Journal of Electrical Engineering and Technology
• /
• v.13 no.3
• /
• pp.1185-1201
• /
• 2018

This paper presents a sensorless direct torque control (DTC) combined with sliding mode approach (SM) and space vector modulation (SVM) to achieve mainly a high performance and reduce torque and flux ripples of a parallel-connected two five-phase permanent magnet synchronous machine (PMSM) drive system. In order to increase the proposed drive robustness and decrease its complexity and cost, the rotor speeds, rotor positions, fluxes as well as torques are estimated by using a sliding mode observer (SMO) scheme. The effectiveness of the proposed sliding mode observer in conjunction with the sliding mode control based DTC is confirmed through the application of different load torques for wide speed range operation. Comparison between sliding mode control and proportional integral (PI) control based DTC of the proposed two-motor drive is provided. The obtained speeds, torques and fluxes responses follow their references; even in low and reverse speed operations, load torques changes, and machines parameters variations. Simulation results confirm also that, the ripples of the torques and fluxes are reduced more than 3.33% and 16.66 %, respectively, and the speed overshoots and speed drops are reduced about 99.85% and 92.24%, respectively.

• Journal of Institute of Control, Robotics and Systems
• /
• v.10 no.11
• /
• pp.1044-1051
• /
• 2004

This paper described a novel locomotion interface that can generate infinite floor for various surface, named as virtual walking machine. This interface allows users to participate in a life-like walking experience in virtual environments, which include various terrains such as plains, slopes and stair ground surfaces. The interface is composed of two three-DOF (X, Y, Yaw) planar devices and two four-DOF (Pitch, Roll, Z, and relative rotation) footpads. The planar devices are driven by AC servomotors for generating fast motions, while the footpad devices are driven by pneumatic actuators for continuous support of human weight. To simulate natural human walking, the locomotion interface design specification are acquired based on gait analysis and each mechanism is optimally designed and manufactured to satisfy the given requirements. The designed locomotion interface allows natural walking(step: 0.8m, height: 20cm, load capability: 100kg, slope:30deg) for various terrains.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.7 no.4
• /
• pp.50-55
• /
• 2008

CNC wire bending machines are used in industries to make a type variety of wire products such as long links. The machines have a long arm device to rotate in order to remove forming errors by flexibility of wire. Generally, the machines which constructed servo motors in the arm have the rotating range of the arm under 360 degree because the servo motors connect with fixed control devices on frame by many cables. The rotating angle under 360 degree limits working speed and forming geometry. Therefore this study developed a gear train to drive a parts in arm and to be independent on arm rotation movement. The developed gear train can transfer four movements to four components in arm and is consists parallel of four pairs of satellite gear trains. This study constructed the CNC wire bending machine with the developed gear train and verified that the gear train could drive internal components independently on arm rotation.

• Proceedings of the Korean Society of Sericultural Science Conference
• /
• 1997.06a
• /
• pp.103-117
• /
• 1997

Decrease in the cocoon production of Japan is drastic because of low price of cocoon, scarcity of successors and so on. To tide over the difficulty, the automation system in the sericulture was discussed and some trials have been conducted by the Ministry of Agriculture, Forestry and Fisheries of Japan. The attempts are based on a low cost artificial diet which does not rely on mulberry leaves. Automatic machines developed until now are a rearing machine constituted with repeated belt conveyor, an reformation type of former rearing machine "Bombyx" and a mounting machine. Running parallel with them, utilization of 20-hydroxyecdysone extracted from a plant to the mounting was also studied to use their machines efficiently in the fields. In conclusion, 10 tons of law cocoon will come to be produced by two persons labor. At present, an automatic rearing system on low cost artificial diet has been developing for the future sericulture.

• Journal of the Korean Society of Industry Convergence
• /
• v.22 no.3
• /
• pp.293-301
• /
• 2019

Based on accurate image processing technology, a system for measuring displacement in ${\mu}m$ for drive error (position error, straightness error, flatness error) at a distance using parallel light and image sensor is developed, and a system for applying this technology development to a large 3D rapid prototyping machine and compensating in real time is developed to dramatically reduce the range of measurement error and enable intelligent 3D production of high quality products.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.22 no.2
• /
• pp.192-198
• /
• 2013

A light-weight enough to be portable yet functionally complete hand-writing machine was designed with hardware assembled and software implemented. The machine simulating human hand-writing not only allows identical massive scriptings on papers in human's proxy, but also offers emotional appearance as if the script was directly hand-written in ink. The features of human writing were independently put together to complete the action of scripting. The writing style could be altered in order to mimic a person's own specific writing style. The present work demonstrates that the assembled hand-writing machine was able to duplicate writing of scripts in almost parallel lines in consistent manner. The machine can be used in the emotional marketing area where seasonal greetings or thank-you cards are to be sent in massive manner. The pen-holding machine can also find an emotional application such as writing a short greeting note on a plastic lid on a disposable coffee cup in coffee-selling stores.

• IE interfaces
• /
• v.11 no.2
• /
• pp.65-78
• /
• 1998

The letters of the automobile tire are usually engraved on the side-wall. The shape of the side-wall is a sculptured surface generated by the rotational sweeping of a profile curve. The letters laid on the side-wall are usually designed by a 2-dimensional CAD. It is impossible to machine the letters on the surface accurately by 3-axis NC machining, because the axis of cutter should be tilted to align with the normal vector of the surface. In this case. the degree of freedom for the machine is at least four. This paper describes an idea for tool path generation of a 4-axis machine by using the 2-dimensional CAD data of the letters and the surface of the side-wall. This study includes the following procedures; (1) measuring the profile of the side-wall surface and curve-fitting of the measured points. (2) the 'non-parallel projection' of the letters on the side-wall, and (3) an inverse kinematics of the 4-axis lettering machine. Procedures in this paper are programmed in C-language on Windows95 environment. With a PC based CNC controller and a 4-axis lettering machine. these are tested sucessfully for the practical use.

• Journal of the Korean Society of Industry Convergence
• /
• v.27 no.1
• /
• pp.87-95
• /
• 2024

This study is on modulo scheduling algorithms for multicore processor in machine learning applications. Machine learning algorithms are designed to perform a large amount of operations such as vectors and matrices in order to quickly process large amounts of data stream. To support such large amounts of computations, processor architectures to support applications such as artificial intelligence, neural networks, and machine learning are designed in the form of parallel processing such as multicore. To effectively utilize these multi-core hardware resources, various compiler techniques are being used and studied. In this study, among these compiler techniques, we analyzed the modular scheduler, which is especially important in one core's computation pipeline. This paper looked at and compared the iterative modular scheduler and the swing modular scheduler, which are the most widely used and studied. As a result, both schedulers provided similar performance results, and when measuring register pressure as an indicator, it was confirmed that the swing modulo scheduler provided slightly better performance. In this study, a technique that divides recurrence edge is proposed to improve the minimum initiation interval of the modulo schedulers.

• Journal of the Korea Convergence Society
• /
• v.12 no.5
• /
• pp.23-29
• /
• 2021

Recent deep learning-based natural language processing studies are conducting research to improve performance by training large amounts of data from various sources together. However, there is a possibility that the methodology of learning by combining data from various sources into one may prevent performance improvement. In the case of machine translation, data deviation occurs due to differences in translation(liberal, literal), style(colloquial, written, formal, etc.), domains, etc. Combining these corpora into one for learning can adversely affect performance. In this paper, we propose a new Corpus Weight Balance(CWB) method that considers the balance between parallel corpora in machine translation. As a result of the experiment, the model trained with balanced corpus showed better performance than the existing model. In addition, we propose an additional corpus construction process that enables coexistence with the human translation market, which can build high-quality parallel corpus even with a monolingual corpus.