In this paper, we propose a method of designing neural networks using biological inspired developmental and evolutionary concept. The living things are best information processing system in themselves. One individual is developed from a generative cell. And a species of this individual have adapted itself to the environment by evolution. Ontogeny of organism is embodied in cellular automata and phylogeny of species is realized by evolutionary algorithms. The connection among cells is determined by a rule of cellular automata. In order to obtain the best neural networks in the environment, we evolve the arrangement of initial cells. The cell, that is neuron of neural networks, is modeled on chaotic neuron with firing or rest state like biological neuron. A final output of network is measured by frequency of firing state. The effectiveness of the proposed scheme is verified by applying it to navigation problem of robot.

In this paper, Fundamental rules governing group intelligence "arrangement" behavior of multiple number of autonomous mobile robots are represented by a small number of fuzzy rules. Complex lifelike behavior is considered as local interactions between simple individuals under small number of fundamental rules. The fuzzy rules for arrangement are generated from clustering the input-output data obtained from the arrangement algorithm. Simulation shows the fuzzy rules successfully realizes fundamental rules of the flocking group behavior.

The objective of this paper is, based upon the principles of artificial life, to induce emergent behaviors of multiple autonomous mobile robots which form from simple local rules to complex global intelligence. Here, we propose an architecture of neural network learing with reinforcement signals which perceives the neighborhood information and decides the direction and the velocity of movement as mobile robots navigates in a group. As results of the simulations, the optimum weights are obtained in real time, which not only prevent from the collisions between agents and obstacles in the dynamic environment, but also have the mobile robots move and keep in various patterns.

Cellular automata are dynamical systems in which space and time are discrete, where each cell has a finite number of states and updates its states by interactive rules among the cell-neighborhood. From the characteristics of self-reproduction and self- organization, it is possible to create a neural network which has the specific patterns or structures dynamically. CAM-Brain is a kind of such neural network system which evolves its structure by adopting evolutionary computations like genetic algorithms (GA). In this paper, we suggest the evolution strategies for the structure of neural networks based on cellular automata.

Many multiagents cooperative problems, such as table transport problem, require several emergent behaviors and a proper coordination of these is essential for successful accompishment of the task. We study in this paper the genetic programming method, called fitness switching, to evolve cooperation strategies of robots in these kind of tasks and show simulation results to demonstrate its effectiveness.

Autonomous mobile robot provide many functions such as sensing, processing, and driving. For more intelligent jobs, more intelligent functions are to be added and the existing functions may be updated. To execute a job autonomous mobile robot has a information of surrounding environment. So, robot uses sonar sensor, vision sensor and so on. Obtained sensor information is used map generation. This paper is focused on map generation using stereo vision system.

This paper describes fuzzy radial basis function networks(FRBFN) extracting fuzzy rules through the learning from training data set. The proposed FRBFN is derived from the functional equivalence between RBF networks and fuzzy inference systems. The FRBFN learn by assigning new fuzzy rules and updating the parameters of existing fuzzy rules. The parameters of the FRBFN are adjusted using the standard LMS algorithm. The performance of the FRBFN is illustrated with function approximation and system identification.

The modular Lateral Information Propagation Networks(LIPN) has been designed. The LIPN has shown to be useful for interpolation of information[3]. The problem is the fact that only the small number of nodes can be implemented in a IC chip with the circuit VLSI technology. The proposed modular architecture is for enlarging the neural network through inter module connections. For such inter module connections, the host(computer or logic) mediates the exchange of information among modules. Also border nodes in each module have capacitors for temporarily retaining the information from outer modules. The LIPN with $4{\times}4$ modules has been designed and simulation of interpolation with the designed LIPN has been done.

This paper presents the matching algorithm with the satellite images using the image warping method. Two stereo images, which are used for the DEM(Digital Elevation Model) extraction, are generally distorted because the images are acquired at different locations and angles. Therefore, the matching Process can't be executed with the original images. To solve this problem, a dynamic triangular image warping method is proposed. At first, the initial matching is executed with seed point, and then, using the matched points from the initial matching, the distorted images is compensated. We experimented this algorithm with the parts of the $6000{\times}6000$ SPOT satellite images. The experiment results show this algorithm is superior to other warping algorithm.

As there comes out better and better solutions for delivering multimedia data on the internet, the number of WebCAM sites which transmits pictures of any spot to the user's system is getting increased. WebCAM service over the internet can be compared with the CATV in that it is beyond the spatial constraints. In this project, we construct a WebCAM site which performs not only the fundamental WebCAM functions of capturing images and of transmitting them periodically but also various kinds of modules carrying out tasks such as image processing, remote control of the camera, processing multiple user requests, making panorama views and so on using Java which is the most popular language for the internet or CGI. The past images are stored in a database implemented for the WebCAM system, so the system can be extended to get over the constraints in time also.

A remote maintenance system, which enables an engineer to check and treat machines installed at remote site with the help of data communication, has been developed. The maintenance of machines with the traveling of engineers is much costly in time and economically. With the developed system, the engineer at the remote site can diagnose malfunctioning machines. The remote maintenance system is composed of video conferencing system, camera control system and signal aquisition system. The developed system has been tested on the 100Mbps ATM network to evaluate its ability to diagnose remote machines.

Data compression of 3 dimensional objects has limitations such as large data size, slow processing time, uncertainties of noise and measurement errors. Recently, along with activities of information transmission in internet, the researches in 3 dimensional information processing become a very important issue as the exchanges of large data are available. In this paper, we propose a data compression algorithm which first estimates the surfaces of 3 dimensional objects and then apply them to the FCV(Fuzzy c-Varieties) Clustering to obtain 3 dimensional edges and vertices, and represent the results.

This paper presents a 3D feature extraction method using rough set theory. Using the stereo cameras, we obtain the raw images and then perform several processes including gradient computation and image matching process. Decision rule constructed via rough set theory determines whether a ceratin point in the image is 3D edge or not. We propose a method finding rules for 3D edge extraction using rough set.

An efficient stereo matching method with variable matching weights is proposed. The edge segment-based stereo matching has been shown to be efficient method. The method includes 5 matching factor with different weights. The ordinary matching weights are not always adequate for every image. Employing different weight sets depending on the complexity shows better matching performance. In this paper, an evaluation criterion for complexity is suggested and the experimental results with the proposed method is shown.

This paper presents the development of depth extraction algorithm for the stereoscopic endoscope data using a stereo matching method. generally, the purpose of existing stereo algorithms is to reconstruct stereo object surface and depth map. but the main purpose of our processing is to give exact depth feeling to doctor showing depth information in some points. for this purpose, this paper presents two stereo matching algorithms which are to measure exact depth. one is using variable window, and the other is reference points-based algorithm for a fast processing.

본 논문은 손등의 정맥패턴을 이용한 개인식별을 위해 개발된 3가지의 알고리즘에 관해 각각의 성능을 비교한 결과를 제시한다. 세가지 방법은 각각 Unsharp Masking을 이용한 이치화 후 정맥과 손등 배경의 면적을 이용한 가중치를 적용한 원형정합 알고리즘[1]과 Kolmogorov Smirnov(KS) statistic[2]을 이용한 매칭 알고리즘을 개선한 방식, 그리고 정맥의 세선화 처리 후 분기점의 좌표, 정맥의 길이, 정맥 가지 사이의 분기각도 등의 특징벡터를 이용한 방법 등이다. 본 연구에서는 전처리 과정에 있어서, 원시영상의 혈관부위와 배경부위의 gray scale 분포가 겹친 상태에서 Unsharp Masking 필터링을 적용한 결과가 기타 다른 전처리 방식보다 우수하게 영상을 강화시킬 수 있음을 확인하였고, 가중치를 이용한 매칭방식이 다른 매칭방식보다 우수함을 확인하였다.

In this paper, we propose a new algorithm based on multi-resolution application of the parameter space to the Hough transform technique. The existing Hough transform technique employs mapping of fixed parameter space in order to extract straight lines from image. One of the difficulties of the existing Hough transform technique lies in the detection of multiple adjacent lines for only one line. Increasing the parameter space from the low level resolution to the high level resolution, our algorithm detects straight line in a stable and efficient fashion. Experimental results are included to verity the performance of proposed algorithm.

A limitation is assumed that In this paper, a generalized method is proposed to extract a period of a motion of on object. To detect a periodic motion, we put restrictions on a stationary camera and on a motion of an object. We ca derive the necessary and sufficient condition that an image sequence consists of the projection of the periodic motion by the affine transformation that is a reasonally good approach to the perspective projection. The difficulty of detecting its periodic motion is to select its have period in sequence and to define its width.

Image coding based on fractal theory presents highly compressed image. In this paper, we discuss about compression of Image using HV partition method. HV partition scheme devides the image adaptively in horizontal and vertical axis. And for reducing the enconding time for the domain-range comparison, we use classification scheme, which uses the order of brightness of the rectangular portion of the image. This paper focused on the technique to reduce coding time which is a problem in traditional fractal compression by adaptive selection of image and its classification method.

In this paper, we treat image segmentation as a graph partitioning problem. and use the normalized cut for segmenting the graph. The normalized cut criterion measures both the total dissimilarity between the different graphs as well as the total similarity within the groups. The minimization of this criterion can formulated as a generalized eigenvalues problem. We have applied this approach to segment static image. This criterion can be shown to be computed efficiently by a generalized eigenvalues problem

To control the remote system in the unstructured environment requires data under certain circumstances. When a machine is dealt with an unstructured environment, new environment structure is to be composed. The stereo vision system can get both the intensity data and the range data. So, in this paper, data architecture of a stereo image is proposed to set them.

This paper presents a lip print recognition by the pattern kernels for a personal identification. A lip print recognition is developed less than the other physical attribute that is a fingerprint, a voice pattern, a retinal blood-vessel pattern, or a facial recognition. A new method by the pattern kernels is pro for a lip print recognition. The pattern kerne function consisted of some local lip print p masks. This function identifies the lip print known person or an unknown person. The results show that the proposed algorithm the pattern kernels can the efficiently realized.

As the state of the art for handling multimedia on the internet is developed a new system called WebCAM which services video and image came out. However, WebMIC site for audio services are still rare. In this work, we developed a WebMIC system which enables anyone to listen audio in real-time through the internet. WebMIC system is implemented by connecting a server with a microphone and the recorded audio files are stored in a Database system. Because the microphones can be distributed sporadically, the developed system can be utilized for managing and advertizing a location. It also can be used to reduce the cost for the existing management system which is almost met by personnel expenses. Anyone who can connect to the internet can get this WebMIC service, so this system can produce a good effect if used in advertising any tourist resort.

Method of diagnosis in oriental medicine, the unbalance of the physiological function of the five viscers and six bowels of the human body is determined from time immemorial with the condition of blood circulation which is performed through blood vessels by the vitality of the heart. In oriental medicine, treatment is largely attempted by adjusting this unbalance. The analysis of pulse wave, which mainly measures the changes in blood flows, is to evaluate the shapes of a pulse wave rather than the quantitative changes like rates and strength of the pulse. This paper presents the development of Hardware System and Pulse Diagnosis Algorithm for automatic diagnosis of the pulse wave. This system makes the precise diagnosis and the objective recording possible.

Auditory Models that are capable of achieving human performance would provide a basis for realizing effective speech processing systems. Perceptual invariance to adverse signal conditions (noise, microphone and channel distortions, room reverberations) may provide a basis for robust speech recognition and speech coder with high efficiency. Auditory model that simulates the part of auditory periphery up through the auditory nerve level and new distance measure that is defined as angle between vectors are described.

This paper propose a new method for the discrete signal : Discrete Wavelet Transform(DWT). This paper is a brief introduction to the DWT and applies the DWT coding for the audio data as an example. We can have a number of hint about the compression algorithm of multimedia resources and the high performance of transmission and storage.

This paper deals with study of Fast Speaker Adaptation Type Speech Recognition, and to analyze speech signal efficiently in time domain and time-frequency domain, utilizes SCONN[1] with Speech Signal Process suffices for Fast Speaker Adaptation Type Speech Recognition, and examined Speech Recognition to investigate adaptation of system, which has speech data input after speaker dependent recognition test.

This paper proposes a method which can resolve the problem of existing fuzzy Pl controller using optimal scaling gains obtained by genetic algorithm. The new method adapt a fuzzy logic controller as a high level controller to perform scaling gain algorithm between two pre-determined sets.

This paper presents recognition and classification of high impedance fault(HIF) patterns in the electrical power systems based on chaotic features. Chaotic features are obtained from two dimensional chaos attractors reconstructed from fault current waveform. The RBFN is trained with the two types of HIF data generated by the electromagnetic transient program and measured from actual faults. The RBFN successfully classifies normal and the three types of fault patterns based on the binary chaotic features.

This paper presents an algorithmic sensor value validation technique that can systematically explore the embedded sensor redundancies in complex physical systems and maximize their utilization in validating sensor values. The confidency criteria are developed for checking the consistency of sensor relationships called Causal Relations. Development results are applied to a tubular type supercritical pressure type thermal power plant with rated operational data to demonstrate the effectiveness of the proposed technique.

This paper discusses a technique that can detect cascading disturbances for automated monitoring and diagnosis systems in power plants. A multi-layered directed AND/OR graph called a disturbance interrelation analysis graph (DIAG) is utilized to represent the relationships among cascading disturbances and trace them. Disturbances that cannot be observed directly from sensors can be traced using techniques similar to interpolations and extrapolations on the DIAG.

In this paper, we propose a digital differential protection of power transformer using intelligent schemes. Intelligent schemes is based on fuzzy logic and neural networks. To enhance the distinction between fault and inrush of conventional approaches, relaying technique by fuzzy logic and neural networks are used. We used transformer inrush currents, external and internal fault signals, which are obtained from EMTP simulation.

The Rotor Stress Indicator is an integrated system of hardware and program components which has been designed to read an assortment of turbine temperature and speed input devices, perform an analysis of the temperature induced stresses and output pertinent temperature and stress information to guide the turbine operator during turbine prewarming, start-ups, load changes, and shut-downs. The purpose of the RSI is to provide guidance to the plant operator during startup, shutdown, loading, and unloading of the turbine. Since the stresses are a function of the temperature changes to which the turbine is exposed, the RSI also provides guidance for operation of the boiler main steam and reheat steam temperatures as they affect the rotor stresses. This may permit more efficient overall boiler turbine start-ups. In this paper, new rotor stress analysis algorithm for RSI is introduced and compared with present system which has been used in thermal power plant.

This paper introduces Two-stage neural network that is capable of recognizing spatio-temporal patterns. First stage takes a spatio-temporal pattern as input and compress it into sparse spatio-temporal pattern. Second stage is for temporal pattern recognition with nonuniform inhibitory connections and different cell sizes. These are basic properties for detecting a embeded pattern in a larger pattern. The network is evaluated by computer simulation.

In this paper, an improved on-line neural network model is suggested. This neural network is designed to store and recall sequence of key strokes in on-line. The network stores incoming patterns as weight connections between series of layers. The layer has a 2-dimensionally distributed neurons where the location of neurons are relevant to the actual location of computer keyboard. To store longer patterns, the network has circulating layer connections and different patterns can be superposed on the same layer. Also, when the patterns are stored over the layers, the starting layer is not fixed but changed by the characteristics of Patterns to increases network capability. The ways how to choose the starting layer during the store and recall process are investigated.

In this paper, an on-chip learning pulse-mode digital neural network with a massively parallel yet compact and flexible network architecture is suggested. Algebraic neural operations are replaced by stochastic processes using pseudo-random sequences and simple logic gates are used as basic computing elements. Using Back-propagation algorithm both feed-forward and learning phases are efficiently implemented with simple logical gates. RNG architecture using LFSR and barrel shifter are adopted to avoid some correlation between pulse trains. Suggested network is designed in digital circuit and its performance is verified by computer simulation.

In this paper, a modified RBF (Radial Basis Function) neural network structure is suggested for the prediction of time series with non-linear, non-stationary characteristics. Conventional RBF neural network predicting time series by using past outputs is for sensing the trajectory of the time series and for reacting when there exists strong relation between input and hidden neuron's RBF center. But this response is highly sensitive to level and trend of time serieses. In order to overcome such dependencies, hidden neurons are modified to react to the increments of input variable and multiplied by increments(or decrements) of out puts for prediction. When the suggested structure is applied to prediction of Lorenz equation, and Rossler equation, improved performances are obtainable.

In this paper, a control algorithm which enables robot to cooperate with human is proposed. The method senses the humanbeing's intention by using force/torque sensor attached at the end effector and moves and cooperates as intended by humanbeing. The method also considers safety of the humanbeing by adjusting and limiting the robot speed automatically. The proposed method is verified its performance by computer simulation and experiments for the 2-DOF DD(Direct Drive) Arm in real-time

The attitude control of a double inverted pendulum with a periodical disturbance at link top is dealt in this paper. The proposed system is consisted of the double inverted pendulum and a disturbance link. The lower link is hinged on the plate to free for rotation in the vertical plane. The upper link is connected to the lower link through a DC motor. The DC motor is used to control the posture of the pendulum by adjusting the position of the upper link. The periodical disturbance can be generated by the additional link attached at the end of link 2 through another DC motor, which is the modeling of a posture for a biped supporting with one leg. The motor for the joint simulates the knee joint(or hip joint) and the disturbance for the legs moving in air. The algorithm for controlling a proposed inverted pendulum is consisted of a state feedback control and a fuzzy logic controller. The fuzzy controller keeps the center of gravity of the biped within the specified range through the nonlinear feedback compensator. The state feedback control takes over the role to maintain a desired posture regardless the disturbance at the link top. In these case, the change of the angle and COG of an upper link is compensated with on-line. Simulations with a mathematical model are conducted to show the validity of the proposed controller.

To improve the visual servoing performance, several strategies were proposed in the past such as redundant feature points, using a point with different height and weighted selection of image features. The performance of these visual servoing methods depends on the configuration between the camera and object. And redundant feature points require much computation efforts. This paper proposes the visual servoing m based on the disturbance observer, which compe the upper off-diagonal component of image fe Jacobian to be null. The performance indices su sensitivity for a measure of richness, sensitiv the control to noise, and controllability are sho improved when the image feature Jacobian is giv a block diagonal matrix. Computer simulation carried out for a PUMA560 robot and show results to verify the effectiveness of the pro method.

There has been extensive efforts about robot path planning. Some major approaches are the roadmap approach, potential field approach and the cell decomposition approach. However, most of the path planning methods proposed so far based on above approaches consider the terrains filled with binary obstacles, i.e., if there exists an obstacle, robot simply cannot pass the location. In this paper, A mobile robot path planning method based on the cell decomposition technique for mobile robot that takes account of the terrain with varing degrees of travers-ability is discussed.

A genetic algorithm for global and local path planning and collision avoidance of mobil robot in unstructured workspace is proposed. The genetic algorithm searches for a path in the entire and continuous free space and unifies global path planning and local path planning. The simulation shows the proposed method is an efficient and effective method when compared with the traditional collision avoidance algorithms.

A control system using a PLC has been developed for a dual tower drier(DTD) in a CANDU type nuclear power plant. This system will replace the existing DTD control system which was implemented with mechanical timers and relays. The new control system makes it possible for an operator to perform more precise time and dew point control for the DTD, thanks to the high efficiency and flexibility of the PLC. The operational cost for the control system is much reduced compared to the existing system.

Plant controller using VME bus is developed. This controller is a PLC designed to control up to 768 I/O units of POSFA PLC which has been developed already by POSCON. In many programming aspects, it adopts IEC 1131, international standard of PLC programming to keep up with international trends. A system software and a controller architecture including CPU board, DSP board and bus extension board are developed to support the IEC 1131, and becomes a base of E.I. controller being developed now.

In this paper, an algorithm to make a configuration system for managing the I/O device of programmable logic controller(PLC) is proposed. Parallel processing architecture is used to deal with a number of I/O devices. From that architecture, a contention problem between processors can arise. To resolve this problem, the configuration system that contains informations about I/O devices is introduced. This configuration system is used to check the contention between processors in the I/O device and also used in program execution.

This paper suggests a method to recognize the various defect patterns of cold mill strip using binary decision tree automatically constructed by genetic algorithm. The genetic algorithm and K-means algorithm were used to select a subset of the suitable features at each node in binary decision tree. The feature subset with maximum fitness is chosen and the patterns are classified into two classes by a linear decision boundary. This process was repeated at each node until all the patterns are classified into individual classes. The final recognizer is accomplished by neural network learning of a set of standard patterns at each node. Binary decision tree classifier was applied to the recognition of the defect patterns of cold mill strip and the experimental results were given to demonstrate the usefulness of the proposed scheme.

Recently, the development of computer achieves a system which is similar to the mechanics of human visual system. The 3D measurement using monocular vision system must be achieved a camera calibration. So far, the camera calibration technique required reference target in a scene. But, these methods are inefficient because they have many calculation procedures and difficulties in analysis. Therefore, this paper proposes a native method that without reference target in a scene. We use the grid type frame with different line widths. This method uses vanishing point concept that possess a rotation parameter of the camera and perspective ration that perfect each line widths into a image. We confirmed accuracy of calibration parameter estimation through experiment on the algorithm with a grid paper with different line widths.

This paper describes an active laser sensing device using laser diodes and optic devices. It is able to detect near targets existed in $360^{\circ}$ directions simultaneously with effectiveness and reliability. This sensing device consists of four laser transmitters and four receivers. Only four transmitter/receiver channels of this near range sensing device are capable of $360^{\circ}$ coverage. The usefulness of this sensing device is confirmed through some experiments for the mock-up targets.

This paper is concerned with an energy optimization for the walking of IWR biped robot. The movement of balancing joints are determined by ZMP(Zero Moment Point) and dynamic properties caused by motion of a swing leg. Therefore, ZMP positions have an important role in walking and guarnateeing the stability of a robot. A genetic algorithm is utilized for solving this problem and finding ZMP with a minimum energy at each sampling time during the walk. In this study, we performed an energy optimization with desired ZMP trajectories and motion of balancing joints.

In this paper, the path planner is presented for a robot to achieve an efficient path forward the given goal position in two dimensional environment which is involved with partially unknown obstacles. The path planner consists of three major components: off-line path planning, on-line path planning, and modification of planned path. Off-line path planning is based on known environment and creates the shortest path. On-line path planning is for finding unknown obstacles. The modification can be accomplished, by genetic algorithm, to be smooth path for preventing slippage and excessive centrifugal force.

This paper deals with the trajectory planning of IWR biped robot using genetic algorithm. The trajectory of a swing leg is generated by 5th order polynomial equation. Velocities and Acceleration properties on a viapoints are needed. These constants are given by heuristic method. The optimal values are determined by G.A to minimize the jerk of a trajectory. As a result, trajectory planning is implemented not on between two viapoints but on a whole interval. Efficient numerical calculation routines and walking algorithms for simulation are accomplished by MATLAB package.

The hybrid control system for a wheeled mobile robot with nonholonomic constraints to perform a cluttered environment maneuver is proposed. The proposed hybrid control system consists of a continuous state system for the trajectory control, a discrete state system for the motion and orientation control, and an interface control system for the interaction process between the continuous dynamics and the discrete dynamics The continuous control systems are modeled by the switched systems with the control of driving wheels, and the digital automata for motion control are modeled and implemented by the abstracted motion of mobile robot. The motion control tasks such as path generation, motion planning, and trajectory control for a cluttered environment are investigated as the applications by simulation studies.

This paper suggests the wheel controller for PWS(Power Wheeled Steering) mobile robot. The proposed controller consists of two parts. To control each motor, the sliding mode controller implemented. This method has robustness about modeling error and disturbance, so the velocity tracking is well guaranteed in the presence of varying load. The design of a fuzzy cross-coupling controller for a PWS mobile robot is described here. Fuzzy cross-coupling control directly minimizes the tracking error by coordinating the motion of the two drive wheels. The fuzzy cross-coupling controller has excellent disturbance rejection and therefore is advantageous when the robot is not loaded symmetrically. The capability of the proposed controller was verified through the computer simulation.

We present a fuzzy model for a robot manipulator and use the model to decide the PD gains of a stiffness controller. Force control applications are extremely difficult to accomplish with such a stiffness robot because robot itself, unknown environment. So we identify a fuzzy model by using Hough transform. We present a method of design of the PD gains of the stiffness controller. We aim at controlling the end-effecor force in the face of uncertainty on the surface stillness. simulation results verify the effectiveness of the proposed strategy.

This paper proposes trajectory tracking of the moving object based on one camera vision system. And, this system proposes a method which robot manipulator grips moving object and predicts coordinate of moving objcet. The trajectory tracking and position coordinate are computed by vision data acquired to camera. Robot manipulator tracks and grips moving object by vision data. The proposed vision systems use a algorithm to do real-time processing.

This paper presents a new collison-free trajectory problem for SCARA robot manipulator. we use artificial potential field for collison detection and avoidance. The potential function is typically defined as the sum of attractive potential pulling the robot toward the goal configuration and a repulsive potential pushing the robot away from the obstacles. In here, end-effector of manipulator is represented as a particle in configuration space and moving obstacles is simply represented, too. we consider not fixed obstacle but moving obstacle in random. So, we propose new distance function of artificial potential field with moving obstacle for SCARA robot. At every sampling time, the artificial potential field is update and the force driving manipulator is derived from the gradient vector of artificial potential field. To real-time path planning, we apply very simple modeling to obstacle. Some simulation results show the effectiveness of the proposed approach.

The intent of this paper is to describe a neural network structure called dynamic neural processor(DNP), and examine how it can be used in developing a learning scheme for computing robot inverse kinematic transformations. The architecture and learning algorithm of the proposed dynamic neural network structure, the DNP, are described. Computer simulations are provided to demonstrate the effectiveness of the proposed learning using the DNP.

This paper presents simulation experiment results of the fuzzy controller using genetic algorithms for robot manipulator. The fuzzy controller consists of 9 quantized levels and 25 fuzzy rules. In the simulations, the population size of each generation is set to be 100. The maximum number of generations is 200. The simulation experiment results show the effectiveness of the proposed the fuzzy controller using genetic algorithms.

Recently we have developed an active stereo head-eye system with vergence, name it KIST HECter(Head-Eye System with Colored Stero Vision), based on the analysis of human's neck and eye motion at visual behavior. Our HECter is a five degree-of-freedom system composed of pan and tilt motion in neck part and independent vergence motion of binocular cameras and commonly shared elevation axis in eye part. And stereo vision Provides two color image, which are processed by powerful each TMS32080 vision board. The shape and size are designed to be almost same as human face. The ability to vergence has significant importance and gives many beneficial merits. On its mechanical implementation we adapt a non-parallelogram 4-bar linkage mechanism since it provides high accuracy in transfering motion and enables compact and flexible design.

This paper describes the development of the SCADA(Supervisory Control and Data Acquisition) system which can be controlled via the Internet. In this paper, the SCADA system is composed of a number of microprocessor-based RTU(Remote Terminal Unit)s, a MMI(Man Machine Interface) host, a SCADA server, and SCADA clients. There are two protocols used in the system. Each RTU and the MMI host are connected by a RS-485 line and CSMA/CD(Carrier Sense Multiple Access / Collision Detection) protocol is used to communicate with each other. TCP/IP(Transmission Control Protocol/Internet Protocol) is used among the MMI host, the SCADA server, and SCADA clients. The equipments installed in the field are controlled by a number of RTUs. The function of the MMI host is to acquire real-time data from RTUs and control them. The SCADA server supports data transfer between the networked MMI host and the SCADA client on the web-server through TCP/lP. Data transfer is possible regardless of the type of network only if there are TCP/lP Winsock-compatible stack driver. The SCADA client is implemented as the shape of web-page by means of JAVA language. Therefore, it runs on a web-browser such as Netscape and Explorer, and allows a number of users to access this SCADA system.

본 논문에서는 원자력 발전소의 제어 통신망을 위해 설계된 통신망 접속장치의 특징을 살펴보고 그에 적합한 통신망 관리의 구조와 서비스를 제안한다. 제안된 통신망 관리는 통신망의 초기화, 파라미터의 조정, 그리고 오류 처리를 담당한다. 제안된 통신망 관리는 물리 매체의 이중화, 통신망 접속장치의 이중화 그리고 시스템 이중화 등을 고려하여 설계되었다.

직류직권모터는 전동지게차와 같은 물류용 전동차에서 사용되는데, 우수한 기동 토오크를 가지는 반면에 파라미터의 열적, 변화가 심하고 마찰과 부하의 비선형성이 존재해 기존의 제어기로는 만족할 만한 성능을 내지 못한다. 본 논문에서는 이를 해결하기 위해 퍼지제어기를 사용한다. 퍼지제어기는 변수의 애매성에 바탕을 두고 제어하기 때문에 이러한 비선형성에 대해 강인하나, 소속함수의 결정과 퍼지규칙의 선정이 어려우며, 체계적인 방법이 존재하지 않는다. 이러한 퍼지제 어의 결점을 해결하기 위해 소속함수는 유전 알고리즘을 통해 자기동조 시키며 퍼지규칙은 오차와 오차변화율의 위상평면을 이용하여 결정한다. 실용성을 검증하기 위해 TI사의 DSP TMS320F240을 이용해 실시스템에 적용했으며, 이를 통해 부하의 변동 및 기준 속도의 변화에도 잘 대처함을 알 수가 있었다.

In this paper, the system which can read the frequency and the power values of the generators, calculate the S.R., display the on-line graph and the off-line graph and manage the database of the generators was developed. This system should be more faster, comfortable and powerful than the present analysis method of the S.R. in real fields.

In this paper, Computer Colour Kitchen System (CCKS) is developed on the basis of Delphi package and robust control concept. CCKS is widely used in the colour dye making industry as a dispensing system which have more advantages than the conventional, when controlling the real processes and systems. This decreases the running time and increases the accuracy and confidence of real system.

In this paper, we discuss the development of monitoring system with data process equipment which transfers data from Remote Terminal Unit(RTU) to monitoring computer. The RTUs sense temperature, pressure and PLC(Programmable Logic Controller) nodes conditions of scrubber in semiconductor manufacture processing. The data Process equipment is connected every RTU and a monitoring computer through serial communication. This equipment receives informations from RTU, process data, and transfers them to monitoring computer. To avoid congestion in data communication, task scheduling algorithm used RT O/S(Real-Time Operating System) is embedded in ROM which is a part of data Process equipment.

Today it is hard to imagine the operation of power network and its facilities without the control systems such as EMS/SCADA/DAS. In KEPCO, data links to the upper layer of control system (SCABA) is required to launched DAS project lately, in terms of the needs of saving cost for the data gathering facilities (RTU) in the substation that previously equipped SCADA RTU. In this Paper we researched the circumstance of the KEPCO SCADA systems for data link and present the scheme of the data link between DAS and SCADA.

The Distribution Automation System(DAS) Master Station(MS) is the core of the system, which has the functions of supervising, data acquisition, processing and man-machine interface. In this paper, we suggested the optimal configuration of the medium/small scale DAS MS in accordance with the consideration of various requirements such as reliability, scalability flexibility, etc.

For a medium/small scale DAS, which using the PCS network as the system communication network, we developed the scheme of communication protocol of DAS between master and remote terminal. This paper presents advanced protocol which has basic frame format and data flow for the medium/small scale DAS, and the procedure of the developing sequence.

This paper proposes a translation method that converts ladder diagrams to binary executable codes for PLC (programmable logic controller)s. A PLC based on general purpose DSP(digital signal processor) validates the method. We performed a benchmark on the system that compares the execution time of the interpretation method and ours. Experimenal result shows how fast this method executes programs that consist of codes generated.

Implementation of a process control language for a pager manufacturing is dealt in this paper. The pager manufacturing process is consisted of a tune and an exam part. In the tune part, three capacitor trimmers are tuned to optimize the state of the pager. In the exam part, message receiving status with the arbitrary calling is tested using a vision system. The program has the ability to reuse the address discarded in the exam part when the, pager testing is not working properly even though there exist the processing gaps between two processes. The system is composed of a personal computer(PC586) and TEM-cells, Flex pager testers, an oscilloscope, a camera, and actuators. Visual Basic running on the Windows 95 is used to implement the control software which has the GUI to make an operator convenient. The validity of applying the completed program to practical machines, which are developing in the KITECH, is shown by using the virtual machine.

In this paper, using tile concept of block digital filtering, and the design procedure of time-interleaved oversampling converter are presented. it is shown that arbitrary sigma-delta A/D converter can be converted into corresponding time-interleaved structure. The TIM structure of this paper is designed by the bilinear transform. To verify the simulation results, a second-order TIM structure A/D converter has been implemented and the design process as well as experimented results are presented.

This paper presents real-time implementation of speech vocoder for PSTN video telephony using ITU G.723 16Kbps ADPCM algorithm. The ADPCM encoder accepts 8-bit PCM compressed signals and expends it to a 14-bit-per-sample. The predicted values are subtracted from encoded signals to produce difference signals. Adaptive quantization is performed on the difference signal to produce a 2-bit, output for transmission over the channel. Computer simulations and experiments were performed to evaluate the performance of the speech vocoder.

In this paper, a new active noise control technique which combines a hybrid active noise control technique and a subband active noise control technique is presented. To reduce the computational burden the active noise control system, a weighted hybrid control technique is applied. This algorithm shows higher stability and good noise attenuation property in either broad band or narrow band active noise control systems in HVAC ducts. Computer simulations were performed to show the effectiveness of the proposed algorithm.

We developed a cement mill control system to improve performances of cement mill and conveyance facilities in view of TPM(Total Productive Maintenance). The system replaces the conventional sequence control relay panel and the mimic panel by the PLC(Programmable Logic Control) and the operation station with a monitor, respectively. In addition, the updated A3A CPU is added to the melsec PLC. All networks of the system are operated through ethernet communications. This system as a domestic product is setup at S company and outperforms other foreign products in views of operation efficiency and process stability.

We propose a robust equalization algorithm to improve the performance of airplane flutter casuing dynamic ghosts. The algorithm is an augumentation of the DFE with training sequence(DFE-TS) with the Modified Stop-and-Go Algorithm(MSGA) based on DFE structure, which is called a DFE-TS/MSGA. This will allow the equalizer to switch itself back to blind mode almost instantaneously when drastic and sudden changes in the channel occur. Test results based on the 8-VSB HDTV receiver show that the proposed algorithm is robust against dynamic ghosts and outperforms the conventional DFE-TS in reducing the intersymbol interference.

Digital watermarking is the techinque which embeds the invisible signal into multimedia data such as audio, video, images, for copyright protection, including owner identification and copy control information. This paper proposes a new watermark embedding and extraction technique by extending the direct sequence spread spectrum technique. The proposed technique approximates the frequency component of pixels in spatial domain by using Laplacian mask and adaptively embeds the watermark considering the HVS to reduce the degradation of Image. In watermark extraction process, the proposed technique strengthens the high frequency components of image and extracts the watermark by demodulation. All this processes are performed in spatial domain to reduce the processing time.

Generally, in the best-known Pulse-type type and Frequency Modulated Continuous Wave(FM/CW)-type RF range measurement systems, the nearest measurable distance between antenna of system and targets is limited to several tens of meters. Moreover, in case of high-speed targets it is more difficult to measure the distance precisely. In this paper we design our own RF(X-band) range(up to 15 meters) measurement system usi code correlation for high- speed targets. It 0 the correlation value between transmitting receiving PN (Pseudo -Noise) codes. So we determine the distance between antenna of s and targets using this correlation value. We fabricated it using MIC techniques and experimental results show that the proposed syste fully qualified for a short range measurement syste

원자력발전소의 증기발생기는 기능상으로도 중요한 역할을 하지만 증기발생기 내의 전열관 손상이나 누설은 바로 방사능 누출의 원인이 되므로 원자력 발전소의 안전성 측면에서 보면 매우 중요한 설비이다. 따라서, 증기발생기 전열관은 정기보수시 마다 전열관 각각에 대해 비파괴검사의 일종인 와전류검사(Eddy Current Testing, ECT)를 통해 건전성을 평가하고 있다. 현재는 신호를 취득하고 평가하는 업무가 별개의 기관에서 수행하고 있으며 이를 위해 현장에 관련 인력과 장비가 이동되어야 한다. 그러나, 취득한 신호를 평가자가 on-line으로 받을 수 있고 그 결과를 이용하여 즉시 피드백이 가능하다면 평가기관의 인력은 현장에 있지 않더라도 평가가 가능하다. 그러므로, 원격 평가 환경을 구축한다는 것은 평가자가 굳이 현장에 가지 않더라도 지역에 상관없이 LAN을 활용하여 전국 어디서든 평가가 가능하도록 하고자 함에 있다.

The operating characteristics of a 3 kW grid-connected photovoltaic (PV) power system was studied by analying annual photovoltaic data. The system performance for grid connection was investigated using a DC/ AC inverter. The results of a demonstration test show that the system utilization rate is 15.6% and the system efficiency is 8.03%.

In this research, the sensor and control system for self detecting masticatory muscle relaxation appliances. A strain gauge is used to measure a strength of tooth clenching force. A bridge circuit and voltage amplifier is designed to amplify measured signals and RF transmitter and receiver is also designed to communicate inner and outer mouth device. The experiments are performed to show the effectiveness of designed system.

This paper presents a method for designing a multiresolution orthogonal wavelet transform matrix and it is extended to the establishment of the wavelet transform domain sign algorithms(SA). It outperforms the conventional sign algorithm, with performance comparable to the LMS algorithm. Together with Daubechies type 1 wavelet, we could also save additional computations which are required in transforming data.

For increasing the quality of SCADA system, the application of image information is demanded from typical SCADA system based on data information. Until now, the remote control and monitoring systems combine data information but now gradually have trends to combine voice, data and image information. So multimedia technology is applied to realize new SCADA system. Depends on these requires, we apply new concepts of communication network and multimedia technology to the SCADA system for electric facility. In this paper, we aim at developing multi media supervisory control and data acquisition system based on data and image information, which used electric power system and security system.

For the exact classification of the arm motion this paper proposes EMG pattern recognition method with neural network. For this autoregressive coefficient, linear cepstrum coefficient, and adaptive cepstrum coefficient are selected for the feature parameter of EMG signal, and they are extracted from time series EMG signal. For the function recognition of the feature parameter a radial basis function network, a field of neural network is designed. For the improvement of recognition rate, a number of radial basis function network are combined in parallel, comparing with a backpropagation neural network an existing method.

With the recent trend to get more into the information age we suppose that the need of information for medical industry policy establishment and new product development of medical instrument will increase rapidly. The object of this research & development is the construction of a database system with information resources in the field of medical instrument for an efficient application of domestic and abroad information in the field of medical instrument and building the foundation for the construction of an unified information management system. In this research we finished the analysis of the collected material of informations for medical instrument industry, design of the database system with object-oriented modelling is done. and we developed pre-product level information management system running on WINDOWS.

In this paper, we interconnected the wavelet transformation and ADPCM to improve the image compression ratio and reduce the loss of information in the reconstructed image. Simulation shows the promising results.

To drive motor or heating machine, it needs the electric power system like the apparatus of inverter. This electric power system obviously comprises power switching devices and drivers to run them. And this system has the topology comprised one/many arm(s), - each arm has high side switching device and low side switching device. Transformer, photocoupler, and HVIC having functions of isolation and level shift which are important thing to drive high side switching device are used as component of drivers in conventional apparatus. Piezoelectric transformers are proposed in this paper, and applied to drive high side swiching device. Through experiments, the possiblities of driving high side switching device are presented and the problems are mooted concurrently. But, we also consider a counterplan for solving the mooted trouble issues.

The container crane is more expensive than other cranes, and is designed for efficient use. Since there are more control variables than other crane has more fault factors. Therefore it is important to reduce these factors efficiently and also to increase the capacity to move more freight in due time. To solve the above problems the velocity relation of the motor will be examined. In this paper the velocity relation of the motor will be expressed in a mathematical equations for efficient control.

A switch is a component with two or more ports that selectively transmits, redirects, or blocks optical power in a fiber transmission line. Our switch uses rotation mechanism using stepping motor, hence the common optical fiber can scan and allign to one of the arrayed N optical fibers to provide optical path by electronic precise control. The developed switch is consisted of switching module and its control module. The performance parameters of a switching loss and a repeatability are considered very important. We performed the study to reduce the switching loss and improve the repeatability of switch. The switch can be widely used as a test instrument of optical device and of optical cable in factory, also of optical cable monitoring systems.

We examine the performance of high-speed non-directed infrared links using pulse-position modulation (PPM) with trellis-coded modulation (TCM) to mitigate the effects of multipath-induced intersymbol interference(ISI) on such links. Using the measured characteristics of multipath indoor infrared channels, we calculate BER curves and ISI power penalties for maximum-likelihood sequence detection of 8-PPM links using rate-2/3 trellis codes operation at 10 and 30Mb/s. Our results indicate that the suggested decoder of PPM with TCM is a very promising decoder on high-speed non-directed infrared channels.

In this paper, we proposed the method to improve the performance of Viterbi decoder by applying Concatenated structure. Proposed decoder for Concatenated Code is designed with inner Viterbi decoder, block deinterleaver and outer Viterbi decoder. Inner Viterbi decoder (K=7, R=1/2) has 8-level soft decision, but outer decoder (K=7, R= 1/2) has 2-level hard decision. Applied interleaving scheme make decoder to have better BER performance in Concatenated code. The designed VLSI shares inner decoder with outer decoder. Because of sharing structure, complexity of decoder can be reduced to half. But it required about twice clock speed.

In this paper, we implemented RS232C serial communication by the 2 wires(Data,GND), and multiple access, by the applying Collision Sensing Multiple Access(CSMA) Protocol. Multiple access is implemented by assigning a unique ID to each controller. The multiple access control operation starts by sending a command packet from a host to another host and the command packet is composed of ID bytes of source and target host computer, data bytes and the check sum, byte. In host computer, after sending command packet, the collision from loop back data. If collision is detected, it means a command packet was collided with another command packet for another host. The packet communication of the controller enables the multiple acces of the controller through the common serial data link. The application of this serial communication through CSMA protocol and the usage of two wires, have an effect on saving the wires and convenient of layout work.

The masticatory muscle relaxation appliance is developed to improve the malhabit of tooth clenching of the wearer. The repeated clenching of the tooth causes excessive attrition of tooth set. The intraoral appliance measures the tooth clenching pressure data and transmits the data to masticatory muscle relaxation appliance. The appliance compares the pressure data with the reference value and generates the warning signal. The relaxation appliance also stores the clenching pressure data for clenching habit analysis. The appliance is designed with a microprocessor, real time clock, nonvolatile read write memory and dual serial communication ports.

In this paper, the implementation of dual rate ADPCM using G.723 16Kbps and 40Kbps speech codec algorithm is handled. For small signals, the low rate 16Kbps coding algorithm shows the same SNR as the high rate 40Kbps coding algorithm, while the low rate 16Kbps coding algorithm shows the lower SNR than the high rate 40Kbps coding algorithm for large signal. To obtain the good trade-off between the data rate and synthesized speech quality, we applied low rate 16Kbps for the small signal and high rate 40Kbps for the large signal. Various threshold values determining the rate are tested for good trade off data rate and speech quality. Also the low pass filter effect of speech input and output devices is simulated at several cut-off frequencies. To simulation result shows the good speech quality at a low rate comparing with 16Kbps & 40Kbps.

Change in lantency of otoacoustic emission(OAE) may indicate clinically and diagnostically important change in the status of the nervous system. A low signal-to-noise ratio of OAE signal makes it difficult to estimate small, transient, time-varing changes in latency. we present an adaptive algorithm that estimates small latency change value even when OAE signal amplitudes are time-varing.

In this paper, Prameters esimated t using AR model in order to approach linearly the heart sound which include the nonlinear characteristic from the characteristics based on a statistical theory. The parameters which is figured out using AR model is a very important information which show the characteristic heart sound In this paper parameters estimated using autocorrelation method and order selected by proposed Akaike[6] method. Compared the similirities of the spectrums between estimated by using AR model and estimated by using FFT method.

The study of 3D image reconstruction re has developed along the progress of computer. Therfore Great deal of research on it have been done. 3D medical image reconstruction techniques are useful to figure out human's complex 3D structures from the set of 2D section. But 3D medical image reconstruction require a lot of calculation, it takes long time and expensive system. That gives a reason to the improvement of study on speed. In this paper. applying the interpolation to only the part which can appear as cube, I come up with a method that calculates the speed by reducing the a mount of calculation.

Recently the optical fiber lines have been widely deployed in the trunk lines and digital data service network. The unexpected failures of optical fiber lines by natural disasters and road construction work have increased rear by rear and the number of damaged telecommunication lines per failures is on increasing. In the meantime, it takes long time to recover the optical fiber lines failures. So, it is important to maintain and monitor the status of optical fiber lines for prevention of optical fiber lines failures. Therefore, KT(Korea Telecom) needs efficient and automatic maintenance system to operate and monitor the optical fiber lines. FLOMS(Fiber Line Operation and Monitor System) had been developed in '96 and pilot system was installed at telephone offices in '97. This paper describes the FLOMS which carries out the functions of efficient maintenance and supports for monitoring optical fiber lines and managing facilities in optical fiber lines.

Fiber optic facilities are increasingly being deployed in loop applications for both busines and residential areas. These facilities support a variety of communications services that include high speed data and video using leading technologies such as Synchronous Optical Network (SONET), Wavelength Division Multiplexing(WDM). This proliferation of fiber optic facilities combined with the increasing pressure on network operators to reduce costs are fostering increased automation to reduce labor costs associated with providing and maintaining communications services. Effective fiber management and accurate records can improve the reliability and integrity of the future telecommunications networks and the quality of customer services. This paper describes the trend forward and the need for the deployment of Fiber Administration System (FAS) into the operations enviroment of a typical network provider.

In this paper, the silicon-nitride membrane structure for IR sensor was fabricated through the etching and the direct bonding. The PTO layer as a IR detection layer was deposited on the membrane and its characteristics were measured. The attack of PTO layer during the etching of silicon wafer as well as the thermal isolation of the IR detection layer can be solved through the method of bonding/etching of silicon wafer. Because the PTO layer of c-axial orientation raised thermal polarization without polling, the more integration capability can be achieved. The surface roughness of the membrane was measured by AFM, the micro voids and the non-contacted area were inspected by IR detector, and the bonding interface was observed by SEM. The polarization characteristics and the dielectric characteristics of the PTO layer were measured, too.

In this paper, we propose the new detecting method for three dimensional piezoresistive silicon accelerometer. Furthermore the accelerometer is formed to have endurance for high temperature by perfect isolation of the piezoresistors using Silicon On Insulator(SOI) wafer. Sensor size are optimized with analytical formulae and extended with FEM simulation for the more detailed results. The accelerometer was fabricated by bulk micromachining techonology. We measured the temperature characteristics and the output characteristics, and the both characteristics were compared with the simulated results

In this study, we formed the piezoelectric film and estimated its characteristics for sensor application. The $Pb(Zr,Ti)O_3(PZT)$ was chosen as piezoelectric material and we used Sol-Gel method to form film. To increase film thickness, the multiple coatings were performed, and the good characteristics obtained in thick film compared to thin film. Because PZT film showed fine etching property as well as other good characteristics, it was thought that it was appropriate material for sensor fabrication.

The bending actuator using zigzag type shape memory alloy springs has been fabricated and characterized. The small sized actuator with outer diameter of 3.0mm and inner diameter of 2.0mm could be implemented because zigzag type spring has advantages for thin wall type actuator over the coil type spring. The measured characteristics of the fabricated bending actuator show the possibility of practical application to micro active bending catheter.

The micro fresnel lens(MFL) was modeled and fabricated on a XY-stage electrostatically driven by comb actuator. The modeled MFL was approximated as a step shape with 4-phase and 4-zone plate. The focal length and diameter of the MFL is 20mm and 912${\mu}m$, respectively. The XY-stage suspending the MFL is designed to generate a large static displacement up to about 20${\mu}m$. On SOI substrates, we first fabricated MFL using the RIE(reactive Ion etching) technology and then patterned and etched bulk silicon to make XY-stage. After the fabrication of all structures on top side of the SOI substrates. $Si_3N_4$ was deposited for passivation of all structures using PECVD(plasma enhanced chemical vapor deposition). All the MFL systems width comb drive actuator were released by KOH etching from the bottom side of the SOI wafer using double-sided alignment technique. In fabrication of MFL, a dry etching conditions is established in order to improve surface roughness and to control the etched depth.

Silicon can be reactive ion etched (RIE) either isotropically or anisotropically. In this paper, a new micromachining technology combining these two etching characteristics is proposed. In the proposed method, the fabrication steps are as follows. First. a polysilicon layer, which is used as the bottom electrode, is deposited on the silicon wafer and patterned. Then the silicon substrate is etched anisotropically to a few micrometer depth that forms a cavity. Then an PECVD oxide layer is deposited to passivate the cavity side walls. The oxide layers at the top and bottom faces are removed while the passivation layers of the side walls are left. Then the substrate is etched again but in an isotropic etch condition to form a round trench with a larger radius than the anisotropic cavity. Then a sacrificial PECVD oxide layer is deposited and patterned. Then a polysilicon structural layer is deposited and patterned. This polysilicon layer forms a pivot structure of a rocker-arm. Finally, oxide sacrificial layers are etched away. This new micromachining technology is quite simpler than conventional method to fabricate joint structures, and the devices that are fabricated using this technology do not require a flexing structure for motion.

Freestanding flexible microstructures fabricated from deposited thin films become mechanically unstable when internal stresses exceed critical values. The residual stress and stress gradient of aluminum thin film were examined to make sure of fabricating the reproduceable aluminium structure. For good shape of micro mirror array and microstructures, the experiment was done varying thickness and deposition rate. As the aluminium film thickness increased from 0.8${\mu}m$ to 1.6${\mu}m$, the stress gradient decreased from 11.62MPa/${\mu}m$ to 2.62MPa/${\mu}m$. The residual stress values are from 42.4MPa to 62.24MPa of tensile stresses.

In this paper, the residual stress and young's modulus of the p+ thin film have been estimated by using the electrostatic resonators. The electrostatic plate resonator with four corrugated bridges and another with four flat ones have been fabricated. The deflection of the plate has been calculated under the induced tension and the residual stress and compared with the dynamic test results. When the young's modulus of the p+ silicon is 125 GPa. The estimated residual stresses of the flat and the corrugated bridges are about 15 MPa and less than 5 MPa, respectively. It has been confirmed that the corrugated structure releases the residual tensile stress resulted from the heavy boron diffusion process.

Measurement equipment, which consists of He-Ne laser, objective lens, CCD camera and position sensitive detector, was developed for the dynamic response of micromirror. This equipment can measure the tilting direction and angle, the rise time and frequency response of micromirror to the input signal. It was found that the rise time of the micromirror was less than 10${\mu}m$ and the resonant frequency was about 25kHz.

Micro gold electroplating structure is fabricated by surface micromachining process. The made structure is clamped-clamped beam and its 1st natural frequency is measured by resonance method. In order to find residual stress, first natural frequency of beam which has tensile stress inside is analysed using Rayleigh's method. Elastic modulus and residual stress are estimated from the measured natural frequency.

In this paper, an effect of drive-in process temperature on the residual stress profile of the p+ silicon film has been investigated. The residual stress profile has been calculated as the fourth-order polynomials. All coefficients of the polynomials have been determined from the measurement of the vertical deflections of the p+ silicon cantilevers with various thickness and the tip displacement of the p+ silicon rotating beam. From the determination results of the residual stress profile, the average stress of the film thermally oxidized at 1000 $^{\circ}C$ is 15 MPa and that of the film oxidized at 1100 $^{\circ}C$ is 25 MPa. The profile of the residual stress through the high temperature drive-in process has a steeper gradient than the other case.

This paper investigates the residual stress of tetraethoxysilane (TEOS) and 7wt% phosphosilicate glass (PSG), which are commonly used as a sacrificial layer or etch mask in the fabrication of microelectromechanical systems (MEMS). In order to measure residual stress, $2{\mu}m$ thick TEOS and PSG stress measurement structures are fabricated. Polysilicon is used as the sacrificial layer. First the residual stress of an as-deposited 7wt% PSG flim and TEOS film are measured to be-0.3115% and -0.435%, respectively, which are quite large. These films are annealed from $500^{\circ}C$ to $800^{\circ}C$. Annealing has the effects of reducing residual stress. In the case of the 7wt% PSG film, the residual stress becomes +0.00715% after annealing at $625^{\circ}C$ for 150 minutes. In the case of TEOS film, the residual stress reduces to -0.2134% after same condition. Incidentally, this condition is the same condition for depositing a $2{\mu}m$ thick polysilicon at $625^{\circ}C$ at our low pressure chemical vapor deposition (LPCVD) furnace.

The characteristics of the electroplated nickel were investigated for the MEMS applications. Nickel sulfate bath was used and saccharine was added as the brightener. The effects of the brightener concentration were investigated by comparison of the surface morphology and the hardness of the electroplated nickel. The polarization characteristics were also investigated. The best results were obtained at 1g/$\ell$ saccharin addition.

This paper describes the fabrication process to fabricate metallic structure of high aspect ratio using LlGA-like process. SU-8 is used as an electroplating mold. SU-8 is an epoxy-based photoresist, designed for ultrathick PR structure with single layer coating [1,2]. We can get more than $100{\mu}m$ thick layer by single coating with conventional spin coater, and applying multiple coating can make thicker layers. In the experiments, we used different kinds of SU-8, having different viscosity. To optimize the conditions for mold fabrication process, experiments are performed varying spinning time and speed, soft-bake, develop and PEB (Post Expose Bake) condition. With the optimized condition, minimum line and space of $3{\mu}m$ pattern with a thickness of $40{\mu}m$ and $4{\mu}m$ pattern with a thickness of $130{\mu}m$ were obtained. Using the patterned PR as a plating mold, metallic structure was fabricated by electroplating. We have fabricated a electroplated nickel comb actuator using SU-8 as plating mold. The thickness of PR mold is $45{\mu}m$ and that of plated nickel is$40{\mu}m$. Minimum line of the mold is $5{\mu}m$. Patterned metallic layer or polymer layer, which has selectivity with the structural plated metallic layer, can be used as sacrificial layer for fabrication of free-standing structure.

The electrical and physical charateristics of Pt-Co alloy thin films on $Al_2O_3$ substrate, deposited by r.f cosputtering respectively, were analyzed with thickness of thin films ($1700{\sim}10000{\AA}$) and increasing annealing temperature ($800{\sim}1000^{\circ}C$). At input power of Pt : 4.4 W/$cm^2$, Co : 6.91 W/$cm^2$, working vacuum of 10 mTorr and annealing conditions of $1000^{\circ}C$) and 60 min, the resistivity and sheet resistivity of Pt-Co thin films with thickness of $3000{\AA}$ was $15{\mu}{\Omega}{\cdot}cm$ and 0.5 ${\Omega}/{\square}$, respectively. The TCR value of Pt-Co alloy thin films was measured with various thickness of thin films and annealing conditions. The optimum TCR value of 3850 ppm/$^{\circ}C$ in temperature range($200{\sim}400^{\circ}C$) is gained under conditions $3000{\AA}$ of thin films thickness and $1000^{\circ}C$ of annealing temperature. The thermal charateristics of Pt-Co micro heaters were analysed with Pt-Co RTD integrated on the same substrate. In the analysis of characteristics of Pt-Co micro heaters, the Pt-Co micro heaters with thickness of $3000{\AA}$ and annealing temperature of $1000^{\circ}C$ had a good linearity and temperature is up to $468.2^{\circ}C$ with 2.1 watts of the heating power.

Automatic yield-test equipment for micro mirror array using image processing was developed. This computerized test equipment can classify the error states of the micromirrors. The test results are displayed on the monitor as a map which shows the error states and position. It is possible to measure yield and reliability with this test equipment for micromirror array using image processing.

The stiction phenomena poses a design constraint in surface micromachining by reducing the releasable size of the microstructure. This problem occurs during the fabrication process of surface micromachined microstructures during the wet etch of sacrificial layers. For the prevention of the sticking problem, the microsctructure is released by sublimation after the substitution of the sacrificial layer etchant with a sublimation material heated above its melting temperature. In the sublimation drying method, the sublimation materials such as p-dichlorobenzene, t-butyl alcohol, and cyclohexane are used. In this paper, a method for designing a sublimation drying system is developed, and its performance is experimentally evaluated.

A single crystal Mn-Zn ferrite was directly etched by focused $Ar^+$ laser beam in $CCl_2F_2$ gas atmosphere. AES has been performed for locally investigating the surface composition of an etched layer. MnCl, ZnCl being created after the substrate and $CCl_2F_2$ chemically reacting was remained in the vicinity of laser irradiation area because of their low vapor pressure. Various patterns using computer were formed on the substrate. The etched grooves and patterned shapes were observed by SEM measurement.

2-D microwell arrays for micro ELISA (Enzyme-Linked Immuno Solvent Assay) system were fabricated using micromachining technology. The materials for the bottom plate, top plate and sidewall of the microwell were used a #7740 glass, gold and silicon respectively considering bio-compatibility and easy fabrication. Cylindrical or groove shape microwells, which have about $100{\mu}m$ depth and $50{\sim}500{\mu}m$ diameter or width, were arrayed. The fabricated microwell array can be applied to the essential part of a biochip when surface modification is made to immobilize cells or biomolecules on the microwell bottom.