In this paper, we suggest a new identification method based on the takagi-sugeno fuzzy model which prepresents an envrionmental stiffness and propose a method to decide PD gains of the PD controller. It is difficult to perform a compliance task due to characteristics of robot itself and uncertain work envronment. Therefore, in this paper, we identify the fuzzy rule by dividing the relationship of input-output data into several piecewise-linear equations using the hough transform which is the one this fuzzy model, we propose a method to design the pD gain. We show the validity of this method by the experiment of tracking the surface of the paper box as an example of variable environment using robot manipulator and force sensing system. As a performance index, we use the settling time, and perform an analysis between conventional PD contorllers and this controller.

Robot manipulators constituing multi-robot system must exert the desired motion force on an object to preserve tghe fine motion of it. The forces exerte on an object by the end-effectors of multi-inducing force and the internal force. Here, motion-inducing force effects the motion of an object, but internal force as lies in the null space of an object coordinate can't effect it. The motion of an object can't track exactly the desired motion by the effect of an object, but internal force as lies in the null space of the effect of internal force component, therefore internal force component must be considered. In this paper, first, under assumption that we can estimate exactly the parameter of dynamics, we constitute paper, first, under assumption that we can estimate exactly the parameter of dynamics, we constitute the controller concerning internal force. And we obtain the internal force as projecting force sensor readings onto the space spanned by null basis set of jacobian matrix. Using the resolved acceleration control method and the fact that internal force lies in the null space of jacobian matrix, we construct the robust control law to preserve the robustness with respect to the uncertainty of mainpulator parameters.

In this paper, the properties and the design procedures of the internal model control (IMC) structures are discussed and a new nonlinear IMC(NIMC) strategy is proposed. The IMC controllers are simply inverse controller in principle but the development of a NIMC poses difficulties due to the inherent complexity of nonlinear systems. Existing design mehtods are a few and not easy to implement. The proposed approach is using multiple linear models, linear IMC controllers, and swiching scheme instead of using nonlinear model/controller. The advantages of the new approach are that we can use linear IMC mehtod which are now well estabilished and need not global nonlinear models.

In this paper, we propsoe an encryption algorithm of image information using a mass transfer function (MTF). The algorithm is based on a diffusion phenomenon of black ink when black ink dropped in the stationary water. We mathematically analyze the phenomenon, in consideration of characteristics of image information, and apply the results of analysis to the security of image transfer phenomenon. The cryptosystem proposed in this paper enables the security services of information in narrow-band channel communication network to be provided. And in transmission of important information, it can secure against intentional disturbance and violation. Also, it can guarantee the safe flow of information.

This paper proposes a KLT-CVQ scheme using PCNN to improbe the quality of the reconstructed images at a given bit rate. By using the PCNN and classified vector quantization, we exploit the high energy compaction and compelte decorrelation capbilities of the KLT, and the pdf (probability density function) shape and space-filling advantages of the vQ to improve the performance of the proposed hybrid coding technique. In order to preserve the preceptual fetures such as the edge components in the reconstructed images, we classified the input image blocks according to the texture energy measures of the local statistics and vector-coded them adaptively, and thereby reduces the possible edge degradation in the reconstructed images. The results of the computer simulations show that the performance of the proposed KLT-CVQ is higher than that of the KLT-CSQ or the DCT-CVQ in the quality of the reconstructed images at a given bit rate.

In this paper, efficient techniques for solder joint inspection have been described. Using three layers of ring shaped LED's with different illumination angles, three frames of images are sequentially obtained. From these images the regions of interest (soldered regions) are segmented, and their characteristic features including the average gray level and the percentage of highlights - refereed to as 2D features - are extracted. Based on the backpropagation algorithm of neural networks, each solder joint is classified intor one of the pre-defined types. If the output value is not in the confidence interval, the distribution of tilt angles-referred to as 3D features - is claculated, and the solder joint is classified based on the bayes classfier. The second classifier requires more computation while providing more information and better performance. The proposed inspection system has been implemented and tested with various types of solder joints in SMDs. The experimental results have verified the validity of this scheme in terms of speed and recognition rate.

In this paper we propose a method that improves PSNR at low bit rate and reduces computational complexity in fractal image coding based on discrete wavelet transform. The proposed method, which uses significant coefficient tree, improves PSNR of the reconstructed image and reduces computational comlexity of mapping domain block onto range block by matching only the significant coefficients of range block to coefficients of domain block. Also, the proposed method reduces error propagation form lower resolution subbands to higher resolution subbands by correcting error of lower resolution subbands. Some experimental results confirm that the proposed method reduces encoding and decoding time significantly and has fine reconstructed images having no blocking effect and clear edges at low bit rate.

In this paper, newly developed algorithm for enhancing images corrupted by white gaussian noise is proposed. In the method proposed here, image is subdivided into a number of subblocks, and each block is separated into cimponents corresponding to signal and noise subspaces, respectively through the signal subspace method. A clean signal is then estimated form the signal subspace by the adaptive wiener filtering. The decomposition of noisy signal into noise and signal subspaces in is implemented by eigendecomposition of covariance matrix for noisy image, and by performing blockwise KLT (karhunen loeve transformation) using eigenvector. To reduce the perceptual noise level and distortion, wiener filtering is implementd by adaptively adjusting noise level according to activity characteristics of given block. Simulation results show the effectiveness of proposed method. In particular, edge bluring effects are reduced compared to the previous methods.

On PR-IV magnetic recording systems, the maximum likelihood sequence detection (MLSD) method is optimum. But it has the problem of the complexity of the structure. The three level detection (TLD) method can be used, which has simpler structure than MLSD, but requires almost twice of power to achieve the same probbility of error as MLSD does. Therefore a new detection method (error controlled detection: ECD) is proposed which has much simpler structure than MLSD and gives much better performance than TLD. The simulation resutls show that the performance of ECD is better than that of TLD by approximaterly 0.5~1.3dB both in linear and nonlinear channels.

This paper is a study on the implementation of speech recognition mobile robot to which the interactive speech recognition techniques is applied. The speech command uttered the sentential connected word and is asserted through the wireless mic system. This speech signal transferred LPC-cepstrum and shorttime energy which are computed from the received signal on the DSP board to notebook PC. In notebook PC, DP matching technique is used for recognizer and the recognition results are transferred to the motor control unit which output pulse signals corresponding to the recognized command and drive the stepping motor. Grammar network applied to reduce the recognition speed of the recogniger, so that real time recognition is realized. The misrecognized command is revised by interface revision through the conversation with mobile robot. Therefore, user can move the mobile robot to the direction which user wants.

In this paper, the method using virtual scheduling suggested algorithm (VSSA) is suggested by considering cell delay variation and token rate of leaky bucket. This method is compared with virtual scheduling algorithm (VSA) and virtual scheduling algorithm with no tolerance excessive peak cell rate. As a result, the research shows that the usage parameter control using vSSA makes quality of service better than the usage parameter control using vSA or VSANT does because the suggested method reduces the violated cell probability of conformed peak cell rate and intentionally excessive peak cell rate.

In this paper, we present a character recognition algorithm in printed korean and english documents including touching characters. We derived two rules to segment and recognize touching characters in the bilingual documents, one from the shape characteristics of korean and english characters of the writing blocks defined in this paper, and the other from the RF (reliability factor) values generated from the classifiers. Overall classification accuracy for the KITE paper of the proposed algorithm was about 96.8% for the english abstract, and about 97.8% for the bilingual parts. Also we confirmed the proposed algorithm significantly improves the accuracy of character segmentation of the actual mixed korean and english documents including touching characters.

In this paper we propose a new controller design method using hybrid fuzzy-neural netowrk and neural network identification in order ot control systems which are more and more getting nonlinearity. Proposed method performs, for a nonlinear plant with unknown functions, hybird identification using a fuzzy-neural network and a neural network, and then a stable nonlinear controller is designed with those identified informations. To identify a nonlinear function, which is directly related to input signals, we can use a neural network which is satisfied with the proposed stable condition. To identify a nonlinear function, which is not directly related to input signals, we can use a fuzzy-neural network which has excellent identification characteristics. In order to verify excellent control performances of the proposed method, we compare the porposed control method with a conventional neural network control method through simulations and experiments with one link manipulator.

흘림체 필기 문자는 문자의 가능한 한도 펜의 움직임을 줄이려는 경제성의 원칙에서 비롯된다. 즉, 다음 획을 쓰기 위해 이동할 때 펜을 들지 않거나, 아예 이동을 생략하거나, 또는 연결된 두 획이 서로 닮아가면서 필기하기 쉬운 단순간 형태로 변화한다. 이러한 변화로 인해, 획이나 자소의 형태가 달라질 뿐만 아니라, 획간이나 자소간의 구분이 매우 어려워진다. 따라서 흘림체의 효과적인 인식을 위해서는 획이나 자소의 정확한 분할에 의존하지 않으면서, 일정한 단위로 분할하여 매칭할 수 있는 방법이 필요하다. 이 연구에서는 구조적인 형태의 단위로 분할하고 매칭하는 '곡률획 모델링 방법(curvature stroke modeling method)'을 제안한다. 곡률획(curvature stroke)은 필기의 회전 방향이 바뀌지 않는 부분획으로 정의되며, 곡률에 따라 선분, 호, 원 등의 형태를 갖는다. 흘려 써진 입력 획들을 곡률획의 나열로 변환하기 위해서는, 필기의 회전을 변화시키는 곳, 급격한 방향 변화를 일으키는 곳, 그리고 지나친 회전을 일으키는 곳 등을 분할한다. 각 참조 자소는 정자체로 입력하여 분할 과정에 의해 생성된 곡률획의 나열로 저장되어 있으며, 인식중에 융합과정을 수행함으로써 매칭을 위한 다양한 곡률획의 나열을 만들어낸다. 이때, 가상 획이 필기되거나 생략될 가능성도 고려한다. 인식의 기본 단위로 곡률획을 사용함으로써, 입력 문자의 불필요한 분할점들을 효과적으로 줄일 수 있고, 또한 자소간의 연결점을 찾기 어려운 경우에도 인접한 두 자소에 걸치는 참조 곡률획을 생성해내기 때문에 정확한 매칭이 가능해진다. 실험 결과, 83.60%의 제 1후보 인식률과 0.99초/자(CPU 클럭: 66MHz)의 처리 시간을 보였다./atom으로 추출되었다. 한편 별도의 추가적인 공정없이 일반적인 에피 성장법을 사용하여 고농도로 붕소가 도핑된 실리콘층 위에 부정합 전위가 없는 에피 실리콘을 성장시켰으며, 이 에피 실리콘의 결정성은 매우 양호한 것으로 밝혀졌다. 또 부정합 전위가 없는 에피 실리콘에 n+/p 게이트 다이오드를 제작하고 그 전압-전류 특성을 측정한 결과 5V의 역 바이어스에서 0.6nA/$cm^{2}$의 작은 누설 전류값을 나타내었다.이었다 5. 쌀의 알칼리 붕괴도는 밀양 맥후작산미가 가장 높았고 호남평야지산미가 가장 낮았는데 비해 아밀로그래프의 호화개시온도는 수원과 이리산미가 가장 낮았던 반면 밀양산미가 가장 높았다. 강하점도는 밀양산미가 가장 낮았고 다음이 이천산미가 낮았던데 비해 계화 및 이리산미가 가장 높았으며 치반점도는 이와 정반대의 경향을 나타내었다. 밥의 점성 /경도비율은 지역간 차이가 유의하지는 않았으나 남부평야산미가 중부평야산미에 비해 다소 떨어지는 경향이었다. 6. 식미와 관련이 있는 쌀 외관품질 및 이화학적 특성을 이용한 주성분 분석에서 전정보의 약 59% 설명이 가능한 제 1 및 제 2 주성분치상의 7개 품종별 6개 산지미의 분포로 보아 품종에 따라서 산지 반응이 달랐는데 대체로 자포니카와 통일형 품종군간 구분과 밀양, 중부평야 및 호남평야의 세 산지간 구분이 가능하였다. 산지내 품종간 미질변이는 남양간척지와 이리산미가 비교적 작았는데 수원산미는 이천과 남양산미의 미질변이를, 계화산미는 이리산미의 미질변이를 거의 포괄하였다.는 산불위험지역의 격자점(15km)내에 최소한 1대의 AWS 설치방안을 제시하였지만, 금후에는 15km내에서도 능선, 계곡 등 구체적인 위치확정을 위한 선행연구가 실시되어야할

Physiological observations reported and the perceptual phenomena that some simple arrays of discrete dots are perceived as a continuous curve, and others are perceived as an angular contour or as consisting of separate goups of dots. The perception of continuous curves versus discondinuous angles is mainly determined by the relative orientations of the dots, i.e., by the angles between sucessive virtual lines, whereas the length of the virtual lines gives little influence. It is important that the perception of dot stimuli is presented by vitual lines between dots in visual information processing. In this paper, on the basis of these facts of physiological observations, some experimental results on the perception of visual lines form dot stimuli are shown. Triples of the dots used in the experiment are similar to those of smith & Vos' physiological experiments.

In this paper, the method for improving the speed of convergence and learning rate of back propagation algorithms is proposed which update the learning rate parameter and momentum term for each weight by generated error, changely the output layer of neural network generates a high value in the case that output value is far from the desired values, and genrates a low value in the opposite case this method decreases the iteration number and is able to learning effectively. The effectiveness of proposed method is verified through the simulation of X-OR and 3-parity problem.

For the decreasing of switching loss, this paper designs the improved buck converter circuit and then compare with the conventional buck converter and finally analyzes the performance. When the switch is turn-off, in the buck converter using the smoothing capacitor, the flow of transient current and peak voltage having the high frequency components was generated because of the instantaneous open-state in the teminals of the switch. However, the buck converter presented in this paper eliminated this disadvantage by using the L-C circuit. Therefore, the higher PF in 0.85 is achieved than that of the buck converter using the smoothing capacitor. Also, due to the decreasing of the harmonic contain rate, the current usage rate is to be higher than that of the conventional buck converter by 4.48dB and the output voltage varient rate for the step response is to be lower 1.32 times than that of the conventional buck converter.