• 제어로봇시스템학회논문지
• /
• 제4권2호
• /
• pp.203-208
• /
• 1998

This paper presents an adaptive fuzzy controller using an neural network and adaptation algorithm. Reference-model following neuro-fuzzy controller(RMFNFC) is invesgated in order to overcome the difficulty of rule selecting and defects of the membership function in the general fuzzy logic controller(FLC). RMFNFC is developed to tune various parameter of the fuzzy controller which is used for the discrete nonlinear system control. RMFNFC is trained with the identification information and control closed loop error. A closed loop error is used for design criteria of a fuzzy controller which characterizes and quantize the control performance required in the overall control system. A control system is trained up the controller with the variation of the system obtained from the identifier and closed loop error. Numerical examples are presented to control of the discrete nonlinear system. Simulation results show the effectiveness of the proposed controller.

• 전자공학회논문지B
• /
• 제29B권8호
• /
• pp.15-23
• /
• 1992

Neural network control has many innovative potentials for fast, accurate and intelligent adaptive control. In this paper, two kinds of neurocontrol architectures for the dynamic control of robot manipulators are developed. One is based on a System Identification and Control scheme and the other is based on the Feedback-Error leaming scheme. Both of the proposed architectures use an inverse dynamic neurocontroller in parallel with a linear neurocontroller. The difference is that the first architecture uses the system identifier to get the signals used for training neurocontrollers, while the second architecture uses a properly defined energy function. Compared with the previous types of neurocontrollers which are using an inverse dynamic neurocontroller and a fixed PD gain controller, the proposed architectures not only eliminate the painful process of the fixed gain tuning but also exhibit superior peformances because the linear neurocontroller can adapt its gains according to the applied task. This superior performance is tested and verified through computer simulation of the dynamic control of the PUMA 560 arm.

• 한국멀티미디어학회논문지
• /
• 제10권12호
• /
• pp.1585-1590
• /
• 2007

This paper introduces adaptive PRML (Partial Response Maximum Likelihood) architecture with PR (a,b,c,d,e) channel type for the improved readability of high-density optical discs with capacity greater than 30GB. The proposed PRML architecture consists of an adaptive equalizer, a Viterbi detector and a channel identifier. Detailed description for each component is included. The architecture is implemented in chip and also confirmed its performance on the test board mounting the chip. Test results show that the proposed 5-tap PRML architecture is well operated, and less than $2{\times}10^{-4}$ of BER (Bit Error Rate) is achieved with radial and tangential tilt margin of ${\pm}0.6^{\circ}$ on self-made 30GB BD at 1x speed.

• 한국항해항만학회:학술대회논문집
• /
• 한국항해항만학회 1998년도 추계학술대회논문집:21세기에 대비한 지능형 통합항만관리
• /
• pp.159-165
• /
• 1998

In this paper the sway-control problem of a container crane is investigated. The control loop is divided into two stages. The first stage is a modified time optimal control for trolley traversing. The velocity command for trolley traversing consists of three components ; a reference velocity and two feedback signals for compensating the deviations of trolley and sway angle from their desired trajectories. For trolley's exact positioning the trolley dynamics is identified via an error equation identifier structure. The second stage is a nonlinear residual sway control that starts at the end of first stage. The control design for the second stage is investigated from the perspective of controling an underactuated system, and the control law combines the feedback linearization and variable structure control.

• 한국기계가공학회지
• /
• 제2권4호
• /
• pp.46-52
• /
• 2003

For position control of electro-hydraulic servo system, single rod cylinders and double rod cylinders are used. Single rod cylinders have simple structure for manufacturing but different volume ratio of two sides induce to non-linearity in process of then mathematical modeling. So only with conventional PID control method it is difficult to control single rod cylinders precisely. For mole precise position control of single rod cylinders, a controller which is inserted a velocity feedback PID controller and MRAC controller are proposed. With experiment control performances of three control methods are compared. In case of experiment, for external varying load to the system, a hydraulic cylinder and a pressure control valve are used. In conclusion a MRAC is considered a suitable control method for external varying load.

• 대한전기학회논문지
• /
• 제45권4호
• /
• pp.574-581
• /
• 1996

The neural network approach has been shown to be a general scheme for nonlinear dynamical system identification. Unfortunately the error surface of a Multilayer Neural Networks(MNN) that widely used is often highly complex. This is a disadvantage and potential traps may exist in the identification procedure. The objective of this paper is to identify a nonlinear dynamical systems based on Self-Organized Distributed Networks (SODN). The learning with the SODN is fast and precise. Such properties are caused from the local learning mechanism. Each local network learns only data in a subregion. This paper also discusses neural network as identifier of nonlinear dynamical systems. The structure of nonlinear system identification employs series-parallel model. The identification procedure is based on a discrete-time formulation. Through extensive simulation, SODN is shown to be effective for identification of nonlinear dynamical systems. (author). 13 refs., 7 figs., 2 tabs.

• 대한의용생체공학회:의공학회지
• /
• 제16권4호
• /
• pp.507-516
• /
• 1995

근전도 신호(electromyogram)의 시변 비정상(time varying nonstationary) 특성은 신호의 정확한 모델링 및 인식에 제약 조건으로 받아들여 졌다. 특히, 최근 들어 장애자들을 위한 보철제어분야에서 근전도 신호를 이용한 기능적 전기 자극을 위한 FES(funcitonal electrical stimulation) 시스템에 있어 근전도 신호의 파라메터 인식은 중요한 요소로서 작용한다. 그러나, 근전도 신호는 자세의 변화 및 근육 피로도 등의 요인에 의해서 시변 비정상 특성을 띠고 있기 때문에 시간에 따라 변하는 인식 파라메터를 정확하게 인식할 수 있는 새로운 알고리즘의 개발과 실시간 처리가 가능한 컴퓨터 하드웨어의 설계가 요구된다. 따라서, 본 논문에서는 시평면의 근전도 신호를 이산 여현 변환(discrete cosine transform)을 이용하여 변환 평면으로 옮긴 다음 상태 방정식(state space equation)을 써서 변환 평면상에서의 AR(autoregressive) 모델을 세우고 주어진 근전도 신호에 대해 모델 파라메터를 추정하였으며, 제안한 알고리즘은 실시간 처리를 위하여 2개의 독립적인 중앙 연산 처리 장치를 갖춘 INMOS사의 IMS T-805 병렬 처리 컴퓨터를 이용하여 동시 다발적인 연산을 수행함으로서 알고리즘의 연산 효율을 높였다. 제안된 알고리즘의 타당성을 검증하기 위해 모델의 추정 오차에 영향을 미치는 입력 자기상관 행렬(input correlation matrix)의 condition number의 변화 및 평균자승오차(mean square error)를 구하여 기존의 SLS(sequential least square) 알고리즘과 비교하였다.

• 제어로봇시스템학회논문지
• /
• 제7권10호
• /
• pp.827-836
• /
• 2001

The PID controller has been widely applied to the most control systems because of its simple structure and east designing. One of the important points to design the PID control system is to tune the approximate control parameters for the given target system. To find the PID parameters using Ziegler Nichols(ZN) method needs a lot of experience and experiments to ensure the optimal performance. In this paper, CMIA(Cell Mediated Immune Algorithm) controller is proposed to drive the autonomous guided vehicle (AGV) more effectively. The proposed controller is based on specific immune responses of the biological immune system which is the cell mediated immunity. To verify the performance of the proposed CMIA controller, some experiments for the control of steering and speed of that AGV are performed. The tracking error of the AGV is mainly investigated for this purpose. As a result, the capability of realization and reliableness are proved by comparing the response characteristics of the proposed CMIA controllers with those of the conventional PID and NNPID(Neural Network PID) controller.

• 한국통신학회논문지
• /
• 제26권12C호
• /
• pp.181-189
• /
• 2001

본 논문에서는 ATM 망의 트래픽 제어 기능 중의 하나인 감시 알고리즘을 제안하고 칩(chip)을 설계하였으며, VHDL을 사용하여 구현하였다. 제안한 알고리즘은 우선순위가 높은 셀의 손실을 최소로 하고, 트래픽의 다중화 및 역다중화 과정에서 발생되는 트래픽의 군집성을 해소할 수 있다는 장점을 갖고 있다. 구현한 칩은 입력 모듈과 UPC 모듈, 출력 모듈의 3부분으로 이루어지며, 편의상, UPC 모듈에서는 메모리 제어시 메모리 어드레스 즉, 커넥션 테이블의 index를 VCI(Virtual Channel Identifier)와 동일하게 할당하였다. 또한 UPC 모듈의 VSA(Virtual Scheduling Algorithm)블럭에서 셀의 도착시간을 계수하는 카운터가 랩-어라운드(wrap-around)할 때 생기는 VSA의 오류를 보정해 줄 수 있는 방법을 제안하였다.

• 대한전자공학회논문지TC
• /
• 제48권11호
• /
• pp.60-66
• /
• 2011

LTE는 다양한 서비스 클래스를 가지고 QoS를 지원한다. 각각의 트래픽 클래스에 대해 BER 요구사항이 크게 다르다. 이는 다양한 클래스의 트래픽에 대해 전송 전력이 상이함을 의미한다. 본 논문에서는 우선순위와 목표 BER을 고려하여 서비스 클래스를 지원하는 새로운 LTE를 위한 스케줄러를 제안한다. CQI와 QCI에 의해 각 사용자당 크래픽 클래스에 대한 목표 BER로 부터 최소 전송 전력이 할당된다. 따라서 사용자의 과거 사용 전송율과 트래픽의 우선순위와 같은 다른 정보를 이용하여 점유 채널의 확률이 결정된다. 서비스 클래스 스케줄링의 시뮬레이션 결과는 최대전송율 및 비례균일과 비교될 것이다. 실험결과는 서비스 클래스 기반의 스케줄링이 전체 시스템 수율을 개선할 수 있음을 보여주고 있다.