• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.57 no.6
• /
• pp.1042-1047
• /
• 2008

As an ultra-precision measurement system the heterodyne laser interferometer plays an important role in semiconductor industry. However the errors of environment and nonlinearity which are caused by air refraction and frequency-mixing separately reduce the accuracy of displacement measurement. In this paper we propose a DFNN(data fusion and neural network) method for error compensation. As a hybrid method of data fusion and neural network, DFNN method reduces the environmental and nonlinear error simultaneously. The effectiveness of the proposed error compensation method is proved through experimental results.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 1994.04a
• /
• pp.55-67
• /
• 1994

Studies on the ship's automatic navigation & berthing control have been continued by way of solving the ship's mathematical model but the results of such studies have not reached to our satisfactory level due to its non-linear characteristics ar low speed. In this paper the authors propose a new berthing control system which can evaluate as closely as captain's decision-making by using the FNN(Fuzzy Neural Network) controller which can simulate captain's decision-making by using the FNN(Fuzzy neural Network) controller which can simulate captain's knowledge. This berthing controller consists of the navigation subsystem FNN controller and the berthing subsystem FNN controller. The learning data are drawn from Ship Handling Simulator (NavSim NMS90 MK III) and represent the ship motion characteristics internally According to learning procedure both FNN controllers can tune membership functions and identify fuzzy control rules automatically The verified results show the FNN controllers effective to incorporate captain's knowledge and experience of berthing.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.18 no.7
• /
• pp.907-912
• /
• 1993

In this Paper we have proposed a new method to implement the interpolation of the functions, using a neural network. The architecture of neural network is a three-layer perceptron and the training algorithm is a modified error back propagation algorithm adding neurons to hidden layer. The interpolated functions are sin(7 X), 3rd order polynomial 0.5$\times$3_2$\times$2+X+2.5 and rectangular pulse 0.99 U (X-0.2) -0.99 U(X-0.8) +0.01, where U(X) is the unit step. The root mean squred errors of the interpolated functions are 0.00258, 0.00164 and 0.00116 respectively.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.30B no.1
• /
• pp.90-98
• /
• 1993

A Gaussian is formed by diffusing a spot excitation. In this paper, a diffusion neural network model is derived from the diffusion equation. And it is shown that a difference of two Gaussians(DOG) may have the same shape as a Laplacian of Gaussian(LOG), A neural network model executing a DOG convolution by diffusing an external excitation is proposed. By this model intensity changes of image may be detected. This model may be implemented economically because each neuron has only four fixed-valued synapes.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.19 no.10
• /
• pp.1894-1899
• /
• 1994

In this paper, a new identification method is proposed for unknown digitally modulated input signals. The proposed identification method is implemented using a self-organized neural network which is based on the characteristic features of the symbol magnitude; the number of symbol magnitude levels, amplitude probability density and adjacent symbol magnitude ratio. The proposed method was performed for 5 QAM signals. The simulation results show that the self-organized neural network can accurately recognize all kinds of patterns even at SNR 8dB. The proposed method can be applied to the intelligent communication system on ISDN and multi-point polling networks.

• The Transactions of the Korea Information Processing Society
• /
• v.5 no.11
• /
• pp.2915-2920
• /
• 1998

This paper propose a new model for evaluationg the whole course of study by using the neural network. Through the experiment, we get the result that our model could evaluate effectively teh state of whole knowledge, because the neural network had the characterisitics such as a generalization and the ability which overcame the weakness of itself.

• The Transactions of the Korea Information Processing Society
• /
• v.5 no.4
• /
• pp.1066-1073
• /
• 1998

In this paper, we propose a new efficient operator named DBAH (difference between arithmetic mean and harmonic mean) and a technique for extracting sketch features through learning fuzzy inference rules with a neural network. The DBAH operator provide some advantages; sensitivity dependence on local intensities and insensitivity on small rates of intensity change in very dark regions. Also, the proposed fuzzy reasoning technique by a neural network has a good performance in extracting sketch features without human intervention.

• Proceedings of the KSR Conference
• /
• 2011.10a
• /
• pp.275-280
• /
• 2011

Due to high-speed of trains, the track deformation increases rapidly and may lead to track irregularities causing the track stability problem. To secure the track stability, the continual inspection on track irregularities is required. The paper presents a methodology for identifying track irregularity using the NARX neural network considering non-linearity in the train structural system. A simulation study has been carried out to examine the proposed method. Acceleration time history data measured at a bogie were re-sampled to every 0.25m track irregularity. In the simulation study, two sets of measured data were simulated. The second data set was obtained by a train with 10% more mass than the one for the first data set. The first set of simulated data was used to train the series-parallel mode of NARX neural network. Then, the track irregularities at the second time period are identified by using the measured acceleration data. The closeness of the identified track irregularity to the actual one is evaluated by PSD and RMSE.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.15 no.7
• /
• pp.4482-4488
• /
• 2014

This paper proposes the creation scheme of a threat map for robot global path planning. The threat map was generated using neural network theory by analyzing the robot's armament state and the menace information of an enemy or obstacle. In addition, the performance of the suggested method was verified using the compared result of the damage amount and existing robot path data.

• Journal of Institute of Control, Robotics and Systems
• /
• v.17 no.11
• /
• pp.1173-1178
• /
• 2011

The power generated by wind turbines changes rapidly because of the continuous fluctuation of wind speed and direction. It is important for the power industry to have the capability to predict the changing wind power. In this paper, neural network based wind power prediction scheme which uses wind speed and direction is considered. In order to get a better prediction result, compression function which can be applied to the measurement data is introduced. Empirical data obtained from wind farm located in Kunsan is considered to verify the performance of the compression function.