제어로봇시스템학회:학술대회논문집
Institute of Control, Robotics and Systems (ICROS)
 기타
Domain
 Machinery ＞ Robot/Automated Machinery
1992.10

An autonomous mobile robot has been developed which can follow a travel map drawn on a monitor screen using a software. The robot works as an exhbition guide making announcement regrading the events to take place or introducing the products on display. This mobile robot computes the world position and heading direction through camera image, ultrasonic and infrared sensors in real time using specially designed algorithm. This mobile robot will be exhibited at '93 TaeJeon EXPO for an exhibition guide, and will be used in a plant or a hospital for materials handling purposes.

This paper proposes a sensorbased path planning method which utilizes fuzzy logic and neural network for obstacle avoidance of a mobile robot in uncertain environments. In order to acquire the information about the environment around the mobile robot, the ultrasonic sensors mounted on the front of mobile robot are used. The neural network, whose inputs are preprocessed by ultrasonic sensor readings, informs the mobile robot of the situation of environment in which mobile robot is at the present instant. Then, according to the situation class, the fuzzy rules are fired to make a decision on the mobile robot action. In addition, this method can be implemented real time since the number of fuzzy rules used to avoid the obstacle is small. Fuzzy rules are constructed based on the human reasoning and tuned by iterative simulations. The effective of the proposed avoidance method is verified by a series of simulations.

In this study, we developed an autonomous AGV which carries materials between workshops. In most of existing AGV systems, when AGV is required to change its navigating path, the guideline or landmark of AGV should be rebuilt according to new navigating path. Using sensors and internal coordinate system, our AGV, however, can navigate along the new path by only changing input parameters of program. On navigating, if AGV meets obstacles, it avoids them and go on to the destination.

Realtime mobile robot controllers usually have been designed with an emphasis on control theory ignoring the importance of system integration. This paper demonstrates that useful mobile robots require a real time controller with a wide range of capabilities in addition to control theory. These capabilities include: pathplanning, position estimation, path tracking control and wheel control. An architectural framework supporting these capabilities has been designed and implemented. Using this frame work, individual modules such as a path planner, a path tracking controller, position estimators, wheel controllers and other cruical elements have been successfully integrated into the control system for the LCAR robot which was developed as a prototype mobile robot in our laboratory. The context of the research, the architecture, its implementation and performance results from experiments are discussed.

In this paper, the problem of trajectory generation for mobile robots is investigated. The robot trajectory is generated so that smooth turns are guaranteed. Also the kinematic constraints of the actual robot are considered and incorporated in the trajectory generation.

This paper describes the development of a 'Generalpurpose ETRI MAP interface module' (GEM) for a MiniMAP network. GEM operates as a MiniMAP node in our FA system. To communicate between GEM and programmable devices such as PLC, CNC, and robot, RS232C is used, which is a traditional method. A MiniMAP host system controls and monitors programmable devices via GEM. The MiniMAP function of GEM is implemented and tested on the basis of the MAP V3.0.

The main purpose of the paper is to derive asymptotic formulae for performance characteristics(throughput, delay) of largescale tokenpassing network with buffered stations and to optimize the buffer capacity with respect to the probability of data generation. We consider two versions of tokenpassing network: uniform and nonuniform tokenpassing time interval. All results obtained are supported by simulations.

The automation system oriented as one of the second year automation projects for the small and medium sized enterprises(SME) was developed for the improvement of the production rate and cut the required manpower in the field of the relay which is one of the small electric components used in various industrial fields. The objectives of this study are not only improving the international competition of the relay itself but also partially solving the technical and financial problems featured by common bottlenecks of the SME for efficient assembly automation. For the purpose of these objectives, several topics are studied as followings.  Analyzing the adjustment process and determining the specification of the automation system.  Determining the layout for the automation system to meet the determined specification.  Detail design of the automation system for relay adjustment and inspection.  Control system design  Automation system development and performance test.

A flexible robotic assembly cell is modeled using Petri net. A simulator is developed and used to predict the optimal status of the system. The assembly cell of flexible manufacturing system(FMS) pilot plant of Automation and Systems Research Institute(ASRI) in Seoul National University is modeled. The system consists of 3 robots, 4 conveyors, automatic guided vehicle(AGV) and autostacker. The simulator is programmed in Turbo C on IBM PC, supporting a simple graphic simulation with pulldown menu. The flexibility of the assembly cell in the FMS plant is guaranteed, since it is possible to predict the optimal status of the system using this simulator.

The IEEE 802.4 priority mechanism can be used to handle multiple data access classes of traffic. Several timers are used to realize the priority mechanism. The performance and stability of a token bus network depend on the assignment of such timers. In this peper, we present a dynamic timer assignment algorithm for the token passing bus network. The presented algorithm has simple structure for realtime applications and adaptively controls the set of initial timer values according to the offered traffic load. The assignment of the set of timers becomes easy due to the presented algorithm. Based on the iterative algorithm, some solutions such as mean waiting time are derived.

In this paper, a method for implementing software for the control system with dual structure in processor module is proposed and implemented to enhance its reliability. In this implementation the multitasking function which is provided by a realtime operating system is applied. The overall softwre is divided into five tasks and is performed in each of the dual processor module, independently. By this, the processor module with dual structure can achieve a control objective and fault diagnostics effectively. An experimental result shows that the backup processor module can be substituted for the primary processor module immediately when it happens to fail, because data relating the failure information are exchanged continuously done via shared memories.

1930년 스웨덴에서 개발에 성공, 네덜란드에서 더욱 발전시킨 ALC(Autoclaved Lightweignt Concrete의 약칭) 는 가볍고, 견고하고, 그리고 시공이 간편한 경제적인 요건들을 충족시키는 건축자재로 세계적으로 널리 사용되고 있으며 , 국내에서는 불과 수년 전부터 연구 개발되고 있는 실정이다. ALC 란 시멘트와 규사, 생석회등 무기질 원료를 고온,고압으로 증기 양생시킨 경량의 기포 콘크리트 제품을 통칭한 것이다. ALC공정은 bulk ALC를 생산하는 batch공정과 이 bulk ALC에 대한 처리 공정으로 크게 나눌 수가 있으며 여기에서는 bulk ALC 처리 공정을 side shield treatment, anticorrosion treatment, curing grate transferer, cutting station, curing car transportation, autoclave traveling platform, 및 packing 의 공정으로 세분하여 각 공정개요 소개 및 PLC(Programmable Logic Controller의 약칭)를 이용한 제어 system에 대하여 설명하고자 한다.

This paper describes the automation of the surface grinder for a grinding wheel by application of programmable logic controller(PLC) with a position sensor, and a limit switch. The control system is designed to provide feeding a workpiece on the turn table and pressing it by the upper disk automatically. In this development the automation of checking the thickness of a grinding wheel is most important. In order to measure the relative displacement, the proximity sensors were employed and the sensitivities of the sensors were investicated and discussed the superiority. It has been shown by model experiment that the automation system of the surface grinder is performed satisfactory.

This paper deals with the design scheme of the custom control circuit for optimizing the interface of peripheral devices to generalpurpose controllers for the high speed digital system. When the various peripheral devices such as memory, I/O devices and buffers which operate at low speed are interfaced to the microprocessor which operates at high speed, inserting the proper wait state to the processor is required. The proposed scheme designed with random logic may be applied to the high performance graphic system like the Xterminal. This circuit provides the flexibility and system independancy for the optimum digital system design.

We design a multielement photodetector to measure the size of particles using the diffracted light energy distribution. The scattered profile measured by the photodetector is sampled by a 32 channel analogtodigital converter. A nonnegative least squares analysis translates the light energy distribution into the corresponding unique particle size distribution. The responses of the particle sizing system are studied theoretically and experimentally.

One of the most difficult problems in depth control for underwater vehicle is the effect of seaway disturbance. One component of the seaway forces is of large magnitude with a relatively narrowband, first order component. The other component is generaly of somewhat smaller magnitude, second order component. Since the magnitude of the first order component is generally much greater than the compensating force that can be generating by the planes, it is undesirable for the controller to generate a control command. In this paper, we disigned adaptive notch filtering system using filter bank structure. Energies of each bandpassed signal are obtained by MA(Moving Average) method and compared to produce center frequency. By adapting this parameter to notch filter, 1st order seaway disturbance can be removed, which lead to the improvement of automatic depth control system.

A portable multichannel functional electrical stimulation(FES) system for the fine control of the paralyzed extremities in spinal cord injury patients is described. This system is composed of a stimulation data creating system, a serial communication device, a 16bit microprocessor, D/A converter of 32 channels and a display device. Stimulation patterns are created from analytical results of integrated EMGs during motion in normal subjects and are stored in the stimulation data creating system as data files. And then the stimulation patterns are sent to the memory in the portable multichannel FES system through serial communication interfacing device. Sophisticated fine control of paralyzed extrimities was realized by transmitting multichannel stimulation patterns to percutaneous intramuscular electrodes, which stimulate the motor function of paralyzed muscle simultaneously. Advantages of this system are as follws: 1) It is possible to modify stimulation patterns in accordance with the patient's situation. 2) This system is small and light.

본 논문은 삼차원 거리영상화 장비중 고속으로 거리영상을 잡을 수 있는 거리영상화 소자를 IC로 시험 제작하여 그 결과를 보였다. 제작된 IC에 대한 실험은 각 부분의 성능과 센서 자체의 정밀도 측정에 중점을 두었다. 각 부분의 성능에 대해서는 빛 감지부분의 신호의 크기, 소자들의 동작성, 정보 저장부분의 입력 전압에 대한 출력 전압의 변형등을 다룬다. 센서 자체의 정밀도는 각 부분에서 발생하는 오차를 측정하였다. 제작된 센서는 아나로그방식으로 정밀도에 제한을 받는 단점이 있어 구조를 달리하여 디지탈 방식으로 접근한 센서를 제안한다. 아나로그 방식에 비해 셀 면적을 줄일 수 있을 뿐 아니라, 정밀도도 높고, 속도도 떨어지지 않아 아나로그 방식 IC 센서의 단점을 극복한다.

Texture is one of the important characteristics that has been used to identify objects or regions of interest in an image. Tire tread patterns can be considered as a kind of texture, and these are classified with a texture analysis method. In this sense, this paper proposes a new algorithm for the classification of tire tread pattern. For the classification, cooccurrence matrix for the binary image is used. The performances are tested by experimentally 8 different tire tread pattern and the robustness is examined by including some kinds on noise.

The flexible arm has considerable structural flexibility. Because of its flexibility, the dynamic nodel is very complex and difficult to get. In this paper, fuzzy logic controller(FLC) of the singlelink flexible arm is proposed, for FLC does not require any mathematical model of the plant. Noncolocated control is used and the choice of linguistic variables are examined. The simulation results are presented to show the possibility of FLC for flexible arm.

In this paper, a fuzzy logic controller(FLC) is designed for the pointing loop of the spinstabilized platform. For the fuzzy inference, a fuzzy accelerator board using the Togai InfraLogic software and digital fuzzy processor(DFP110FC) is designed, and a validation of an algorithm for fuzzy logic control is also presented. The pointing loop of the spinstabilized platform using FLC has better performance of step responses than a proportional controller in case of same loop hain through the software simulation and the experiment of implemented hardware.

The fuzzyneural hybrid control system utilizing the fuzzyneural network(FNN) will be presented in this paper. The basic structure of the controller is the parallel combination of a conventional Pcontroller and a FNN. Such a combination can guarantee the stability of a plant at initial stage before the rules are completely created. And a method how to automatically tunning the parameters of the FNN will be proposed with error backpropagation(BP) algorithm. Finally the effectiveness of the proposed strategy will be verified by computer simulations using a two DOF robot manipulator.

This paper presents an autotuning algorithm of scaling factor in order to improve system performance. We define the scaling factor of fuzzy controller as a function of error and error change. This function is tuned by the output of performance evaluation level utilizing the error of overshoot and rising time. Simulation results show that the proposed algorithm has good tuning performance for a system with parameter change.

Fuzzy theory has shown good control performance for nonlinear system that is difficult to be controlled by the conventional controller. Backpropagation neural network can interpolate output without the priori knowledge of its dynamics. In this paper, we proposes a FuzzyNeural Controller. The Fuzzy Control by deterministic rule may not be sensitive for uncertain conditions and has a disadvantage of setting the rule by repeatedly experience. To solve such problems, we construct Self organizing FuzzyNeural Controller which can reorganize the fuzzy rule according to the state of system. Experimental results show that proposed FuzzyNeural Controller has better performance than conventional controller(PID) has especially rising time and overshoot characteristics.

This paper presents the configuration, HILS procedure and performance simulation results of the RPV autopilot including a strapdown AHRS. Real time hardwareintheloop simulation was performed by using a 3 axis flight motion simulator alonged assumed flight trajectory of the RPV. Being compared with the result of the 6 DOF simulation, the HILS results showed that the performance of the autopilot was satisfactory.

Joint probabilistic data association filter(JPDAF) for multitarget tracking was developed for realtime implementation, while it abandoned an algorithm for track initiation. In this paper, we propose three features for track initiation that can be adapted to the JPDA filter. In addition, with the proposed approaches, the performance of track maintenance is evaluated in the case of tracks being near. To eliminate the abundant false tracks, we exploit the simple method using the state error covariances. Simulations are performed to demonstrate the efficiency of the proposed approaches.

This paper describes a new approach that can extract the relative shift vectors between two aerial image sequences for implementing the visual navigation system. This method minimizes the noise included in the aircraft attitude information that represents the changes of the aircraft attitude using the statistical method and Kalman Filtering method. This result can be used to find the relative shift vectors which are independent of the attitude changes indicating the true trajectories of aircraft. We applied this method to the image about which we had already known the information of aircraft attitude, and that result showed that the errors were minimized successfully.

The purpose of this paper is to calculate the dynamic derivatives of single rotor Helicopter in forward flight. From trim condition, the equation of motion is derived, and we can calculate the dynamic dervatives. The results were compared with flight test data. The phase angle and stick displacement are obtained and compared at the trim condition.

In this paper, the complex nonlinear dynamics of a satellite is obtained. And it is shown that several limitations exist when the magnetorquer is used as an active actuator to attitude control. Such limitations cause a delayed convergence of pitch and roll angle. The simulation results insure that the roll angle bias is dependent on the z axis spin rate. And a heuristic algorithm is applied to control the attitude libration through the computer simulations.

In this paper, the HardwareInTheLoop Simulation(HILS) of missile systems are studied. The HILS is an effective performance evaluation technique that bridges the simulation fidelity gap between analytic alldigital simulations and actual flight tests of missile systems. The HILS may be required to perform system integration tests, performance evaluation at system or subsystem level. Major elements of this HILS facility will include the flight table, simulation computers, I/O computer and peripheral equipments. HILS of missile systems typically involve computer modeling of flight dynamics coupled with a hardware guidance and control(G/C) systems. This paper describes a real time performance evaluation technique of a G/C system, Development of a HILS for a Autopilot of SAM G/C will be used as an example.

In this paper, we derive the optimal inputconstrained control law which minimizes predictive control objective function subject to input constraints. In order to obtain the closed form of control law, three suboptimal methods are proposed and evaluated by simulation.

In this paper, an adaptive generalized predictive control(GPC) algorithm which minimizes a Nstage cost function is proposed. The resulting controller is based on GPC algorithm and can be used in unknown plant parameters as the parameters of one step ahead predictor are estimated by recursive least squares method. The estimated parameters are extended to G,P, and F amtrix which contain the parameters of N step ahead predictors. And the minimization of cost function assuming no constraints on future controls results in the projected control increment vector. Hence this adaptive GPC algorithm can be used for either unknown system or varing system parameters, and it is also shown through simulations that the algorithm is robust to the variation of system parameters. This adaptive GPC scheme is shown to have the same stability properties as the deterministic GPC, and requires small amount of calculation compared to other adaptive algorithms which minimize Nstage cost function. Especially, in case that the maximum output horizon is 1, the proposed algorithm can be applicable to direct adaptive GPC.

This paper proposes a method by which the estimated values of the unknown parameters of system are converged to the true values in finite time using adaptive deadbeat controller. After those are converged to the true values, the deviation from these values do not virtually exist or, if any, extremely small. Also we apply this technique of deadbeat convergence to a system contaminated with white noise or colored noise. It is shown that the estimated parameters of those systems approach the true values in finite time even though the performance do not match perfectly with the system without noises.

The PID controllers which is widely used in the process industry are poorly damped when the dynamic process contains significant dead time or when there are random disturbances acting on the plant. GPC is known to be more superior than conventional selftuning algorithm in overcoming above problem and prior choice of model order. In this paper, we propose the method which determine the parameter of PID controller from minimization of GPC criterion. The controller has emplicit scheme which is comprised of parameter estimation and PID control design. Simulation results show the performance of the proposed selftuning PID controller.

In this paper, each section of an Electrolytic Tinning Line is mathematically modelled with the coupling effect accounted for. Based on the modelling, PI and feedforward controllers are designed for speed control or tension control of each section of the process line. It is shown through simulatios that the tension of the stip is controlled precisely along the entire line and the contituety of the process is insured when using the efficient control of loop towers.

The classical control theory has been developed successfully for the design of a system controller and has evolved continually. Even though sophisticated simulation techniques and software packages are available, there is still some difficulty in the design of a complex system controller at the desk. So the trial and error method is sometimes used to design a new controller, but it requires excess time and cost. This paper suggests a controller design method through the experimental approach. The basic concept is to adjust gradually the design parameters of the controller to the simulation results and experimental data of a similar real system. This method will be a very useful and easy way to design an accurate and/or optimal controller for a real plant while reducing time and giving a good solution at a reasonable cost.

A mathematical model of the blowdown type wind tunnel is developed in order to design the controller which controls the stagnation pressure being used to obtain the setpoint Mach Number. The motion of compressible fluids in the tunnel is modeled using the onedimensional gasdynamics. The time responses of the wind tunnel states, such as pressures, mass flow rates, and valve open area, are investigated by digital computer simulation. By the simulation study it is shown that the real blowdown wind tunnel can be simulated by the obtained mathematical model.

In this paper, a two ports machine(TPM) model for discrete event dynamic systems(DEDS) is proposed. The proposed model is a finite state machine which has two inputs and two outputs. Inputs and outputs have two components, events and informations. TPM is different from other state machine models, since TPM has symmetric input and output. This symmetry enables the block diagram representation of the DEDS with TPM blocks, summing points, multiplying points, branch points, and connections. The graphical representation of DEDS is analogous to that of control system theory. TPM has a matrix representation of its transition and information map. This matrix representation simplifies the analysis of the DEDS.

This paper describes a learning method of neural network controllers. The learning method improves the performance of indirect learning mechanism in the neurocontrol of nonlinear systems. To precisely identify dynamic characteristics of the plant by utilizing a limited prior information we propose a new energy function which takes advantage of the proportional relationship between outputs of the plant and those of neural networks.

We investigate a neural network as a dynamic system controller when system characteristics are abruptly changing. The shape of sigmoid functions are determined by autotuing method for the optimum sigmoid function of the neural networks. By using information stored in the identifying network a novel algorithm that can adapt the control action of the controller has been developed. Robustness can be seen from its ability to adjust large variations of parameters. The potential of the proposed method is demonstrated by simulations.

We propose a method of generating data to train a neural network controller. The data can be prepared directly by an iterative learning technique which repeatedly adjusts the control input to improve the tracking quality of the desired trajectory. Instead of storing control input data in memory as in iterative learning control, the neural network stores the mapping between the control input and the desired output. We apply this concept to the trajectory control of a two link robot manipulator with a feedforward neural network controller and a feedback linear controller. Simulation results show good generalization of the neural network controller.

Multilayered neural networks are applied to the identification and control of nonlinear dynamical system. Traditional adaptive control techniques can only deal with linear systems or some special nonlinear systems. A scheme for combining multilayered neural networks with model reference network techniques has the capability to learn the nonlinearity and shows the great potential for adaptive control. In many interesting cases the system can be described by a nonlinear model in which the control input appears linearly. In this paper the identification of linear and nonlinear part are performed simultaneously. The projection algorithm and the new estimation method which uses the delta rule of neural network are compared throughout the simulation. The simulation results show that the identification and adaptive control schemes suggested are practically feasible and effective.

This paper describes a neural network based control scheme with MRAC. The system consists of two neural network; one is for identifier and the other is for controller. Identification is firstly performed to learn the behavior of the nonlinear plant. Neural net controller is next trained by backpropagating the error at the output of plant through the identifier. Also the training method used in this paper repeatedly updates weights of neural network to track the reference model.

In the development of a high speed and light weight manipulator, it is necessary to consider the structural elasticity of a robotic arm. The analysis of the infinite mode dynamic of robotic arm must be performed to obtain the finite mode modelling to achieve the feasible controller design of the robotic arm. The modelling procedure of the robotic arm is also illustrated. The controlled mode of the modelled dynamic can be derived by truncating the higher vibrational mode to result in the low order system for the sampling in the control signal is confined to the higher mode. And it is controlled by the pole assignment which can compensate the unmodelled dynamic effects. The unmodelled dynamic can result in the instability of the controlled system, which is known as spillover. The controller design of the low order system is simulated by the pole assignment and optimal control theory.

In this paper, we propose an optimal method for the tracking a trajectory of the endeffector of flexible robot arms with multiple joints. The proposed method finds joint trajectories and joint torques necessary to produce the desired endeffector motion of flexible manipulator. In inverse kinematics, optimized joint trajectories are computed from elastic equations. In inverse dynamics, joint torques are obtained from the joint equations by using the optimized joint trajectories. The equations of motion using finite element method and virtual work principle are employed. Optimal control is applied to optimize joint trajectories which are computed in inverse kinematics. The simulation of flexible planner manipulator is presented.

This paper presents an adaptive control scheme for flexible joint robot manipulators. This control scheme is based on the Lyapunov direct method with the arm energybased Lyapunov function. The proposed adaptive control scheme uses only the position and velocity feedback of link and motor shaft. The adaptive control system of flexible joint robots is asymptotically stable regardless of the joint flexibility value. Therefore, the assumption of weak joint ealsticity is not needed. Also, joint flexibility value is unknown. Simulation results are presented to show the feasibility of the proposed adaptive control scheme.

This paper presents a new stable composite control law for the flexible joint robot manipulators, which incorporate the additional stabilizing control law with sliding property. The singularly perturbated models include inertia moments functions of the deformations of actuator. The newly defined fast controller variable is computed from the corrected reducedorder model without additional computational loads. The simulations for 2 DOF flexible joint manipulator show that the proposed schemes are more stable than conventional one, and especially effective for the manipulator with high jointflexibilities.

Recently, the digital circuit technique has had great success and gained importance in all industries. For the measurement of degree of freshness of plant and animal's tissue, many devices has been developed. But it's data was inaccurate and it had difficulty in making database because it was depend on analog circuit technique. The purpose of this paper is the development of a device based on digital circuit for measurement of freshness degree of wet fish such that its accuracy is very trustworthy and it can be measured in a short time. The device is developed by using microcomputer with some interface modules of AD/DA converter and digital circuit with IC modules and its effectiveness has been evaluated through two experimental results of electrical Qvalue test and chemical Kvalue test.

This paper deals with the controller design problem having input filter. To reduce the excessive overshoot in the setpoint response, a design method with filtered input is proposed here and we show that it can be reduce the overshoot to the specified value based on simple formula. Analytic results and computer simulation show that the contorller design method suggested in this paper is superior to the conventional solution of reducing large overshoot.

This paper describes a duplexed process control system for a boilerturbine system of power plant, which is composed of supervisor, controller, and simulater subsystems. Its embeded POL(Problem oriented language) as a structured control language enables this system to do a realtime distributed control and fault diagnosis by simple programming with the actual implementation of advanced control algorithms such as PID autotuning and GPC, etc, the performance of overall system has been greatly enhanced.

Recently, electronic engine control system is used in many automotives for high efficiency and low pollution. In order to perform these requirements, fuel injection control, spark timing control, knock control, exhaust gas recirculation control and idle speed control should be implemented. In this paper, idle speed control system using microcontroller is developed, which is compact in hardware, but powerful in software performing efficient control and various compensations for engine condition and environments. If idle speed is low engine operation is not smooth, reversely if high, fuel consumption is increased. Therefore idle speed must be maintained as low as possible within the scope that ensures smooth operation of engine. Also, an engine signal simulator, which generates various signals from engine, is realized for test facility.

There are lots of disturbance in the super heater temperature control system of power plant boiler as follows. 1.Burner light off. 2.Excess Air. 3.Burner tilt. 4.G.R fan flow. Temperature control system of super heater in the power plant has delay time about 5 min. So it is difficult to control the super heater temperature in the power plant. This paper show us the application of domestic development DCS to control the super heater temperature in seoul #5 thermal power plant unit.

In this study, we propose a shared storage system which is more efficient policy than MIS(Mixed Intermediate Storage) policy for relatively rare storage system and can be called CIS(Common Intermediate Storage) policy. Using this strategy, we develop a new completion time algorithm and apply it to two kinds of optimal or near optimal scheduling method; combinatorial search and simulated annealing method. We also extend this strategy to other storage policy, for example MIS policy, with PLC(Programmable Logic Controller) logic and on/off action of electronic valves. It thus can be accepted as a basic form of FMS(Flexible Manufacturing System) for operating various storage policies. Finally we suggest the interlocking block to compansate for the shortcoming of CIS policy, i.e, complication of operation and safety, resulting in a basic batch process automation mode.

Many dynamic mechanical systems, such as partsfeeders, walking machines, and percussive power tools, are described by equations of motion which are discontinuous. The discontinuities result from kinematic constraint changes which are difficult to foresee, especially in presence of impact. A simulation algorithm for these types of systems must be able to algorithmically predict and detect the kinematic constraint changes without any prior knowledge of the system's motion. This paper presents a rulebased approach to the prediction and detection of kinematic constraint changes between bodies with arc and line boundaries. The developed algorithm's ability to accurately and automatically detect the unpredicted changes of kinematic constraints is demonstrated with a numerical example.

This paper presents the simulation architecture and strategy for dynamic simulation of chemical process and describes key features of developed dynamic simulation system, MOSA(MultiObjective Simulation Architecture). A plant structure may be partioned into several strong coupling units, called cluster. If this cluster is solved simultaneously, it is possible to simulate whole plant without introducing convergence problem of tear streams. In this study, a flexible modular approach based on clusters was proposed as a promising architecture for dynamic chemical process simulator.

In this paper, a high performance graphics system is suggested and its hardware architecture and software structure are described. The developed graphics system is a multiprocessing system that uses 6 i860 RISC CPU's and supports PHIGS language in a hardware level. The software is programmed with respect to the graphics pipeline and the software modules are distributed into each processor for the optimization of the performance. The implemented graphics system can draw about 100,000 3D polygons second.

In this paper, we propose dynamic hybrid control method which takes the manipulator dynamics into consideration and extend to two cooperating robots. The first step is the linearization of the manipulator dynamics and the second step is the design of position/force controllers for the linearized model which takes account of both the command response and the robustness of the controllers to modeling errors and disturbance. We also consider load sharing for each robot.

We describe about the graphic simulation system which supports the determination of efficient multirobot motions during cooperation. For the construction of the simulation software for multirobot motions, two problems are presented. First problem is that all the robot motions must be determinded using both the desired object motions and the holonomic constraints with the object. To find the robot motions combined with the various object motion path, the robot motions are derived from the desired object path instead of a master robot path. Therefore robot motions can be easily modifiable with the various object motions. This type of motion determination is different from that of the masterslaves method using the master robot motions. The other is that the developments of robot application softwares need a heavy cost when the participated robots or the tasks given to the robots are changed. Based on objectoriented programming paradigm, we present useful software objects describing robot application programming environment. The objectoriented programming paradigm increases the software reusability, reliability, and extensibility, and also provides the structual concepts to cope with the various demands of robot application programming.

The load distribution problem of two cooperating robots grasping one object is studied. The optimal joint torque needed for the desired motion is obtained by using a new objective function. A new objective function is defined for the minimization of joint torque effort and internal force. The optimal solution can be found by geometrical approach and analysis using the concept of force ellipsoid. Simulation results are presented with 6DOF PUMA robots.

In this paper, an unilateral teleoperation system using an universal master arm is developed. This system is composed of an universal master arm, a slave arm and a telerobot controller. The universal master arm has a vertically articulated type link structure, while an industrial robot is used as a slave arm. As the shapes of master arm and slave arm are different, the workspace mapping is needed, which maps the workspace of master arm to that of slave arm. Experimental results show that the slave arm of the developed system effectively follows the operator's motion.

At present, the manufacturing industry is in a process of great change of circumstances like meeting demands to involve a great variety of types and shorter product life and thus more flexible manufacturing. These changes cause the larger number of different tools and frequent tool changes, which lead to the considerable losses in productive time and the high amount of capital tied up in the tool area. In our country, for the most part, the individual tool are still being presetted according to a data sheet and the measured values are entered, output in a list manually or via punched tapes. This usually takes a considerable amount time and lead to a high error rate. This paper describes a computer controlled tool data management system combined with the bar code tool identification labeled on cutting tools.

This paper constitutes the simulator of the serial production line using Extended Petri Nets. We analyze each operations and effect of machine down, calculating the performance measures for improving the capability of manufacturing system. The performance measures are system efficiency, average utilization of machines, average buffer level, and so on. We intend to present the information of the operation states and various problem occurring in the systems.

The tire manufacturing process demands classification of tire types when the tires are transferred between the inner processes. Though most processes are being well automated, the classification relies greatly upon the visual inspection of humen. This has been an obstacle to the factory automation of tire manufacturing companies. This paper proposes an effective vision systems which can be usefully applied to the tire classification process in real time. The system adopts a parallel architecture using multiple transputers and contains the algorithms of preprocesssing for character recognition. The system can be easily expandable to manipulate the large data that can be processed seperately.

본 논문에서는 연속 공정자동화를 위한 시스템 제어장치 중 공정의 제반운영 및 제어상태를 감시하기 위하여 공정상태를 나타내는 여러가지의 data를 그래픽 모니터상에 graphics로 표시해 주는 User Interface System인 Operator Terminal의 설계 및 구현을 다룬다. 이를 위하여 Operator Terminal은 출력장치로 VGA board와 color monitor, 입력장치로는 mouse와 function keyboard, 통신을 위한 RS232C serial port를 갖는 IBMPC AT급의 mother board로 구성된다. 또한 program 저장은 상위 computer의 HDD를 이용한다. 그래픽 편집 및 모니터링을 위한 software는 한글 MS Windows 환경에서 구현한다.

This paper deals with the scheduling and simulation for FMC(Flexible Manufacturing Cells). In order to achieve CIM, there is a critical need to link factory level and machine level. The primary functions performed by this link for all jobs issued to the shop floor and cell include shortterm scheduling and dynamic operational scheduling. Here, hierarchical control structure is introduced to define these functions. And Intelligent scheduling through expert module is adopted for efficiency of FMC operation. Computer simulation reveals that expert scheduling method is better than heuristics in various performance index.

In this thesis, a selftuning control method based on Variable Structure System technique for tracking control of DirectDrive motor is presented. The selftuning control could not make the tracking error zero in the transient period. This tracking error may be due to disturbances or the error in parameter identification. To overcome this problem, a selftuning control method based on discrete time VSS technique is presented. The STC based on VSS technique gives good tracking performance of the reference signal in the transient period. The proposed controller is robust to parameter errors and disturbances. The performance of the proposed controller is compared with that of simple STC through digital computer simulation.

In this paper, the author proposed FLVSC(Fuzzy Logic Variable Structure Controller), of which control rules are extracted from the concepts of VSC(Variable Structure Control). FLC(Fuzzy Logic Controller) based on linguistic rules has the advantages of not needing of some exact mathematical model for plant to be controlled. The proposed method has the characteristics which are viewed in conventional VSC, e.g. insensitivity to a class of disturbances, parameter variations and uncertainties in sliding mode. In addition, the method has the properties of FLC  noise rejection capability etc. The computer simulations have been carried out for a DC servo motor to show the usefulness of the proposed method and the effects of disturbances and parameter variations are considered.

This paper presents a local trajectory generation method which is based on a sequence of reference posturevelocities and the efficient low level control algorithm which constructs the complete smooth curve from the trajectory specification. The reference trajectory generator(RTG) which is in between the local path planner(LPP) and the robot motion controller(RMC) generates a sequence of setpoints for each path segments from the LPP and pass it to the RMC. The RMC controls the motions of vehicle which should follow the sequence. In the feedback controller of VMC, the method which compensates robot posturevelocity error correctly is used. These methods are implemented on indoor autonomous vehicle, 'ALIVE' mobile robot. The ALIVE mobile robot system is implemented on the 32bit VME bus system: the two VME CPU's are used for RTG and RMC, while the 80C196KCbased VME board is used for motor controller.

The DC(DirectCurrent) servo motor has widely used for many application areas, FA(Factory Automation), OA(Office Automation) and home applications. But DC servo motor needs periodical inspection because it has brush and commutator. Recently, AC servo motor has expanded it's application areas due to for the development of the power semiconductor and control technology. But it has large torque ripple for it's small number of commutation. And it also has cogging torque due to permanent magenet rotor. Therefore it can't run balence rotarion. Many torque ripple reduction methods are published. In this paper, phase advanced method adopted for torque ripple reduction of AC servo motor. In this research, AC servo motor torque characteristic variation surveied under the phase advance control through the computer simulation. Under the simulation, the load inertia varied from 0.0001[Kg.m
$^{2}$ ] to 0.0314[Kg.m$^{2}$ ]. The result os nonlinear simulation, torque and speed ripple of AC servo motor under the phase advance control reduced approximately 50[%] and 10[%]. And maximum torque of AC servo motor under phase advance control condition increased about 5[%] as compare with fixed switching time. 
When a velocityaided strapdown inertial navigation system is loaded into a torpedo subjected to an extraneous force by the current, odometer measurement errors occur seriously. In order to compensate for navigation errors induced by large odometer biases, the Kalman Filter with separate bias estimator is applied, which separately estimates an unknown bias, and corrects the state estimate produced by the biasfree Kalman Filter to reflect the effect of the bias estimate.

A method for optimal route planning is presented with the assumption that the overall defended area is known in terms of threat potential function. This approach employes tangent plane to reduce the dimension of the state space for optimal programming problems with a state equality constraint. Onedimensional search algorithm is used to select the optimal route among the extermal fields which are obtained by integrating three differential equations from the initial values. In addition to being useful for the route planning through threat potential area, the trajectory planning will be suitable for general twodimensional searching problems.

In this paper, we present a method to design a coupled autopilot for STT missiles which have severe aerodynamic crosscoupling. The aerodynamic model is derived in the meneuver plane and, based on that model, an autopilot scheduled by the normal acceleration and the estimated bank angle is designed. Bank angle is obtained by a simple estimator. With the proposed autopilot, it is shown by computer simulations that induced moments are properly compensated and the performance is supiorior to the conventional autopilot.

In this paper, a model reference adaptive control algorithm is applied to the design of the normal acceleration controller for missiles with nonminimumphase characteristics. The method used in this paper is due to Ohkubo. In this scheme, a feedforward compensator is designed first so that the extended system becomes minimumphase and after that an adaptive control algorithms is designed for the extended system. The feedforwrd compensator is obtained by solving the robust stabilization problem. It is shown that the performance of the designed controller is satisfied via computer simulation.

A practical adaptive tracking filter for a maneuvering target is proposed in this paper by combining a modified input estimation technique with pseudoresiduals and a decoupled tracking filter in lineofsight Cartesian coordinate system. Since the adaptive tracking filter has decoupled structure and computes maneuver input estimates for each axis separately, it requires much less computations compared with the coventional tracking filter with MIE technique without degrading performance. Also, since pseudomeasurement noises in lineofsight Cartesian coordinate system are much less correlated compared with those of inertial Cartesian coordinate system, the proposed tracking filter produces less false alarms or miss detections to improve the performance.

This paper presents a realtime algorithm for an infrared seeker to find the real target automatically against various background noises without changing the reticle configuration. The modeling technique of infrared sources and analysis results of the various source types based on the FFT algorithm are included. Futhermore, a neural network is used to recognize the source type using the results of FFT analysis. The evaluation of target recognition for cases which can happen in real situation is also treated.

As an approach to design the intelligent controller, this paper proposes a new FNN(Fuzzy Neural Network) control method using the hybrid combination of fuzzy logic control and neural network. The proposed FNN controller has two important capabilities, namely, adaptation and learning. These functions are performed by the following process. Firstly, identification of the parameters and estimation of the states for the unknown plant are achieved by the MNN(Model Neural Network) which is continuously trained online. And secondly, the learning is performed by FNN controller. The error back propagation algorithm is adopted as a learning technique. The effectiveness of the proposed method will be demonstrated by computer simulation of a two d.o.f. robot manipulator.

We propose a control architecture for the fuzzy logic control of robot manipulators and a rule base derivation method for a fuzzy logic controller(FLC) using a neural network. The control architecture is composed of FLC and PD(positional Derivative) controller. And a neural network is designed in consideration of the FLC's structure. After the training is finished by BP(Back Propagation) and FEL(Feedback Error Learning) method, the rule base is derived from the neural network and is reduced through two stages  smoothing, logical reduction. Also, we show the performance of the control architecture through the simulation to verify the effectiveness of our proposed method.

The optimal FIR filter and smoother for estimating the state of 2D statespace model are derived. Using the 2D state covariance propagation rule suggested in this paper, the efficient algorithm to find the impulse response of the optimal FIR filter and smoother can be found. The filter and smoother can be applied to the image restoration problems and some examples of the application will be given.

Diagnostic prototype expert system was developed by analyzing the measured acoustical data of automobile. For the utilities of this system, 1/3 octave filter(bandpass filter) and A/D converter were used for data acquisition and then information was analyzed using signal processing technique and pattern recognition by Hamming network algorithm. In order to raise the reliability of the diagnostic results, fuzzy inference technique was applied and, the results were displayed as graphical method to help the novice in diagnostic field. The validation of this diagnostic system was checked through experiments and it showed and acceptable performance for diagnostic process.

In this paper, discretetime H
$_{\infty}$ controller design in .delta. domaion using Normalized Coprime Factor plant description is proposed and the loopshaping method developed by Mcfalane[2], which is known to be very simple and systematic method, is adopted here in H$_{\infty}$ controller design. In particular, we show that .delta. H$_{\infty}$ controller proposed here is a unified form for the continuous and discretetime cases.es. 
The conventional closedloop guidance commands are generated from a simplified point mass model for real time operations. In real situations, the generated guidance commands are applied to the original rigid body. This can cause attitude instability of the vehicle. In this paper, in order to solve the attitude instability problem in the guidance system sense, the influence of the guidance commands on a launch vehicle attitude is derived quantitatively. The checking method of the attitude stability conditions that uses Liapunov theorem is proposed, and the attitude stabilizing method is also proposed. The attitude stability is accomplished by subtracting the influence of the guidance commands that destabilize the vehicle attitude. The closedloop guidance commands generated from the simplified point mass model may destabilize the vehicle attitude, which is verified through simulations. In this case, the vehicle attitude can be always stabilized with the proposed attitude stabilizing method without additive fuel consumption.

The nonrational transfer function of a BernoulliEuler beam, as an important component of a flexible structure, is analyzed. The true pattern of zeros of that transfer function is investigated as a function of sensor and actuator seperation. Translational displacement sensors are used for two cases in which a force input and a moment input are seperately applied. When the displacement sensor is located at a certain point, the first pair of zeros on the real axis of the splane arrive at the origin and cancel the rigidbody mode. The location of the translational displacement sensors on the beamat which the rigidbody mode of the beam is unobservable is analyzed as the center of percussion and is uniquely located for each case. If sensor is moved beyond such a point, a pair of zeros appear on the imaginary axis and move away from the origin along the imaginary axis of the splane.

This paper defines the relationship between PI type fuzzy control system and conventional PI control system, and discusses the relationship of parameters and control action in fuzzy controller. The tuning algorithm that updates ouput variable scaling factor of fuzzy controller is proposed .The proposed sheme is applied to the simulations of 2 selected dynamical plants. The simulation results show that the controller is effective in controlling dynamical plants.

The inference of fuzzy controller can be considered a mapping from the controller input to membership value. The membership value, a kind of weight, has a role to decide if the input is appropriate to the rule. The membership function is described by several values, which are decided by a learning method. The learning method is adopted from adaptive filtering theory. The simulation shows the proposed fuzzy controller can learn linear and nonlinear functions. the structure of the proposed fuzzy controller becomes a kind of neural network.

A neural model predictive control strategy combining a neural network for plant identification and a nonlinear programming algorithm for solving nonlinear control problems is proposed. A constrained nonlinear optimization approach using successive quadratic programming cooperates with neural identification network is used to generate the optimum control law for the complicate continuous/batch chemical reactor systems that have inherent nonlinear dynamics. Based on our approach, we developed a neural model predictive controller(NMPC) which shows excellent performances on nonlinear, modelplant mismatch cases of chemical reactor systems.

In this paper, the optimal regulating algorithm are developed by adjusting the scaling  factor and estimating the linguistic control rule according to the results which are attaining from the first defined parameters, and the fuzzy level controller is realized usign the algorithm. The proposed controller apply to the fuzzied level control process, the cement grinding plant. The results shows that the control performance is greatly improved.

In this study, we proposed the fuzzy modeling method and designed a modelbased logic controller for Activated and Sludge Process(A.S.P.) in sewage treatment. The identification of the structure of fuzzy implications is carreid out by use of fuzzy cmeans clustering algorithm. And to identify the parameters of fuzzy implications, we used the complex and the least square method. To tune the premise parameters automatically the complex method is implemented. The modelbased fuzzy controller is designed by rules generated from the identified A.S.P. fuzzy model. The feasibility of the proposed approach is evaluated through the identification of the fuzzy model to describe an inputoutput relation of the A.S.P.. The performance of identified modelbased fuzzy controller is evaluated through the computer simulations.

On the generation of robotic assembly sequences using disassemblability based on separable directionThis paper presents a method for the automatic generation of assembly sequences based on the recursive extraction of a preferred part along with the verification of its disassemblability. To verify the disassemblability of the part we inference the geometric precedence constraints using a method of extracting separable directions for the part and determine the disassemblability cost by the separability and stability cost. The proosed method provides a local optimal solution of finding a cost effective assembly plan, and the feasibility of calculating robot motion programs by evaluating separable directions in flexible manufacturing application. A case study is given to illustrate the concepts and procedure of the proposed scheme.

We are now doing research for the drawing check of local parts in mechanical drawing made by a CAD system. It needs the recognition of drawing elements with respect to the local parts. Because, we usually abbreviate the dimensioning in the mutually related drawing elements. This paper is concerned with a computer aided supporting system to the dimension check and recognition of local parts in mechanical drawings. This sytem has been applied to some examples and we have confirmed the feasibility of this checking method.

In this paper, an operating system is designed for NC controllers. Host computer is the IBMPC, and the controller panel is designed in graphics mode to look easily for users. Pulldown menu is selected, which is useful to minimize a mistake made by users, so users can use NC controller more effectively. Also, many functions are proposed, such as NC editor for programming the NC program directly, simulation function for testing editted NC program, and new keypad that appropriates for the functions of OS.

This paper deals with an exit strip gauge controller design for a cold rolling mill process with one input and two outputs. In order to attenuate the exit strip gauge variation caused by the Backup roll eccentricity and the entry guage variation, an H
$_{\infty}$ controller design methodology is used to keep the frequency weighted mixed sensitivity function small. Effectiveness of the proposed controller is demonstrated by a computer simulation.. 
In GMA welding processes, monitoring and control of weld quality are extremely difficult problems. This paper describes a neural networkbased method for monitoring and control of weld pool sizes. First, weld pool sizes are estimated via a neural estimator using multipoint surface temperatures, which are strongly related to the formation of weld pool, and then controlled using the estimated pool sizes. Two types of controllers using the pool size estimator are designed and tested. To evaluate the performance of the designed controllers, a series of simulation studies was performed.

Kim, DongIl;Song, JinIl;Kim, SungKwan;Lee, ChoongHwan;Lee, YunSuk;Kang, Moon;Na, SangKeun;Lim, YongGyu;Nam, KiJun 537
This paper presents Samsung CNC (Computer Numerical Controller) system with an intel 80486/487 as the main CPU and a 32 bit floating point DSP(Digital Signal Processor) TMS320C30 as the motion control CPU. The Samsung CNC system diverse userfrienly characteristics such as multitasking, powerful menu system, internal PLC system, and 2/3 dimensional graphics in wire and solid mode. The main CPU executes central processing program, user interface program, interpreter, BMI, etc while the motion control CPU carries out some interpolations, acceleration/deceleration, and PID control algorithm with feedforward terms. Complex interpolations except linear and circular ones are performed on the main control CPU. The experimental results for the circular interpolation under linear acceleration/deceleration shows that the proposed CNC system can be widely used in controlling machining centers with good machining accuracy. 
The phased array antenna has the ability to perform adaptive sampling by directing the radar beam without inertia in any direction. The adaptive sampling capability of the phased array antenna allows each sampling time interval to be varied for each target, depending on the acceleration of each target at any time. In this paper we design a threedimensional adaptive tracking algorithm for the phased array radar system with a given set of measurement parameters. The tracking algorithm avoids taking unnecessarily frequent samples, while keeping the angular prediction error within a fraction of antenna beamwidth so that the probability of detection will not be degraded during a track update illuminations. In our algorithm, the target model and the sampling rate are selected depending on the target range and the target maneuver status which is determined by a maneuver detector. A detailed simulation is conducted to test the validity of our tracking algorithm for encounter geometries under various conditions of maneuver.

In this paper and recursive version of orthogonal ARMA identification algorithm is proposed. The basic algorithm is based on GramSchmidt orthogonalization of automatically selected basis functions from specified function space, but does not require explicit creation of orthogonal functions. By using two dimensional autocorrelations and crosscorrelations of input and output with constant data length, identification algorithm is extended to cope slowly timevarying or ordervarying delayed system.

In this paper the normalized innovations squared of a Kalman filter is used to detect a bias level in prediction errors due to target accelerations. The probability density function of the normalized innovation squared is obtained for a steady state Kalman filter, and it is used to calculate the detection probability of the bias level. A typical example is given to compute the detection probability.

Bearing information of target is used critically for target tracking in underwater environment. In passive sonar, target bearing measurements are obtained by processing the acoustic signal emanating from the target. PDA tracking algorithm is usually applied in this case since bearing measurements have several peaks due to interference with other acoustic sources or reflections from underwater media. In this paper, we propose a modified PDA algorithm adopting new probabilistic distributions of the number, position, and amplitude of peaks based on the analysis of real data. This algorithm is tested on real and artificially generated data. The computer simulation result shows improvement of the tracking performance.

It is tragic that the Korean Airline Boing 747, KE007, wandered hundreds of miles off course into Soviet airspace and was shot down on September 1, 1983. The exact cuases are not known yet. Thus, speculation centers on human error or faulty procedure of three Litton LTN72R inertial navigation systems(INS) with which the KAL KE007 was equipped. The inertial platform must be aligned before the INS can be used as a precision inertial navigation system. This analysis checks a possibility that the navigation errors are caused by a wrong INS alignment procedure assuming it is done at Anchorage. Possible causes for the navigational position error, such as alignment errors and gyro drift errors, are analyzed through inertial navigation system error prapagation simulations. A set of misalignment angle is estimated to determine what degree of alignment errors are required to cause the navigation error assuming that the accident is caused by the INS misalignment.

A guidance loop of the missile system which uses a command to line of sight(CLOS) guidance law is designed based on LQG theory. In the environment of the severe tracking system noise, the system requires small aerodynamic control fin travel and small miss distance simultaneously. Results from a sample airframe shows good performance against a randomly maneuvering target.

In this paper, the optimal homing guidance problem is investigated for the general missile/target models described in the statespace. The closedform solution of the optimal guidance law derived, and its asymptotic properties are studied as the timetogo goes to infinity or zero. Futhermore, several approximate solutions of the optimal guidance law are suggested for realtime applications.

In this paper, we show that various navigation constant values in PNG law can result in as optimal gains when we introduce proper timevarying weighting functions into the cost function of an optimal control problem. we then apply this idea to the guidance problem where we are required to achieve a given impact angle as well as the zero miss distance. As a result, we obtain a set of optimal guidance laws each of which could be related to a navigation constant in PNG. Some basic properties of these guidance laws are also presented.

The 3dimensional PPNG(Pure proportional navigation guidance) law was proposed about forty years ago, but the satisfactory analysis of its performance has not been presented since then. In this note, we prove under some reasonable assumptions that the missile guided according to this 3dimensional PPNG law can always intercept a randomly maneuvering target if (1) the target acceleration varies with a certain bound, (2) the navigation constant is selected large and (3) the initial heading error is small. We introduce a Lyapunovlike method that proves to be a very powerful tool in obtaining our results.

An integrated guidance scheme for guided weapon system is described in this paper. Against conventional guidance methods, this method combines an autopilot and a guidance law. The controller is designed using LQ regulator whose performance index is different from other optimal guidance laws. Since dynamics of the system is considered in the derivation, the controller performance is improved. By simulation, the suggested method shows better performance in minimum distance sense than conventional guidance schemes such as Bang Bang guidance or Pursuit Guidance. Since the suggested method provides smooth rudder deflection in contrast to the conventional method, the load on a energy source of the system can be greatly lessened.

In CLOS guidance, feedback compensation of the Coriolis acceleration is used to reduce miss distance. This paper presents the effects of the bandwidth of target and missile on the optimal Coriolis acceleration compensation. A state space formulation of CLOS guidance is used to implement CLOS guidance in feedback form. And the LQR control method is applied to find the optimal feedback gain. From the analysis of the Riccati equations of the optimal control, the following facts are observed: When the target is agile, the optimal gain is reduced, since the compensation becomes ineffective. The missile bandwidth also affects the Coriolis accleration compensation. Narrower missile requires more compensation for the Coriolis acceleration.

In this paper, we consider an optimal guidance problem with both the terminal impact angle and control constraints in addition to the usual zero miss distance constraint. We first present the optimal solution of the problem for the missile of an arbitrary order, and show that it is a linear combination of a step response and a ramp response of the missile. Therefore the usual practice of using the control obtained by saturating the optimal solution for the case of unlimited control may result in a large terminal miss. A method called the initial command saturation is suggested to reduce this terminal miss, where the control in the initial phase of guidance is forced to be saturated until a certain condition for a guidance variable is met.

For hydroforming process, determination of backup fluid pressure in chamber is one of the most essential tasks. In this paper, we present a backup pressure estimation system which estimates the backup pressure of hydroforming process utilizing a multilayered neural network. The neural network learns the nonlinear relation ship between the backup pressure and the geometric state variables of hydroforming process. The proposed method does not necessitate sophisticated analysis on hydroforming process but some geometric intuition. The experimental results show that the neural network well approximates the nonlinear relationship between the backup pressure and the geometric state variables of hydroforming process, thus giving the good estimation of backup pressure vs punch stroke curve.

We propose a new control architecture which consists of a PI controller and a neural network(NN) controller connected together in parallel. This architecture is well adapted to a wide range of uncertainties and variations of systems. The NN controller is learned through weights of the emulator which identify the dynamic chracteristics of the systems. A performance evaluation level of two NN's decides automatically which controller of the two controllers will be used mainly. The PI controller operates mainly during learning phase of the NN controller whereas a good performance is obtained from the NN controller only, when the NN controller is learned sufficiently.

In pattern recognition, the conventional neural networks contain a large number of weights and require considerable training times and preprocessor to classify a transformed patterns. In this paper, we propose a constrained pattern recognition method which is insensitive to rotation of input pattern by various degrees and does not need any preprocessing. Because these neural networks can not be trained by the conventional training algorithm such as error back propagation, a novel training algorithm is suggested. As such a system is useful in problem related to calssify overse side and reverse side of 500 won coin. As an illustrative example, identification problem of overse and reverse side of 500 won coin is shown.

Mathematical modeling is majorly divided into three parts: the derivation of models, the fitting of models to data, and the simulation of data from models. This paper focuses on the parameter optimization which is necessary for the fitting of models to data. The method of simulated annealing(SA) is a technique that has recently attracted significant attention as suitable for optimization problem of very large scale. If the temperature is too high, then some of the structure created by the heuristic will be destroyed and unnecessary extra work will be done. If it is too low then solution is lost, similar to the case of a quenching cooling schedule in the SA phase. In this study, therfore, we propose a technique of determination of the starting temperature and cooling schedule for SA phase.

The position control for DC motor under the unpredictable load variations is presented. Neural network controller trained to deal with this problem provide the estimates of system parameters. Pole placement is also performed in accordance with them. The proposed method is validated through computer simulation.

Recently many dynamics control algorithms using robot dynamic equation have been proposed. One of them, Kawato's feedback error learning scheme requires neither an accurate model nor parameter estimation and makes the robot motion closer to the desired trajectory by repeating operation. In this paper, the feedback error learning algorithm is implemented to control a robot system, 5 DOF revolute type movemaster. For this purpose, an actuator dynamic model is constructed considering equivalent robot dynamics model with respect to actuator as well as friction model. The command input acquired from the actuator dynamic model is the sum of products of unknown parameters and known functions. To compute the control algorithm, a parallel processing computer, transputer, is used and realtime computing is achieved. The experiment is done for the three major link of movemaster and its result is presented.

In this paper, a hybrid position/force control scheme is proposed. The control scheme modifies the position command for force control against constraint surface of environment and is very simply designed and implemented. The merits of the control scheme are that it can cope with change of constraint conditions and small position inaccuracy of the environment. A constraint surface position observer is also proposed to reduce disturbances on controlled force.

Trajectory tracking control problems are described for a twolink robot manipulator with artificial rubber muscle actuators. Under the assumption that the socalled independent joint control is applied to the control system, the dynamic model for each link is identified as a linear secondorder system with timelag by the step response. Two control laws such as the feedforward and the computed torque control methods, are experimentally applied for controlling the circular trajectory of an actual robot manipulator.

This paper presents a modelbased control scheme for a robot manipulator to track a desired trajectory as closely as possible in spite of a wide range of manipulator motions and parameter uncertainties of links and payload. The scheme has two components: a nominal control and a variational control. The nominal control, generated from direct calculation of the manipulator dynamics along a desired trajectory, drives the manipulator to a neighborhood of the trajectory. Then a discretetime PID regulator is designed based on the linearized dynamic model and Linear Quadratic(LQ) method, which generates the variational control that regulates perturbations in the vicinity of the desired trajectory.

This paper presents to accomplish successfully a multiinput real time control by applying control hierarchy for sliding mode of multijoint manipulators whose nonlinear terms are regarded as disturbances. We could simplify the dynamic equations of a manipulator and servo system, which are composed of linear elements and nonlinear elements, by assuming that nonlinear terms, which are Inertia term, gravity force term, Coriolis force term and centrifugal force term, are external disturbance. By simplifying that equation, we could easily obtain a control input which satisfy sliding mode of multiinput system. We proposed a new control input algorithm in order to decrease chattering by changing control input according as effect of disturbance if a control response become within allowance error range. In this experiments, we used DSP(Digital Signal Processor) controller to suppress chattering by time delay of calculation and to carry out real time control.

In this paper, the control of a DC servo motor position control is implemented with fuzzy controller using IBM PC AT. In addition to the fuzzy controller with the error and change of error inputs PI controller is applied as well to improve the steady state error response. It shows the above configuration results in a good control characteristics.

In this paper, we present a DSPbased high dynamic performance torque control scheme of variable reluctance motors(VRM's) for DD(Direct Drive) robots via function inversion technique. The VRM with our controller behaves like DC motors, and hence developed torque tracks given torque command accurately with no torque ripples. Furthermore, our torque control algorithm ensures the production of maximum constant torque under maximum current limitation, minimizes power loss in each phase resistance, and takes magnetic saturation effect into account. Also, since our control algorithm is represented in the form of lookup table, it can be easily implemented with simple digital circuits and this tabular design method is computationally more accurate and simpler compared to the prior methods.

This paper presents a velocity regulation scheme for a DC motor subjected to random torque and velocity measurement noises of white noise type as well as unknown constant load torque (bias). The scheme separately estimates an unknown bias in addition to state estimation by the biasfree Kalman Filter, and reflects the effect of the bias estimate to the armature input voltage such that velocity variations be regulated. It is shown via computer simulations that the performance of the present scheme is superior to that of the conventional analog PI regulator.

In industry, the speed control of singlephase induction motor in domestic use is generally controlled by a simple ONOFF or PID control method. However, in this case, in order to have a good speed regulating characteristics, itself should be modified in accordance with the optimum PID factors which are varied each time operating speed changes. Shortening the development time and saving the cost which are needed to modify the controller is a major problem to be solved now in industry. In order to alleviate the above difficulties, it is proposed to apply adaptive control technique using MRFAC(Model Reference Following Adaptive Control) for the speed control of singlephase induction motor which has scarcely been studied. In this paper, the above speed control technique is achieved using MCS96 one chip micro controller with a good speed control characteristics and it is expetted to open a wide application field in the speed control of singlephase induction motor in the future.

Solar arrays and antennas of the satellite are usually stowed within the dimensions of the launchvehicle fairing and deployed in the orbit. To solve such multibody dynamic problems, differential equations and algebraic equations are simultaneously solved, and special solution techniques are required. In this paper, Lagrange multipliers associated with the constraints are iteratively computed by monotonically reducing an appropriately defined constraint error vector, and the resulting equation of motion is solved by a wellestablished ODE technique. Defomable bodies as well as rigid bodies are treated, and applications to satellite solar arrays are explained.

Implicit guidance algorithm can deal with the trajectory error rapidly, but it has to save much data. If, however, the control variable is represented by a specific function form, a few parameters will suffice to define the control variable. In this paper, we study the method of updating the parameters of the control function for the reduction of trajectory errors. The method proposed here does not require much memory for guidance.

This study concerns about the orbit prediction and orbit determination of Korean future communication satellite, called "Moogunghwa", which will be motioned in the geostationary orbit. Perturbation effect on the satellite orbit due to nonspherical geopotential term, lunar and solar gravity, drag force of the atmosphere and solar radiation pressure was investigated. Cowell's method is used for orbit prediction. Orbit determination was performed by using EKF which is suitable for realtime orbit determination. The result shows that the characteristics of the satellite orbit has drift. So the periodic control time and control value in the view of the periodic of error can be provided. The orbit determination demonstrated the effectiveness since the convergence performance on the position and velocity error , and state error standard deviation is reasonable.easonable.

In this paper, MicroGenetic algorithms(.mu.GAs) is proposed on a minimumtime path planning for robot manipulator, which is a kind of optimization algorithm. The minimumtime path planning, which can allow the robot system to perform the demanded tasks with a minimum execution time, may be of consequence to improve the productivity. But most of the methods proposed till now suffers from a significant computation burden and can't often find the optimal values. One way to overcome such difficulties is to apply the MicroGenetic Algorithms, which can allow to find the optimal values, to the minimumtime problem. This paper propose an approach for solving the minimumtime path planning by using MicroGenetic Algorithms. The effectiveness of the proposed method is demonstrated using the 2 d.o.f plannar Robot manipulator.

In the field of assembly processes, nonrigid parts such as wires, tubes, gaskets and 0rings cannot be assembled automatically. And although many researches have been made for rigid part mating, there are not substantial studies in flexible parts assembly field. In this paper, assembly stages of flexible parts are classified and some analysis are made. FEM was used to estimate the relationship between deformation and reactive forces. An assembly algorithm adopting reciprocal twisting motion was proposed and the assembly tool design methodology was discussed.

본 논문은 주조공정후 주물의 라이져(Riser)나 버어(Burr) 등을 제거하는 페틀링(Fettling) 작업의 자동화에 대해서 기술한다. 이 작업을 주물의 마무리 작업이라 하는데 이 분야는 작업상의 특수성과 적용 로봇의 미개발로 아직 산업현장에서 실용화 되지 못한 분야이다. 본 논문에서는 그라인딩 작업 환경과 이 분야에 적용되는 로봇의 필요조건에 관해서 기술하고 이 분야에 적합하도록 제작된 MC 560 로못을 소개한다. 주물제작의 마무리 작업은 로봇 이외에도 여러종류의 가공기계와 주변기기를 요구한다. 본 논문은 주물의 재질과 형상에 따른 가공기계 선택방법과 주변기기의 설계 등을 기술하며 자동화 시스템의 구성요소, 작업공정, 작업효과 등에 관해서 논한다.

An intelligent robot control system is developed, which is based on extensible hardwares and softwares. The system could be used to test advanced and complex real time application programs to avoid constraints on present robot control system in executing a complex or precise algorithms, due to the limitation of hardware and software. In this paper we used the RCCL(Robot Control C Library) on SUN4 as a supervisory system that plays the path planning and manmachine interface. And we used VxWORKS as a real time OS on a VME bus CPU equiped with some interface boards. Two systems were connected through the Ethernet network. We used the 4 axis manipulator, FARA, developed by Samsung Electronics Co.

A hydraulic gun/turret servo system requires fast and robust controller performance because of severe operating condition and precise target tracking objective. Digital controllers are able to satisfy this requirement due to high speed electronic device. The purpose of this study is to compare with preEPU with newdesigned optimal, adaptive controllers by simulating nonlinear hydraulic simulation program. The designed digital controller shows good tracking performance and robustness to disturbance.

A transient hydraulic flow rate computation scheme is described here so that the transient hydraulic flow rate can be determined using the dynamic pressure measurements at the ends of a straight flowline with a dynamic, model of the hydraulic line. Simulation results indicate that the method is relatively robust to realistic levels of uncertainties in the fluid properties.

In this study, heavy loads driving servo control systems, which are composed of electrohydraulic servovalve, hydraulic motor/cylinder, gear box and link mechanism, are investigated for implemention. To predict the performances of the systems, modelling and simulation with some nonlinearities are carried out. Simulation results are compared with experimental results.

The Stewart platform is one example of a motion simulator which generates 6 DOF motion in space by 6 actuators connected in parallel. The present SISO controllers are designed to track displacement command of each actuator computed from reference 6 DOF motion of platform by Stewart platform inverse kinematics. But this type of control can't cope with external load variation, geometric configuration of motion simulator, and different dynamic behavior of 6 DOF motion. In this paper, a multivariable controller using H optimal control theory is designed for linerized simulator model with each actuator driving force as control input and platform 6 DOF motion as measured output. Nonlinear simulation result of the H
$_{\infty}$ MIMO controller is not satisfied in steadystate characteristics. But the proposed H$_{\infty}$ + PI control scheme shows acceptable performance.e.e. 
This paper proposes a new linear robust state feedback controller for the linear discrete time systems which have uncertainties in the state and input matrices. The uncertainties need not satisfy the matching conditions, but only their bounds are needed to be known. The proposed controller is derived from the linear quadratic game problem, which solution is obtained via the modified algebraic Riccati equation. The controller guarantees the robust performance bound. The bound of the solution and the condition of the uncertainties, which can stabilize the uncertain system are explored.

The modeling and control design schemes are developed for maglev systems with a combined lift and guidance. First, bond graph techniques are applied for modeling these multienergy domain systems more logically and systematically. And the stability loop via pole placement and the performance loop via loopshaping LQ control are designed. The suggested controller satisfies the required characteristics of stability and performance simultaneously. Finally, the robustness of the synthesized maglev control system is evaluated for the variations of air gap and vehicle mass through computer simulation.

In this paper, we proposed a mixed H
$_{2}$ /H$_{\infty}$ optimal controller design method using the parameterization of H$_{2}$ suboptimal controller. The method is based on the minimization of H$_{2}$ performance measure with an H$_{\infty}$ norm constraint. We also derived the necessary and sufficient conditions for existence of solution from the decoupled Riccati equations. And the designed controller has statespace representation.n. 
본 논문에서는 미지의 불확실한 시스템에 적용할 수 있는 제어 알고리즘을 제안하였다. n차 수 단일 입출력 시스템의 상태를 "uncontrolled" 상태와 "controlled" 상태로 분리시키며 아울러 uncontrolled 부시스템이 임의의 n1개의 특성근(eigen values)을 갖도록 하는 변환 행렬을 유도하고, 이를 이용하여 새로운 강인한 제어기를 설계하였다. 적응 제어 이론과 가변 구조 제어 이론의 개념을 도입하여 새로운 제어 법칙을 제안하였으며 제안된 제어기의 안정도는 Lyapunov 안정도 이론을 이용하여 증명하였다. 컴퓨터 시뮬레이션 결과 제안된 적응 제어 알고리즘이 강인성과 안정성을 갖음을 확인할 수 있었다.

The hydrodynamic coefficients estimation problem is important to develop an underwater vehicle and design a controller for it. In this paper, an identification theory, the Extended Kalman Filter, is applied to this parameter estimation problem. In the case that a process noise is not used, all of the parameters are almost exactly converged to the true values respectively. When a process noise is used, all of the parameters are converged to the true values, too, although some parameter estimates are slightly biased. The comparisons of the two trajectories between those generated by the true parameters and those by the estimated parameters show that the parameter estimation problem is wellsolved.

It is necessary to estimate hydrodynamic coefficients to design the autopilot system and motion simulator of submersible vehicle. In this paper, an algorithm was designed to estimate hydrodynamic coefficients of submersible vehicle. Using this algorithm, the hydrodynamic coefficients were estimated from measurements of full scale trial. The estimated hydrodynamic coefficients were used for the design of an autodepth controller(ADC) of submersible vehicle, and the resulting ADC are proved to have a better performance than the previous one.

This paper describes the autodepth control system with depth control tank for low speed submersible vehicle that can be used for both near surface and deeply submerged keeping operations. The PDA control algorithm is used to design controller and adaptive notch filter is designed to eliminate the dominant frequency of seaway. The computer simulations demonstrate the excellent depth keeping performance of the controller under seaway effects.

We present a multisensor fusion method in positioning control of a robot by using fuzzy logic. In general, the vision sensor is used in the gross motion control and the force/torque sensor is used in the fine motion control. We construct a fuzzy logic controller to combine the vision sensor data and the force/torque sensor data. Also, we apply the fuzzy logic controller to the peginhole process. Simulation results uphold the theoretical results.

When a mobile robot moves from one place to another, position error occurs due to the limit of accuracy of robot and the effect of environmental noise. In this paper. an accurate method of estimating the position and orientation of a mobile robot using the camera calibration is proposed. Kalman filter is used as the estimation algorithm. The uncertainty in the position of camera with repect to robot base frame is considered well as the position error of the robot. Besides developing the mathematical model for mobile robot calibration system, the effect of relative position between camera and calibration points is analyzed and the method to select the most accurate calibration points is also presented.

Readings from robotic sensors are somewhat uncertain. This uncertainty problem makes it difficult to employ the sensor feedback controlled robots widely in real industrial sites, In this paper, redundant sensor fusion techniques are discussed to effectively overcome the sensor uncertainty, A weighted averaging technique is proposed under static and dynamic sensing environments. Proposed technique is tested by the experiments of stereoscopic 3d position measurements.

Much experience and knowledge is needed in welding because there are many working parameters and quantitative description is difficult. Therefore, introduction of expert system based on such data base has been required. In this study, welding heat and shape of bead was controlled by fuzzy inference with the welding condition, position error and voltage and current error of robot. For this, torch trajectory of robot was generated by modeling the working data with CAD and then welding was carried out through down loading to robot. And working error was controlled by online communication.

Due to the inherent complexity in the CIM operation, achieving the triple, meeting duedates, maximizing machine utilization, and maximizing system throughput simultaneously is practically impossible. Targeting the smalltomedium size industries of Korea, we propose an exrert system that 'provides a good and practical solution to the CIM operation problems. Heavy consideration has given to the realtime and dynamic nature of CIM in the development process of the system. The system is under testing stage at KUFMS, model CIM plant.

Some concepts and methods for information system as a part of CIM system in ship production are proposed in this paper. The characteristics and problems focused on product design and production planning and control are analyzed. A conceptual framework for CIM in ship production is presented, which is focused on a product model for information system. A methodology for the practical use of information for each managerial class is also presented.

The objective of cell operation in CIM is to maximize system efficiency, not the cell itself. We introduce a cell operation scheme that pursues the direction. For specific cases, work center and AS/RS are closely investigated. Structures of their operation mechanism and methods of job scheduling are introduced along with an expert system developed for the scheme. The cell operation softwares developed are now under test at K.U.FMS, a model CIM plant.

Distortion analysis of digital filters for the ECG signal processing is presented. Several band pass and band reject filters are designed for the analysis. Computer simulations are performed to compare the distortions of the Butterworth type filters and linear phase optimal fitters. The designed filters are applied to power line interference cancelling in ECG signals.

A novel steepest descent adaptive filter algorithm, which uses the instantaneous stochastic gradient for the steepest descent direction, is derived from a newly devised performance index function. The performance function for the new algorithm is improved from that for the LMS in consideration that the stochastic steepest descent method is utilized to minimize the performance index iterativly. Through mathematical analysis and computer simulations, it is verified that there are substantial improvements in convergence and misadjustments even though the computational simplicity and the robustness of the LMS algorithm are hardly sacrificed. On the other hand, the new algorithm can be interpreted as a variable step size adaptive filter, and in this respect a heuristic method is proposed in order to reduce the noise caused by the step size fluctuation.

This paper presents the design of an automated EEG analyzing system. The design considerations including processing speed, A/D conversion, filtering, and waveforms detection, are overviewed with the description of the associated EEG characteristics. The architecture of the currently implemented system consists of a pcontroller based frontend signal processing unit and a host computer system. The data acquisition procedures are described along with a couple of illustrations of the acquired EEG/EOG signal.

Tracking performance depends on the quantity of the measurement data. In the KalmanBucy filter and other trackers, this dependence is well understood in terms of the measurement noise covariance matrix, which specifies the uncertainty in the value of measurement inputs. In this paper, we derived approximated error covariance matrix to evaluate the dependence of target detection probability and false alarm probability in the presence of uncertainty of measurement origin.

There are many specific requirements in the actuation, system for high speed underwater vehicle, such as size, weights, power etc.. In this paper, a high performance compact hydraulic actuation system to satisfy such requirements was designed. The controller of the system was designed using both the conventional PID and VSC which were known to have reliability, robustness respectively. The performance analysis was done for the designed actuation system through computer simulation.

The disturbance and time delay can often cause a significant error in the estimation of trajectory of a underwater vehicle. The time delay considered in this study is due to the delayed rudder response to the rudder input from the guidance control part. The simulation tests are performed on maneuver with constant rudder angle, zigzag maneuver, diveclimb maneuver, and corridor pattern maneuver. The results are compared with those of without delay cases.

The problem of robust control of a underwater vehicle subject to variation of a real parameter and velocity is considered. The controller set which stabilized perturbed plant is chosen using numerical gradient method and the controller is used for nominal performance and robust performance. Simulation results are presented to show that the precise montion control of the controller is accomplished under perturbation in the system.

An elevator group supervisory control logic is investigated to supervise multiple elevators, ensuring their efficient operations. In this paper, a predictive fuzzy control logic of group elevator system is developed for coping with multiple control objects and uncertainty of system state. Simulation of this control logic shows considerable improvements of system performance by the reduction of average waiting time and long wait rate.

This paper presents a method on applying Genetic Algorithm(GA), which is a wellknown high performance optimizing algorithm, to construct the selforganizing fuzzy logic controller. Fuzzy logic controller considered in this paper utilizes Sugeno's hybrid inference method, which has an advantage of simple defuzzification process in the inference engine. Genetic algorithm is used to find the optimal parameters in the FLC. The proposed approach will be demonstrated using 2 d.o.f robot manipulator to verify its effectiveness.

A fuzzy controller is designed for compensating the crosscoupling effect of induced roll due to the dynamic characteristics of three fin torpedo. Since the utilization of fuzzycoprocessor has many interfacing problems with typical microprocessors of the guidance and control unit, the simplified fuzzy inference method based on nonfuzzyprocessor is proposed to implement fuzzy controllers of three fin torpedo. This method provides a flexible rulebase design to guarantee the robust control. The good potential of the proposed design is shown through realtime simulations using both a mathematical model on AD100 computer and an implemented controller on Intel 80C186/80C 187 microprocessors employing 12bit A/D converter.

It is shown that there exists a nonlinear mappping which transforms features and their changes to the desired camera motion without measurement of the relative distance between the camera and the part, and the nonlinear mapping can eliminate several difficulties encountered when using the inverse of the feature Jacobian as in the usual featurebased visual feedback controls. And instead of analytically deriving the closed form of such a nonlinear mapping, a fuzzy membership function (FMF) based neural network is then proposed to approximate the nonlinear mapping, where the structure of proposed networks is similar to that of radial basis function neural network which is known to be very useful in function approximations. The proposed FMF network is trained to be capable of tracking moving parts in the whole work space along the line of sight. For the effective implementation of proposed IMF networks, an image feature selection processing is investigated, and required fuzzy membership functions are designed. Finally, several numerical examples are illustrated to show the validities of our proposed visual servoing method.

There has been continuous effort to achieve total automation of the power plants. But due to complexity, nonlinearities and timevarying properties of the system, no success has been reported if conventional and/or modern control theories are applied. In spite of this, the startup and shutdown operations are successfully performed by skilled human operators who are able to utilize a great wealth of knowledge and past experiences. In this paper, in order to automate the shutdown operation of power plants, it is proposed that the operation be performed by a more efficient method than the current used, through dividing the total process into several subprocesses, introducing checkpoints, and using fuzzy logic. For this, fuzzy logic controllers and fuzzy decisionmakers are designed and the validity is shown by simulation via a set of piecewise continuous shutdown models.

In this study, it is developed the interactive DNC(Direct Numerical Control) system, in using RS232C cable and auxiliary computer, through the diagnosis of planning process and information evaluation. this DNC system recognize the Manufacturing planning and control it. This DNC system has a different notation. It can be done by an operator who hasn't knowledge about personnel computer. It is operated with automatic planning and measurement tec. by operator, using part program on the NC(Numerical Control).

This paper presents a methodology for automatic feature extraction used in a vision system of FMC (flexible Manufacturing Cell). To implement a robot vision system, it is important to make a feature database for object recognition, location, and orientation. For industrial applications, it is necessary to extract feature information from CAD database since the detail information about an object is described in CAD data. Generally, CAD description is three dimensional information but single image data from camera is two dimensional information. Because of this dimensiional difference, many problems arise. Our primary concern in this study is to convert three dimensional data into two dimensional data and to extract some features from them and store them into the feature database. Secondary concern is to construct feature selecting system that can be used for part recognition in a given set of objects.

This paper presents a comprehensive methodology of automatic feature recognition for rotational parts. The parts of interest are rotational without deviation. In stage one, IGES formatted daa of 2D engineering drawings are transformed into three independent graphs through upperhalf profile. In stage two, recursive evaluations are made to recognize features and relevant information is acquired. Finally,an example is evaluated to demonstrate the recognition power of the developed algorithm for rotational features.

The manual procedure of drafting engineering design is fast becoming obsolete. One of the important areas in CAPP is to automate tolerance assignment which is encountered in mechanical part design. This paper presents an approach for automating tolerance assignment using CAD database and tolerancing database in CAD drawing. The system is consisted of four major functions feasture extraction, feaure inferencing, rulebased tolerance allocation, and automatic upating. Auto CAD R.II is employed as CAD system and a computer progran is developed by using Auto LISP on PC386.

We deals with stationary layout control system of AGV. It is a intelligent contorl system to be wholly charged control PC a layout information and guided command and to be controlled a vehicle driving, steening, safety of natural functions of AGV. Fieldbus concentrator of stationary layout control system serves control command from control PC and status information of AGV. Telegram software monitors transmitted command and status information through IR (Infrared) modem. Epecially it is possible to easily network to use not an exclusive controller of AGV but

In this paper we investigate the connection between the continuoustime FIR(finite impulse response) filter and the corresponding discrtettime FIR filter with fast sampling. The interconnection is established by formulating the discretetime case using delta operators which has superior numerical properties in discretizing prcedure. The aim of this paper is to present a unified FIR filter using the .delta.operator and to show that, as sampling interval .DELYA. aperator to zero, the results of this filter converge to the corresponding continuoustime reuslts, which implies that the unified FIR filter unifies continoustime FIR filter and discrtettime FIR filter.

Resently there have been increasing interests in adaptive identification and control of flexible structures. In this paper, vector channel lattice filters and their applications to parameter identification of flexible structures are studied. Numerical examples are given to show its performace to estimate the natural frequencies of 5mass system. It is observed that vector channel lattice filter convetges quickly and identifies modal frequencies even when some of them is unobservable for some measurements. Experimental results demonstrated the ability of the lattice filter to identify the natural frequencies and the damping ratios of cantilever beam and pipe.

ARMA fast Transversal filter(FTF) algorithm solves the extended least squres estimation problems in a very efficient way. But unfortunately, it exhibits a very unstable behavior, due to the accumulation of roundoff errors. So, in this paper, two effective method to stabilize ARMA FTF algorithm is proposed. They are based on the analysis of the propagation of the numerical errors according to a first order linear model. The proposed methods modify the numerical properties of the variables responsible for the numerical instability, while proeserving the theoretical form of the algorithm. The proposed algorithms still have the nice complexity properties of the original algorithm, but have a much more stable brhavior.

VSS identification approach is based on following concept, i.e. while in sliding motion, the switching of control inputs refects system uncertainites. Therefore, if there exist some operations that make the information form the switiching control inputs be achievable, then the unknown parameters can be actually identification mechanisms which can fully make use of the available information. Two different types of VSS identifiers are taken into consideration. The first type uses adjustable model whose structure is similar to that of identified systems. From the viewpoint of contro, this type of VSS identifiers may be regraded as direct identifier vecause the identified system is handled as an open loop. On the other hand, if the identified system is controlable in the sense of VSS(sliding mode can be generated through chosing control inputs), the second type of VSS identifier, the indirect VSS identifier, can be constructed according to the linerized system strucutre while staying in sliding mode. Therefroe it can be applied to some nonlinear systems which are not linear in parametric space by general identification algorithms, whereas linear in parametric space when sliding mode is existed.

In operational pont of view, the batch reactor is quite different from the continuous reactor in that it is completely in unsteady states. This makes process variables swing over wide ranges and the process disclose its nonlinerarities. For the most part these nonlinearities are due to reaction heat. Accordingly it is very important to know the informations of reaction heat. This paper presents the method of reaction heat measurement in industrial batch reator which furnishes the limited measurable points. The informations of reaction heat are utilized in modeling of the batch reactor as well as deciding the stability and control variables.

A costsaving feeding rate contro method for a batch reaction process is proposed by reconfigurating the flow rate control loop where the feeding rate is estimated by differentiating weight measurement. While the existing control methods require either a flow rate measurement device or a flow rate estimator, the proposed method is devised not to require neither of these facilities by directly controlling the tank weight with a timevarying set point. Experimental evaluation in a pilotscale unit shows that the proposed method can significantly reduce the actuator hunting of the existing method where the flow rate is estimated by differentiating the weight measurement.

Three different reflux policies are compared for a batch distillation process in which a fixed recovery with a given average purity of the distillate is required ; the first, for the constant distillate purity ; the second, for the constant reflux ratio ; finally, for the optimal reflux policy which gives the minimum operation time. The optimal reflux policy was obtained using pontryagin's maximum principle. Througy the numerical simulations for the three different binary mixtures, it was found that the time advantage of the optimal reflux operation over the constant overhead composition operation varies form 10.0 to 22.4% and the advantage over the constant reflux operation varies from 1106 to 36.6% in the three cases considered.

A batch reaction calorimeter is the device for estimating reaction heat along time. Reaction heat formation data are very imoprtant for batch reactor behavior analysis and operatoin. So, many types of reaction calorimeter have been introduced and used. In this study, we analyze the dynamic charateristics of the batch reaction calorimeter that is designed in our laboratory. And we introducde suitable temperature control system. As a result, we analyze reatin heat formation data obtained by simulation and introduce its application examples.

This paper deals with the solution of inverse kinematics of the industrial FANUC robot with IBM PC386. The inverse kinematics of FANUC robot cannot be solved by the algebraical method, because arm matirix T
$_{6}$ is very complex and 6joint angles are associated with the position and the approach of endeffector. Instead we fuund otehr 5joint angle by and algebraical method after finding .THETA.$_{1}$ value by a numerical method.d. 
This study treats the method for kinematic modeling of the biped walking robot, for synthesizing various gait trajectories, and for calculating adequate values of the joint torque inside the stable region. To synthesize various and anthropomorphic walking easily, the gait trajectory is specified by a set of ten walking prameters, and the trunk motion equation is derived by the zero moment point and the gait trajectory. By distributing ground reaction force and moment reduced at the zero moment point to the both feet, the joint torque equation can be derived readily, and according to this equation, the joint torque to stable walking can be computed.

A dynamic modeling and analysis for the Stewart platform type of parallel robot is addressed. The dynamic modeling is performed based on the method of Kinematic Influence Coefficients(KIC) and transfering of the generalized coordinates. The optimum geometric configurations of the system that minimize the actuating forces at the linear actuator are found for several trajectories by using the optimization technique.

The mobile robot may be composed of a robot and an AGV. In this paper, the ynamics of a robot part, such as reaction forces or required torques, are studied with assumption of the AGV in motion. These forces of a mobile robot are different from those of an industrial robot. A new robot is deigned and manufactured for the mobile robot.

The purpose of this paper is to perform the engine driving well with controling the IAC(Idle Air Control) valve which controls the demandair when the abrput increasing or decrasing and the idling, the fuel by controlling the injector, and the spark timing. This plant is the complex system because this should be controlled each other and each one affects other. We have controlled this system effectively by using Personal Computer in order to reduce the exhaust gas and improve the engine performance.

In order to implement the digital PID control algorithm, it is necessary to consider the effect of the finite word length(FWL). In this paper, we show the FWL effect in the digital PID controllers. The conception analyse the effects of the signal quantization error in the digital PID algorithm and the coefficient wordlength determined from performance criteria with the statistical wordlength concept. Throughout this paper, it is dealt with the type of controller structure based delta operator the delta operator has such advantages are superior rounfoff noise perfoff noise performance, more accurate coefficient repersentation, and less sensitive control law.

In recent years, induction motor is applied for several industrial actuatung parts instead of direct current motor because of the robust construction, nonexpensive and maintenancefree actuator etc. and having capability of speed control according to development of power electrounics and microprocessor techniques. In the paper, a microprocessorbased digital control approach for spped control of induction motor system is presented by considering a simple modelling equation as the system expression equation of induction motor and using the self tuning control and torque effdforward control method. As the model equation of the induction motor system, we use a second order differential equation which is well known in the modeling equation is induced form the control theory stand point such tath we can describe usually the motor system connected by inverter, generator and load etc. The effectiveness of the control system composed by the above mentioned design concept is illustrated by the expermental result in the presence of step reference change and generator load variation.

The paper treated the oscillation control of cargo rope and the position control of cart for a tower crane system by adopting the optimal regulator method. Since the tower crane has nonlinearity and it is very unstable when the cargo is mobiling, an insensitive control system with respect to oscillation of cargo rope and disturbance such as wind is required. In this paper, model equation of the tower crane is induced by using Lagrange equation and it is linearized at equilibrium point. The real time control of tower crane syste is implemented by 16bits microcomputer with A/D and D/A convetters to illustrate the application of the adopted optimal desgn method.

In the paper, a hybrid control approach for the swingup control of a rotary type inverted pendulum is treated using onechip microcomputer. The control approach is composed by a scheduling logic control for swing up control and the linear state feedback control to achieve the disired invertedstate of the pendulum. The experimental cystem has been implemented by a 16bit onechip microcomputer with 3096 opu as the digital controller incorporating the above mentioned control approach.

In this paper, we present a solution to the PID tuning problem by optimizing a GPC(General Predictive Control) criterion. The PID structure is ensured by constraning the parameters to a feasible set defined by the discretetime Euler approximation of the ideal continuoustime PID controller. The algorithm is ectended by incorporating heuristic rules for selection of the significant design parameters. The algorithm has been successfully tested and some results are prewented.

본 논문에서는 다중 로보느의 협조제어(Coordinated Control)를 위한 로보트 콘트롤러의 설계에 대해서 연구한다. 첫 부분에서는 다중 로보느의 연구배경 및 연구동기에 대해서 논의하고 이어서 Coordinated Task를 묘사하기 위한 Programming Primiitive Set을 정의하며 구현에 대해서도 논의한다. 특히 Motopn Primitive는 synchronous(Coordinated Motion), Asynchronous Motion, Conditional Motion, 특수 Motion으로 분류하고, 각각의 궤적계획 및 구현에 대해서도 간단히 논의한다. 특히 본 논문에서는 외부의 변화하는 환경에 효과적으로 적응할 수 있게 하기 위하여 Vision센서, Encoder신호와 Limit센서, Force센서 등의 다양한 외부 센서를 융합 처리할수 있는 다중 로보트 제어 시스템을 개발하였다.

This paper proposed LM method for solving binpicking problem in robot vision. It has the processing steps such as image enchancement, image thresholding, region labelling, and moment computation. To determine a target object form bined objects, the modified labelling method is used. To determine position and orientation of holdsite, the moment method is used. Finally, some experiment results are illustated and compared with the results of conventional shrinking algorithm. The proposed LM method has reduced processing time.

In this paper we proposed the systematic method of reducing the order of controller with robustness. State space formulae for all controllers is found by solving two coupled Jlossless coprime factorizations and model reduction problem. To reduce the order of controller, balanced truncation and Hankel approximation are used.

This paper deals with design method of the model following servo system in which optimal regulator probelm is used to design the controllers that make the step/ramp response of the plant be keptclose to a specified ideal step/ramp response of the model. The characteristics of this system is robust in the presence of the specified disturbances or the partameter perturbations of the plant. Especially, by direct feedforward compensator from the reference input the steady state offset of plant output response is excluded and the transient response is improved. Examples are give and the results of the design of the model follwing servo systems are verified by the computer simulation.

Loop shaping techniques are developed for the LQG/LTR controller design of singular multivariable sytems. One approach is to use the mode form of plant and the other is to replace the eigenvalues at 0 by ones at .epsilon.(.rarw.0). These two concepts for the target filter loop design are applied to a flight autopilot. And it is shown that these techniques are effective ones for the desired loopshaping of singular multivariable systems.

Time Delay Controller(TDC) is a model following controller which uses input and output values and state variables to estimate additional quantity of dynamics due to external disturbances and/or model parameters variation at some past instant. TDC is very robust against parametric uncertainty whil it is not robust against unmodeled dynamics even showing instability. To solve this problem a stability anlysis is performed and a compensation technique using reduced order observer, AntiFiltering Compensator(AFC), is proposed for a case in which the high order kinown dynamics is deliberately ignored. If the ignored dynamics causes instability of the TDC control system, AFC is shown to be indispensible fot a stable implementation of TDC.

This paper presents an algebraic approach for finding a dynamic state feedback controller when the linear multiinput system with delays in both state and input is controllable. In the timedelay case, controllability of the system does not always imply that system is cyclizable. Therefore, reduced order augmentation systems which is cyclizable as the timevarying case are considered. It is possible to construct feedback contorl systems by using singleinput methods.

A new sliding surface for a viaraible structure control(VSC) law is emplyed to achieve fast and robust path tracking in a class of secondorder nonlinear unceratin dynamical systems. The surface onitialy passes arbitrarily given initial conditions and subsequently moves towards a predetermined surface via rotaiting or/and shifting. We call it as a moving sliding surface(MSS). The surface is then incorporated with the VSC law which is constructed by imposing the sliding conditions in a special way. We primarily enforce the conditions which assume that once the system state is on a sliding surface that it is driven towards the zero state. Using the VSC law associatied with the MSS, it is shown that the tracking behavoirs are remarkably improved in the sene of the fastness and the robustness.

Two types of digital repetitive control systems are analyzed and designed to reduce the error spectrum including not only harmonic but also nonharmonic components. First, a novel gain scheduling algorithm is suggested for conventional and modified repetitive controller is scheduled to reduce the infinite norm of error in frequency domain. For this, the relative error transfer function is mewly defined as the ratio of the error spectrum for the system with a repetitive controller to the error spectrum for the system with a repetitive controller to the error spectrum for the system without a repetitive controller. Secondly, as an alternative of a repetitive control system with the gain scheduling, a repetitive control system with higher order repetitve function is analyzed and designed, where instead of equal weightings, weightings of the higher order repetitive function is determined in such a way that the infinite norm of relative error transfer function is minimized. To show the validities of proposed methods, computer simulation results are illustrated for a typical disk drive head positioning servo system.

One of the undesirable nonlinear phenomenon called 'wind up' occours when the integrator in the controller and the saturated actuator interact. Large overshoot, slow response, instability, limit cycle and jump resonance are the characteristics of wind up phenomenon. Several 'antiwindup' compensators have been developed to prevent some of the aforementioned nonlinear characteristics such as instabilituy and limit cycle, but none has studied the effect of antiwindup compensator on the jump resonance. In this paper, we developed an analyitcal method to design the compensator to prevent not only limit cycle but also jump resonance. An illustrative example is included to show the compensator eliminates jump resonance of effectively.

A new scheme of OFVSC(Output Feedvack Variable Structure Controller) is proposed for the servo control system. The main structure of proposed control scheme is composed of servo compensator and dynamic switiching function. By the use of dynamic switiching function the assumption of full state availability can be removed and the disturbances which does not satisfy the matching condition cna be rejected. And the servo compensator which is designed for each output variable the robustness for the all type of disturbances. And the performances of proposed control system are evaluated through simulation studies for a numerical example.

In ths paper a new active assembly algorithm for chamferless precision parts mating, is considered. The successful assembly task requires an extremely high position accuracy and a good knowledge of mating parts. However, conventional assembly mehtod alone makes it difficult to achieve satisfactory assembly performance because of the complexity and the uncertainties of the process and its environments such as imperfect knowledge of the parts being assembled as well as the limitation of the devices performing the assebled as well as the limitation of the devices performing the assembly. To cope with these problems, a selflearning rulebased assembly algorithm is proposed by intergaring fuzzy set theory and neural network. In this algortihm, fuzzy set theory copes with the complexity and the uncertainties of the assembly process, while neural network enhances the assembly schemen so as to learn fuzzy rules form experience and adapt to changes in environment of uncertainty and imprecision. The performance of the proposed assembly algorithm is evaluated through a series of experiments. The results show that the selflearning fuzzy assembly scheme can be effecitively applied to chamferless precision parts mating.

Extracting the process planning information from the CAD data is the key issue in integrated CAD/CAM system. In this paper, we develop algorithms for extracting the shape and setup configuration for NC machining of prismatic parts. In determining the workpart shape, the minimumenclosing condept is applied so that the material waste is minimized. To minimize the number of setups, feature based algorithm is developed considrint the part shape, tool shape, and tool approach direction. The validity and effectiveness of the developed algorithms were tested by computer simulations.

Process planning is a key feature that an intelligent CAD/CAM system should possess. In this paper, we address the problem of determinig process configuration to manufacture the free surfaces. For analyzing the surface, the method of surface subdivision is used. The developed algorithm evaluates the processability of the given surface by applying the three/four/fiveaxes procedures sequentially. To illustrate and test the developed algorithm, numerical simulations are presented.

This paper describes the systematic control method of process information transfer and machine cell control in FMS implementation. We have constructed an experimental FMS computer network and control system. The system hardware consists of host computer to manage process data and information transfer of machine cells, cell control computers to control machine cells(NC lathe, machining center). On the other hand, software is made up of oredr management module, NC program searching and generation module, NC part program error check module and cell control module. In this study, we could arrive at conclusion as following : The first, each task could be accomplihed by the efficient information transfer in hierachical computer network. The second, data base system of part programs and process control data is needed for the efficint information transfer and production management. Lastly, expansion of FMS control system could be achieved by the hierachical and decentralized computer control system.

In order tominimize fixing error of workpieces, circle, prismatic, sphere, cylindrical and sculptures types. Modification Rule by Indexing Table and Modification Rule by NC Program are developed for machining centers by using touch trigger probes. The Modification Rule by Indexing Table meas the alignment of workpiece to NC program through degree of freedoms of indexing table. The Modification Rule by NC Program is the alognment of NC program to workpiece stetp condition via the generation of NC progarm. A postprocessing module is alos developed for generating NCpart program(User Macro) to compensate for machining errors in end milling and boring processes. Developed method are verified by experiments.

Existing CAD systems do nto provide the advanced function for systematic checking of design and drafting errors in mechanical drawings. This paper describes a method for sytematic checking of global parts in mechanical drawings. The checking items are deficiency and redundancy of dimensions, inputerrors in dimension figures and symbols, etc. Checking for deficiency and redundancy of global dimensions has been performed applying Graph Theory. This system has been applied to several examples and we have confirmed the feasibility of this checing method.

Modern chemical processes are becoming more complicated. The sophisticated chemical processes have needed the fault diagnosis pxpert systems that can detect and diagnose the fault diagnosis expert systems that can detect and diagnose the faults of some processes and give and advice to the operator in the event of process faults. We present the TimeDelay Neural Network(TDNN) approach for online fautl diagnosis. The online fault diagnosis system finds the exact origin of the fault of which the symptom is propagated continuously with time. The proposed method has been applied to a pilot distillation column to show the merits and applicability of the TDNN.