• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.11a
• /
• pp.57-58
• /
• 2009

최근 MPEG에서 HD(High Definition) 해상도 이상의 초고해상도 비디오를 위한 HVC(High-performance Video Coding) 표준화에 대하여 논의가 되고 있다. 이런 흐름 속에서 ETRI에서 보유중인 베리어 방식의 CCD를 사용하여 NHK에서 제작된 RG1G2B 4K 영상은 dark current error, bias error, flat error 등에 의한 영상 자체의 열화가 많아 HVC 연구를 위한 영상으로 사용되기에는 무리가 많다. 가장 이상적인 해결방안은 NHK에서 제작한 카메라 자체에 열화제거를 위한 장치들을 설치하여 규칙적인 열화를 제거한 영상을 확보하는 것이지만, 특수 제작된 카메라여서 이 방법은 불가능하다. 본 논문은 이 NHK의 영상을 wavelet 기반의 denoise filter를 응용하여 영상의 열화를 일정부분 제거하면서 영상의 디테일이 최대한 유지되도록 하여, 기존의 영상보다 깔끔한 8K UHD 영상을 획득하는 방안을 제시한다.

• 제어로봇시스템학회:학술대회논문집
• /
• 2002.10a
• /
• pp.33.1-33
• /
• 2002

The INS provides high rate position, velocity and attitude data with good short-term stability while the GPS provides position and velocity data with long-term stability. By integrating the INS with GPS, a navigation system can be achieved to provide highly accurate navigation performance. For the best performance, time synchronization of GPS and INS data is very important in GPS/INS integrated system. But, it is impossible to synchronize them exactly due to the communication and computation time-delay. In this paper, to reduce the error caused by the measurement time-delay in GPS/INS integrated systems, error compensation methods using separate bias Kalman filter are suggested for both the...

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 2003.04a
• /
• pp.21-26
• /
• 2003

The well-conditioned observer in a stochastic system is designed so that the observer is less sensitive to the ill-conditioning factors in transient and steady-state observer performance. These factors include not only deterministic issues such as unknown initial estimation error, round-off error, modeling error and sensing bias, but also stochastic issues such as disturbance and sensor noise. In deterministic perspectives, a small value in the L$_2$ norm condition number of the observer eigenvector matrix guarantees robust estimation performance to the deterministic issues and its upper bound can be minimized by reducing the observer gain and increasing the decay rate. Both deterministic and stochastic issues are considered as a weighted sum with a LMI (Linear Matrix Inequality) formulation. The gain in the well-conditioned observer is optimally chosen by the optimization technique. Simulation examples are given to evaluate the estimation performance of the proposed observer.

• International Journal of Control, Automation, and Systems
• /
• v.4 no.3
• /
• pp.351-358
• /
• 2006

The proposed inertial measurement unit (IMU) is composed of accelerometers only. It can determine a vehicle's position and attitude, which is the Gyro-free INS. The Gyro-free INS error is deeply affected by the sensor bias, scale factor and misalignment. However, these parameters can be obtained in the laboratory. After these misalignments are corrected, the Gyro-free strap-down INS could be more accurate. This paper presents a compensator design for the strap-down six-accelerometer INS to correct misalignment. A calibration experiment is taken to get the error parameters. A simulation results show that it will decrease the INS error to enhance the performance after compensation.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2008.06a
• /
• pp.525-526
• /
• 2008

For measuring nonlinear Vehicle Modeling based on 7-Depth Sensor, the neural networks are proposed m adaptive and in realtime. The structure of it is similar to recurrent neural networks; a delayed output as the input and a delayed error between the output of plant and neural networks as a bias input. In addition, we compute the desired value of hidden layer by an optimal method instead of transfering desired values by backpropagation and each weights are updated by RLS(Recursive Least Square). Consequently, this neural networks are not sensitive to initial weights and a learning rate, and have a faster convergence rate than conventional neural networks. This new neural networks is Error Estimated Neural Networks. We can estimate nonlinear models in realtime by the proposed networks and control nonlinear models.

• Journal of Institute of Control, Robotics and Systems
• /
• v.6 no.1
• /
• pp.15-24
• /
• 2000

Stand-alone global position system receiver based on C/A code tracking generates position error of 100m mainly due to the selective availability and ionospheric and tropospheric delay errors. The differential GPS is the most commonly used method for removing those bias range error components. The relative GPS, although somewhat restrictive in its use, is ideally suited to the car monitoring system for improved Automatic Vehicle location, especially where the DGPS infrastructure is not available. The RGPS does not require any additional hardware, facility or external infrastructure and can be operated within the system with existing host computer and communication link. This paper presents detailed description of the RGPS concept and its implementation for real-time data processing. Performance of RGPS is evaluated with real data and is compared with DGPS.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• v.1
• /
• pp.133-138
• /
• 2006

Singular value decomposition (SDV) approach is applied to the observability analysis of GPS/INS in this paper. A measure of observability for a subspace is introduced. It indicates the minimum size of perturbation in the information matrix that makes the subspace unobservable. It is shown that the measure has direct connections with observability of systems, error covariance, and singular structure of the information matrix. The observability measure given in this paper is applicable to the multi-input/multi-output time-varying systems. An example on the observability analysis of GPS/INS is given. The measure of observability is confirmed to be less sensitive to system model perturbation. It is also shown that the estimation error for the vertical component of gyro bias can be considered unobservable for small initial error covariance for a constant velocity horizontal motion.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.28 no.11
• /
• pp.1621-1628
• /
• 2004

This paper proposes a design methodology for the robust observer using the eigenstructure assignment in multi-output systems so that the observer is less sensitive to the ill-conditioning factors such as unknown initial estimation error, modeling error and measurement bias in transient and steady-state observer performance. The robustness of the observer can be achieved by selecting the desired eigenvector matrix to have a small condition number that guarantees the small upper bound of the estimation error. So the left singular vectors of the unitary matrix spanned by space of the achievable eigenvectors are selected as a desired eigenvectors. Also, this paper proposes how to select the desired eigenvector based on the measure of observability and designs the observer with small gain. An example of a spindle drive system is simulated to validate the robustness to the ill-conditioning factors in the observer performance.

• Wind and Structures
• /
• v.32 no.2
• /
• pp.161-167
• /
• 2021

In this study, the artificial neural network (ANN) method was used for estimating the monthly mean wind speed of Sivas, in the central part of Turkey. Eighteen years of wind speed data obtained from nine measurement stations during the period of 2000-2017 at 10 m height was used for ANN analysis. It was found that mean absolute percentage error (MAPE) ranged from 3.928 to 6.662, mean bias error (MBE) ranged from -0.089 to -0.003, while root mean square error (RMSE) ranged from 0.050 to 0.157 and R2 ranged from 0.86 to 0.966. ANN models provide a good approximation of the wind speed for all measurement stations, however, a tendency to underestimate is also obvious.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.21 no.3
• /
• pp.109-125
• /
• 1996

Biases and errors in the human being's reasoning process have been studied continuously by the researchers, especially psychlogists and social scientists. These bias phenomenon is classified on the basis of the origin, i. e. motivation and cognition. Furthermore the necessity of research on the bias in the management and management information system areas in increased more and more recently, which have their academic backgrounds in the psychology and social science. The biased information stream is transformed into the systematic error due to the motivation and cognitive bias of human-being, then its resulting phenomena are as follows; 1. the availability of salient information 2. preconceived ideas or theories about peoples and event 3. anchoring and perseverence phenomena. In order to reduce the information errors, Satty suggested the Analytic Hierarchy Process (AHP) that is the subject of this paper and that is widely used for evaluation of complex decision making alternatives. THerefore this paper studies AHP's effects and its limitations in applying to the management area. Thus this paper compared the performances of the 3 models : 1 the traditional additive regression model. 2 regression model using the factor score, and 3 the regression model with AHP. As a result, 3 models produce the different outcomes.