This paper presents an adaptive optimal learning controller for uncertian robot systems which makes use fo simple DNN(dynamic neural network) units to estimate uncertain parameters and learn the unknown desired optimal input. With the aid of a lyapunov function, it is shown that all that error signals in the system are bounded and the robot trajectory converges to the desired one globally exponentially. The effectiveness of the proposed controller is hsown by applying the controller to a 2-DOF robot manipulator.

This paper proposes a modified dynamic programming for finding the correspondence between right and left images. A dynamic programming is based on the intensity of images for stereo matching. But htis method is intended to mismatch at uniformed intensity region. To reduce thd mismatching, the stereo images are segmented to various regions with respective uniform intensity, and the different cost function has applied to the segmented region during the dynamci programming. Cost function costains jump cost. And jump cost included two parameter .alpha. and .beta. which have influence on minimum cost path. Experimental results show that the 3D shape of some stereo pairs cna be finely obtained by this proposed algorithm.

In this paper, we propose a new approach for recovering 3-D shape and extracting the reflectance properties of surface from intensity images. Photometric stereo method(PSM) is genrally based on the direct illumination. In this paper, the reflectance function is derived by interoduceing the indirect diffuse illumination in PSM and then applied to hybrid reflectance model which consists of two components; the lambertian and the specular reflectance. Under the hybrid reflectance model and the indirect diffuse illumination circumstance, the reflectance properties of sample surface can be extracting by normal sampler and then 3-D shape of an object can be recovered based on extracting reflectance properties. This method is rapid because of using the reference table and simplifies the restriction condition about the reflectance function existing in prior studies. Th erecovery efficiency in our method is better than that in prior studies. Also, this method is applied to various types of surfaces by defining general reflectance function.

We present an efficient technique, bidirectioanl inertial maximum cost search technique, for th edetection and segmentation of circular shaped objects using the spatial information around the neighborhood of the boundary candidates. This technique searches boundary candidates using local pixdl information such as pixel value and its direction. And then to exclude pseudo-boundary caused by shadows or noises, the local contrast is defined between the clique of the boundary candidates and the cliques of the background. In order to effectively segment circular shaped boundary, the technique also uses the curvature based on trigonometirc function which determines circular shaped boundary segments. Since the proposed technique is applied to the pixel cliques instead of a pixel itself, it is proposed method can easily find out circular boundaries form iamges of the PCB containing circular shaped parts and the trees with round fruits compared to boundary detection by using the pixel information and the laplacian curvature.

Skin color reappearance problem in color processing system is necessary to transform a specific color sand improve color reappearance quality as a reference color. The skin color has been situated as an important memory color not only in our lives but also in color application systems such as TV. Thus, skin color reappearance problem is more important than other color processing problem. In this paper, we propose a skin color reappearance algorithm for color enhancement in TV which use phase detector to detect the skin colors at real-time from 3.58 MHz color burst signal and color signal, comparators to discriminate the types of skin color, and micom to reprodece standard skin colors for races.

In this paper, we propose an efficient algorithm for adaptation of tracking windwo, which improves tracking performance of a correlation-based video tracker by rejecting background effect originated from a time-varying target. Th eproposed adaptation algorithm ajdusts the size of a tracking window by using the ratio of spatial gradient power in target region to that in backgorund region, which is especially adequate for a correlation-based tracker. Experimental results for synthetic and real image sequences show that the proposed method adapts a tracking window well to a time-varying target and so greatly suppresses background effect, which makes improvement of trakcing performance.

This paper proposes a multiresolution surface parameter estimation method for range images. Based on robust estimation of surface parameters, it approximates a patch to a planar surface in the locally adaptive window. Selection of resolution is made pixelwise by comparing a locally computed homogeneity measure with th eglobal threshold determined by te distribution of the approximation error. The proposed multiresolution surface parameter estimation method is applied to range image reconstruction. Computer simulation results with noisy rnag eimages contaminated by additive gaussian noise and impulse noise show that the proposed multiresolution reconstruction method well preserves step and roof edges compared with the conventional methods. Also the segmentation method based on the estimated surface parameters is shown to be robust to noise.

In radar signal processing, an A/D converter with sufficient dynamic range and high sampling speed is required to detect the weakest target signals in heavy clutter and ECM environments. As the sampling frequency increases, the amount of digital data transfered to the signal processing module is also increased. To overcome these massive data transfer burden, we need an A/D conversion module with an enough data transfer rate. In this paper, we proposed an implementation scheme of a new A/D conversio module that can be used in multi-mode 3-D phased array radar signal processing system, and evaluated the performance. The proposed A/D conversion module is implemented with a standard A/D converter and a 6U-standard VME bus.

buck type converter doesn't appear when an input voltag eis lower than an output voltage. This is the main reason the buck converter has not been used for high power factor converters. In this paper, soft switching high power factor buck converter is proposed. This converter is composed of diode rectifier, input capacitor can be small enough to filter input current, buck converter with loss less snubber circuit. Converter is operated in discontinous conduction mode, turn on of the switching device is a zero current switching (ZCS) and high powr factor input is obtianed. In addition, zero voltage switching (ZVS) at trun off is achieved and switching loss is reduced using loss less snubber circuit. The capacitor used in the snubber circuit raised output voltage. Therefore, proposed converter has higher output voltage and higher efficiency than conventional buck type converter at same duty factor in discontinous conduction mode operation. High power factro, efficiency, soft switching operation of proposed converter is veified by simulation using Pspice and experimental results.

This paper analyzes the steady-state and dynamic characteristics of the watkings-johnson converter. the major contributions of this paper includ identification of the following characteristics unique to the watkins-johnson converter: (1) The output voltage of the converter is nearly constant at the continous conduction mode (CCM), and at discontinousou conduction mode (DCM) it decreses linerly as the output curretn increases. (2) The control-to-output transfer function is a second-order one with a left-half-plane (LHP) zero determined by the combination of the output capacitor and equivalent series resistor of the converter. This LHP zero signinificantly improves the stability of the converter. (3) The control-to-output transfer function reduces to a first-order one, as the converter moves from CCM to DCM. (4) The parastic resistance of the inductor does not cause any significant influence on poles and zeros of the transfer function.

This paper presents motor speed and revolution angle detection method using a sinusoidal AC tacho-generator. The 2-phase or 3-phase output tacho-generator can be adopted, and its' output voltages must have sinusoidal waveforms. Because the detection algorithm is simple, the proosed method can be implemented with analog devices or microprocessor conveniently. And the proposed method has a very short detection delay time. Especially in the analog implementation, there is no delay time without the settling time of analog devices. With the experimental results, it is verified that the proposed method can acculately detect the instantaneous motor speed and revolution angle over the wide ranges.