• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1996년도 한국자동제어학술회의논문집(국내학술편); 포항공과대학교, 포항; 24-26 Oct. 1996
• /
• pp.944-947
• /
• 1996

Realization of autonomous agents that organize their own internal structure in order to behave adequately with respect to their goals and the world is the ultimate goal of AI and Robotics. Reinforcement learning gas recently been receiving increased attention as a method for robot learning with little or no a priori knowledge and higher capability of reactive and adaptive behaviors. In this paper, we present a method of reinforcement learning by which a multi robots learn to move to goal. The results of computer simulations are given.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제49권11호
• /
• pp.591-598
• /
• 2000

In this thesis, a digital tracking controller is proposed for multi-axis position control system. Tracking and contouring error exist when the machine tool moves along a trajectory in multi-axis system. The proposed scheme enhances the tracking and contouring performance by reducing the errors. Also, an optimal tracking controller reduces the tracking error by the state feedback and the feedforward compensator reduces the effects of a nonlinear disturbance such as friction or dead zone. The proposed control scheme reduces the contour error which occurred when the tool tracks the reference trajectory. Finally, the performance of the proposed controller is exemplified by some simulations and by applying the real XY servo system.

• 한국정밀공학회지
• /
• 제30권12호
• /
• pp.1313-1320
• /
• 2013

An ultraprecision multi-axis machine tool has been designed and developed in our laboratory. The machine tool has four moving axes which are composed of three linear axes and one rotational axis. It has a gantry type structure and the Z-axis is on the X-axis and the C-axis, on which a workpiece is located, is inside the Y-axis. This paper shows control performance improving method and procedure for the ultra-precision positioning control of a hydrostatic bearing guided linear axis. Through improvements of electrical and mechanical components for the control system such as control electronics and oil pumping systems, the control disturbing noise is decreased. Also by the frequency domain analysis of control system those problem-making system components are identified and modified with analytical methods. The controller is analyzed and designed from frequency domain data and system information. In the experimental control results the nanometer order control result is successfully presented.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.438-443
• /
• 2005

Handy paint rollers with simple or no patterns are generally used to transcribe its design to a wall just after painting. However, the types of the patterns are limited to several conventional ones, so that interior planners' or decorators' demands are gradually tending to getting attractive roller designs. In order to obtain abundant kinds of the roller designs, a new advanced 3D machining method should be established for cylindrical models. In this paper, a post-processor that can generate suitable NC data is proposed for multi-axis NC machine tools with a rotary unit. The 3D machining system with the post-processor is also presented for an attractive interior decorating. The machining system allows us to easily transcribe the relief designs from on a flat model to on a cylindrical model. The effectiveness of the proposed 3D machining system using the post-processor is demonstrated through some machining experiments.

• 소성∙가공
• /
• 제17권4호
• /
• pp.257-262
• /
• 2008

The control of wire temperature is very important in the fine wire drawing process. The wire speed should be increased, and the wire temperature should be dropped as much as possible. Up to now, the process design of wire drawing process depends on the experiences of experts. In this study, a wire drawing process design method was proposed to increase the productivity. The proposed method of this study includes the pass schedule and the design of a multi pass wire drawing machine. A pass schedule was performed based on the calculation of the wire temperature. Also, a new multi pass wire drawing machine was manufactured to apply the designed pass schedule. Through the wire drawing experiment, the effectiveness of the proposed process design method was evaluated. The final drawing speed was increased from 1,100m/min to 2,000m/min without deterioration of final drawn wire.

• 대한설비관리학회지
• /
• 제23권4호
• /
• pp.29-38
• /
• 2018

Through process quality information, the time required for process quality analysis has been drastically shortened, the process defect rate has been reduced, and the manufacturing lead time has been shortened and the on-time delivery rate has been improved. Therefore, The purpose of this study is to develop a quality information analysis system model that effectively shortens the time required for process quality analysis in automobile safety belt parts manufacturing process. As a result of experiments on communication operation between manufacturing execution system (MES) quality server, injection machine control computer, injection machine programmable logic controller (PLC) and terminal, in analyzing quality information, the conventional handwriting input method took an average of 20 minutes, but the new multi-network method took about 2 minutes on average. In addition, the process defect rate was reduced by 13% and the manufacturing lead time was shortened from 28 hours to 20 hours. The delivery compliance rate improved from 96 to 99%.

• 한국정밀공학회지
• /
• 제26권4호
• /
• pp.33-38
• /
• 2009

• 한국멀티미디어학회논문지
• /
• 제6권5호
• /
• pp.937-944
• /
• 2003

다중 에이전트 시스템에서 가장 중요한 문제는 여러 에이전트가 서로 효율적인 협동(coordination)을 통해서 목표(goal)를 성취하는 것과 다른 에이전트들과의 충돌(collision) 을 방지하는 것이다. 본 논문에서는 먹이 추적 문제의 목표를 효율적으로 성취하기 위해 새로운 전략 방법을 제안한다. 제안된 제어 전략은 다중 에이전트를 제어하기 위해 강화 학습을 이용하였고, 에이전트들 간의 거리관계와 공간 관계를 고려하였다.

• 한국정보통신학회논문지
• /
• 제9권5호
• /
• pp.1039-1044
• /
• 2005

원격 제어 모니터링 시스템을 위한 임베디드 리눅스 시스템의 다중 채널 직렬 장치를 설계하였다. 제안된 장치는 다중 채널 직렬 장치의 효과적인 데이터 처리를 위해 내부에 FIFO, Sate Machine, 인터럽트로 구성되어 있으며 임베디드 리눅스에서 효과적인 디바이스 프로그램을 가능하게 한다. 다중 채널 직렬 장치의 기존의 접근 방식은 개별 직렬 장치를 일일이 접근하고 있으나 본 장치에서는 다중 채널 직렬 장치를 한 개의 직렬 장치로 간주하여 접근하는 것과 같은 효과를 볼 수 있다. 본 장치는 다중 채널 직렬 입출력 구동에 효율적이며 기존의 다중 채널 직렬 장치 보다 접근 속도가 빠르다. 본 장치는 수질 측정과 같은 센서들 사이의 거리가 멀거나 센서와 모니터링 시스템의 사이의 거리가 먼 원격 제어 모니터링에 효과적이다.

• 대한임베디드공학회논문지
• /
• 제10권5호
• /
• pp.325-333
• /
• 2015

Vision based night-time vehicle detection has been an emerging research field in various advanced driver assistance systems(ADAS) and automotive vehicle as well as automatic head-lamp control. In this paper, we propose night-time vehicle detection method based on multi-class support vector machine(SVM) that consists of thresholding, labeling, feature extraction, and multi-class SVM. Vehicle light candidate blobs are extracted by local mean based thresholding following by labeling process. Seven geometric and stochastic features are extracted from each candidate through the feature extraction step. Each candidate blob is classified into vehicle light or not by multi-class SVM. Four different multi-class SVM including one-against-all(OAA), one-against-one(OAO), top-down tree structured and bottom-up tree structured SVM classifiers are implemented and evaluated in terms of vehicle detection performances. Through the simulations tested on road video sequences, we prove that top-down tree structured and bottom-up tree structured SVM have relatively better performances than the others.