• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.27 no.6
• /
• pp.59-67
• /
• 2013

This paper presents a neural net type process model for enhancing learning compensation function in hot strip finishing rolling mill. Adequate input and output variables of process model are chosen, the proposed model was designed as single layer neural net. Equivalent carbon content, strip thickness and rolling speed are suggested as input variables, and looper's manipulation variable is proposed as output variable. According to simulation result using process data to show the validity of the proposed process model, neural net type process model's outputs give almost similar data to process output under same input conditions.

• 제어로봇시스템학회:학술대회논문집
• /
• 1993.10b
• /
• pp.534-540
• /
• 1993

In this paper, we present an iterative learning method of compensating for position sensor error. The previously known compensation algrithms need a special perfect position sensor or a priori information about error sources, while ours does not. To our best knowledge, any iterative learning approach has not been taken for sensor error compensation. Furthermore, our iterative learning algorithm does not have the drawbacks of the existing iterative learning control theories. To be more specific, our algorithm learns a uncertain function inself rather than its special time-trajectory and does not request the derivatives of measurement signals. Moreover, it does not require the learning system to start with the same initial condition for all iterations. To illuminate the generality and practical use of our algorithm, we give the rigorous proof for its convergence and some experimental results.

• Journal of the Korea Society of Computer and Information
• /
• v.23 no.10
• /
• pp.23-31
• /
• 2018

Recently, there have been many studies on machine learning. Among them, studies on reinforcement learning are actively worked. In this study, we propose a controller to control bicycle using DDPG (Deep Deterministic Policy Gradient) algorithm which is the latest deep reinforcement learning method. In this paper, we redefine the compensation function of bicycle dynamics and neural network to learn agents. When using the proposed method for data learning and control, it is possible to perform the function of not allowing the bicycle to fall over and reach the further given destination unlike the existing method. For the performance evaluation, we have experimented that the proposed algorithm works in various environments such as fixed speed, random, target point, and not determined. Finally, as a result, it is confirmed that the proposed algorithm shows better performance than the conventional neural network algorithms NAF and PPO.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.9
• /
• pp.57-66
• /
• 1994

In this paper, we present an iterative learning method of compensating for position sensor error. The previously known compensation algorithms need a special perfect position sensor or a priori information about error sources, while ours does not. to our best knowledge, any iterative learning approach has not been taken for sensor error compensation. Furthermore, our iterativelearning algorithm does not have the drawbacks of the existing interativelearning control theories. To be more specivic, our algorithm learns an uncertain function itself rather than its special time-trajectory and does not reuquest the derivatives of measurement signals. Moreover, it does not require the learning system to start with the same initial condition for all iterations. To illuminate the generality and practical use of our algorithm, we give the rigorous proof for its convergence and some experimental results.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.21 no.9
• /
• pp.2456-2469
• /
• 1996

A digital satellite communication channel has a nonlinearity with memory due to saturation characeristis of the high poer amplifier in the satellite and transmitter/receiver linear filter used in the overall system. In this paper, we propose a complex radial basis function network(CRBFN) based adaptive equalizer for compensation of nonlinearities in digital satellite communication channels. The proposed CRBFN untilizes a complex-valued hybrid learning algorithm of k-means clustering and LMS(least mean sequare) algorithm that is an extension of Moody Darken's algorithm for real-valued data. We evaluate performance of CRBFN in terms of symbol error rates and mean squared errors nder various noise conditions for 4-PSK(phase shift keying) digital modulation schemes and compare with those of comples pth order inverse adaptive Volterra filter. The computer simulation results show that the proposed CRBFN ehibits good equalization, low computational complexity and fast learning capabilities.

• Proceedings of the KIEE Conference
• /
• 1996.11a
• /
• pp.386-388
• /
• 1996

This paper presents the learning controller for robot manipulators to track the desired trajectory exactly. The learning controller, based on the Lyapunov theory, consists of a fixed PD action and a repetitive action for the purpose of feedforward compensation which is adjusted utilizing a linear combination of the velocity and position errors. The learning controller Is often used In case of the desired trajectories are periodic tasks, and has advantage that it periodically converges to zero even if we don't know the exact dynamic parameters. In this paper, we show that the position and velocity errors of robot manipulators converge to zero sa time goes infinite for the input is periodic function and show a good trajectory tracking performance In the cartesian space.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.14 no.3
• /
• pp.63-68
• /
• 2014

This paper related with the I-CMA (Improved-CMA) algorithm that is possible to compensates of phase in CMA adatpve equalizer which is used for the elemination of intersymbol interference in the multipath fading and band limit characteristics of channel. The new cost function is proposed for the eliminate the amplitude and phase simulataneous by modifying the cost fuction for get the error signal in present CMA algorithm. It has a merit to the algorithm simplicities and eliminats the PLL device for phase compensation after equalization. For proving this, the recovered signal constellation that is the output of equalizer output signal and the residual isi and Maximum Distortion charateristic learning curve that are presents the convergence performance in the equalizer and the overall frequency transfer function of channel and equalizer were used. As a result of computer simulation, the I-CMA has more good compensation capability of amplitude and phas in the recovered constellation. But the convergence time is slow due to the simultaneously phase compensation.

• Journal of Power Electronics
• /
• v.11 no.6
• /
• pp.922-930
• /
• 2011

In this paper, two radial basis function neural networks (RBFNNs) are used to dynamically identify harmonics content in converter waveforms based on the p-q (real power-imaginary power) theory. The converter waveforms are analyzed and the types of harmonic content are identified over a wide operating range. Constant power and sinusoidal current compensation strategies are investigated in this paper. The RBFNN filtering training algorithm is based on a systematic and computationally efficient training method called the hybrid learning method. In this new methodology, the RBFNN is combined with the p-q theory to extract the harmonics content in converter waveforms. The small size and the robustness of the resulting network models reflect the effectiveness of the algorithm. The analysis is verified using MATLAB simulations.

• Journal of Biomedical Engineering Research
• /
• v.11 no.1
• /
• pp.157-162
• /
• 1990

In this study, a peripheral auditory model is used as a frequency feature extractor and a recurrent network which has recurrent links on input nodes is constructed in order to show the reliability of the recurrent network as a recognizer by executing recognition tests for 4 Korean place names and syllables. In the case of using the general learning rule, it is found that the weights are diverged for a long sequence because of the characteristics of the node function in the hidden and output layers. So, a refined weight compensation method is proposed and, using this method, it is possible to improve the system operation and to use long data. The recognition results are considerably good, even if time worping and endpoint detection are omitted and learning patterns and test patterns are made of average length of data. The recurrent network used in this study reflects well time information of temporal speech signal.

• Journal of the Korean Society for Precision Engineering
• /
• v.15 no.11
• /
• pp.76-82
• /
• 1998

In this study, the Ball screw and Servo motor Mechanism is considered as a High Speed Tool Feed System for the machining of a piston of a reciprocating engine. For the machining of a piston, that shapes oval, high speed servo mechanism is needed as a positioning of a cutting tool, and the stroke of tool is 0.1 mm ~ 1 mm. Ball screw and servo motor Mechanism is available very much because this mechanism is used widely in general machine. This Mechanism has been designed with the use of the decrease in mass and partial wear of the ball screw for high speed positioning of tool. Also the periodic learning control method with the inverse transfer function compensation has been applied to the positioning control for the high accuracy positioning of tool. These applications lead the achievement of the machining of a piston with an accuracy of 5${\mu}{\textrm}{m}$ at 2500 rpm in CNC turning.