제어로봇시스템학회:학술대회논문집
Institute of Control, Robotics and Systems (ICROS)
 기타
Domain
 Machinery ＞ Robot/Automated Machinery
1998.10

Robots utilized in the field of welfare or agriculture should be light in weight and flexible in structure. A pneumatic actuator has properties such that it is more powerful compared with a motor of same weight, and that it is flexible, clean and unexplosive. In this paper we propose a new structure of the pneumatic actuator with twodegreeoffreedom. By using proposed pneumatic actuators, we can easily construct multidegreeoffreedom pneumatic manipulators. Here we constructed a fourteendegreeoffreedom pneumatic dual manipulator. The performance of the dual manipulators is confirmed through experiments for ballhandling with impedance control. In the experiments several control schemes, including the decentralized control and the simple adaptive control (SAC), were used. The results show that a flexibility of the pneumatic actuator is appropriate to accomplish the coordinative motion of the right and left arms of the robot.

A tracking controller is presented for RLFJ(rigid link flexible joint) robot manipulators with only position measurements. The controller is developed based on the integrator backstepping design method and on the two observers: the first is simple linear form observer for the filtered link velocity errors and the other for the actuator velocities. The proposed controller achieves exponential tracking of link positions and velocities while keeping all internal signals bounded. It also guarantees exponential convergence of the estimated signals to their actual ones. Finally, simulation results are included to demonstrate the tracking performance.

In this paper, a target approachable forceguided control with adaptive accommodation for the complex assembly is presented. The complex assembly (CA) is defined as a task which deals with complex shaped parts including concavity or whose environment is so complex that unexpected contacts occur frequently during insertion. CA tasks are encountered frequently in the field of the manufacturing automation and various robot applications. To make CA successful, both the bounded wrench condition and the target approachability condition should be satisfied simultaneously during insertion. By applying the convex optimization technique, an optimum target approaching twist can be determined at each instantaneous contact state as a global minimum solution. Incorporated with an admissible perturbation method, a new CA algorithm using only the sensed resultant wrench and the target twist is developed without motion planning nor contact analysis which requires the geometry of the part and the environment. Finally, a VMEbus based realtime control system is built to experiment various CA task. Tinsertion task as a planar CA and doublepeg assembly task as a spacial assembly were successfully executed by implementing the new forceguided control with adaptive accommodation.

This paper proposes a new compliant contact control strategy for the robot manipulators accidentally interacting with an unknown environment. The main features of the proposed method are summarized as follows: First, each entry in the diagonal stiffness matrix corresponding to the task coordinate in Cartesian space is adaptively adjusted during contact along the corresponding axis based on the contact force with its environment. Second, it can be used for both unconstrained and constrained motions without any switching mechanism which often causes undesirable instability and/or vibrational motion of the end effector. Third, the adjusted stiffness gains are automatically recovered to initially specified stiffness gains when the task is changed from constrained motion to unconstrained motion. The simulation results show the effectiveness of the proposed method by employing a twolink direct drive manipulator interacting with an unknown environment.

Mizoguchi, Hiroshi;Hidai, KenIchi;Goto, Yoshiyasu;Teshiba, Masashi;Shigehara, Takaomil;Mishima, Taketoshi 26
This paper proposes a novel method to efficiently develop GUI based control software for a legged mobile robot. Although GUI is convenient it is a very burden to both a computer and its developer. In case of the mobile robot, these problems are more serious. The proposed method solves these problems by separating GUI from control software. An implementation based upon the proposed method demonstrates its effectiveness. 
Path Planning is one of the important fields in robot technologies. Local path planning may be done in online modes while recognizing an environment of robot by itself. In dynamic environments to obtain fluent information for environments vision system as a sensing equipment is a one of the most necessary devices for safe and effective guidance of robots. If there is a predictor that tells what future sensing outputs will be, robot can respond to anticipated environmental changes in advance. The tracking of obstacles has a deep relationship to the prediction for safe navigation. We tried to deal with active contours, that is snakes, to find out the possibilities of stable tracking of objects in image plane. Snakes are defined based on energy functions, and can be deformed to a certain contour form which would converge to the minimum energy states by the forces produced from energy differences. By using point algorithm we could have more speedy convergence time because the Brent's method gives the solution to find the local minima fast. The snake algorithm may be applied to sequential image frames to track objects in the images by these characteristics of speedy convergence and robust edge detection ability.

A nonlinear attitude model of a satellite with thrusters, magnetic torquers and a reaction wheel cluster is developed. Then the linearized version of this satellite attitude model is derived far the attitude hold mode. For comparison purpose, various control methods are considered for attitude control of a satellite. We consider a proportional derivative controller which is actually used in the remote sensing satellite, KOMPSAT. Then a comparison is made with an H
$_2$ controller, an H$\sub$ $\infty$ / controller, and a mixed H$_2$ / H$\sub$ $\infty$ / controller. The analysis and numerical studies show that the proportional derivative controller's performance is limited in the sense that the pitch angle cannot approach zero. The simulations also show that among three control methods (H$_2$ control, H$\sub$ $\infty$ / control, and mixed H$_2$ / H$\sub$ $\infty$ / control) H$_2$ control has the fastest response time, H$\sub$ $\infty$ / control has the slowest and mixed H$_2$ / H$\sub$ $\infty$ / control comes in between the first two control methods. On the other hand, H$\sub$ $\infty$ / control used least amount of control effort while H$_2$ control required the most. 
In this paper, a stochastic approach based on a Monte Carlo simulation method for the design of a guidance and control (G & C) system of an automatic landing flight experiment (ALFLEX) vehicle is presented. The aim of this study is to design a G & C system robust against uncertainties in the vehicular dynamics. In this study, uncertain parameters and disturbances are treated as random variables in the Monte Carlo simulation. Then, some controller gains in the G & C system are tuned to satisfy conditions concerning the states at touchdown. The proposed method was applied to the ALFLEX vehicle. The simulation results shored the effectiveness of the present approach.

This study deals with guidance and control laws for an optimal reentry trajectory of a vertical landing reusable launch vehicle (RLV) in the future. First, a guidance law is designed to create the reference trajectory which minimizes propellant consumption. Then, a nonlinear feedback controller based on a linear quadratic regulator is designed to make the vehicle follow the predetermined reference trajectory, The proposed method is simulated for the first stage of the HII scale rocket.

Star tracker placement configuration is proposed and the properness of the placement configuration is verified for star tracker's sun avoidance angle requirement. Precision attitude determination system is successfully designed using a gyrostar tracker inertial reference system for a candidate LEO spacecraft. Elaborate kalman filter formulation for a spacecraft is proposed for covariance analysis. The covariance analysis is performed to verify the capability of the proposed attitude determination system. The analysis results show that the attitude determination error and drift rate error are good enough to satisfy the mission of a candidate spacecraft.

B.Widrow established fundamental relations between the leastmeansquare (LMS) algorithm and the digital Fourier transform［1］. By extending these relations, we proposed the short time spectra analysis system using the LMS algorithm［2]. In that paper, we used the normal LMS algorithm on the thought of dealing with only real analytical signal. This algorithm minimizes the real meansquare by recursively altering the complex weight vector at each sampling instant. But, the short time spectra analysis sometimes deals with the complex signal that is outputted from complex analog filter. So, in order to optimize and develop this methods, furthermore it is necessary to derive an algorithm for the complex analytical signal. In this paper, we first discuss the new adaptive system for the spectra analysis using the complex LMS algorithm and then derive convergence condition, time constant of coefficient adjustment and frequency resolution by extending the discussion. Finally, the effectiveness of the proposed method is experimentally demonstrated by applying it to the measurement of transfer performance on complex analog filter.

This paper presents a lip print recognition by the pattern kernels for a personal identification. A lip print recognition is developed less than the other physical attributes of a fingerprint, a voice pattern, a retinal blood/vessel pattern, or a facial recognition. A new method is proposed to recognize a lip print bi the pattern kernels. The pattern kernels are a function consisted of some local lip print pattern masks. This function converts the information on a lip print into the digital data. The recognition in the multiresolution system is more reliable than recognition in the singleresolution system. The results show that the proposed algorithm by the multiresolution architecture can be efficiently realized.

This paper describes a method for recognition of twodimensional position of an object by use of aerial ultrasonic sensor and signal processing technique, which would become a help for blind person or selfmobile robot. First, we have developed a method for measuring the time difference between the transmitted and the received burst wave by use of one ultrasonic transmitter and three receivers. Secondly, a new method is developed for measuring the distance to an object by use of Msequence correlation method. Thirdly, a measurement method to obtain the position of an object is described by use of phasearrayed ultrasonic sensor, which gives us a widerange position determination in a short time.

This paper proposes a novel computer human interface, named Virtual Wireless Microphone (VWM), which utilizes computer vision and signal processing. It integrates realtime face tracking and sound signal processing. VWM is intended to be used as a speech signal input method for human computer interaction, especially for autonomous intelligent agent that interacts with humans like as digital secretary. Utilizing VWM, the agent can clearly listen human master's voice remotely as if a wireless microphone was put just in front of the master.

For ANC systems applied to aircrafts or passenger ships, engines from which reference signals are usually measured are located so far from seats where main part of controllers are placed. It can make feedforward ANC scheme difficult to implement or very costly. Feedback ANC algorithms which do not require reference signals and use error signals alone to update the filter, are usually sensitive to measurement noise ' and impulse noise. In this paper, reference signal needed for the feedforward control is not measured directly but generated with the estimated frequencies. Cascade adaptive notch filter (ANF), which has the low computational burden, is used to implement ANC system in real time. Several ANFs of order 2 are connected in series to estimate multiple sinusoids. Computer simulations and experiments in the laboratory for verifying efficacy of the proposed algorithm are carried out.

We have developed an eigenvalue method for impedance computed tomography to improve the illconditioning problem. We have compared the performance of this method and the balancing method by computer simulations. As a result, it was proved that this method is better than the balancing method very much. It was found that the initial value condition is not so severe to obtain good images.

In this paper the authors propose a model about interaction of inner modules of autonomous robot which is possible to team walking action without external and explicit supervisor signal. A main feature of the model is that completed and fixed module for estimating robot's motion parameter by utilizing binocular parallax can be a supervisor for the module to team the walking action.

We introduce an autonomous flying system using a modelhelicopter. A feature of the helicopter is that autonomous flight is realized on the lowcost compact modelhelicopter. Our helicopter system is divided into two parts. One is on the helicopter, and the other is on the land. The helicopter is loaded with a vision sensor and an electronic compass including a tilt sensor. The control system on the land monitors the helicopter movement and controls. We firstly introduce the configuration of our helicopter system with a vision sensor and an electronic compass. To determine the 3D position and posture of helicopter, a technique of image recognition using a monocular image is described based on the idea of the sensor fusion of vision and electronic compass. Finally, we show an experiment result, which we obtained in the hovering. The result shows the effectiveness of our system in the compact modelhelicopter.

This paper describes the realtime implementation of an adaptive controller fur the robotic manipulator. Digital signal processors(DSPs) are special purpose microprocessors that are particularly powerful for intensive numerical computations involving sums and products of variables. TMS320C50 chips are used in implementing real time adaptive control algorithms to provide an enhanced motion for robotic manipulators. In the proposed scheme, adaptation laws are derived from the improved Lyapunov second stability analysis based on the direct adaptive control theory. The adaptive controller consists of an adaptive feedforward controller and feedback controller. The proposed control scheme is simple in structure, fast in computation, and suitable for realtime control. Moreover, this scheme does not require any accurate dynamic modeling, nor values of manipulator parameters and payload. Performance of the adaptive controller is illustrated by simulation and experimental results for a assembling robot.

This paper proposes a graphical user interface for industrial robot systems. Previous user interfaces for industrial robot systems were based on the text. In order to enable operators to handle robots more efficiently, a set of graphical tools is provided. The graphical tools contain a control panel for operating robots and compiling robot programs, a graphical teaching panel for handling virtual robots and a graphical monitoring panel for checking robot status. Furthermore, the proposed GUI can be used to operate remote robots because it has network utilities. This system consists of the virtual mode and the real mode. The user can handle a 3D virtual solid model of the robot in the virtual mode and an actual robot in the real mode.

In this study, we try to coincide virtual robot system in an OLP(offline programming) with actual robot system even though kinematic differences between them are made. The virtual robot in the OLP may be modeled according to kinematics of the actual robot system. However, it is a complicated problem to find exactly all kinematic parameters of actual robot and environment. In this paper, an automated calibration method is proposed In order to find some kinematical parameters which are necessary for the modeling of a robot and environment in the OLP. It is applicable to SCARA robot for assembly task. In this method, a wellmarked worktable of environment Is regarded as reference coordinate frame. The robot detects some marks on the worktable through sensors attached to the endeffector. The necessary parameters are calculated from the data of the robot joint variables when the robot detects the mark. The model in the OLP is modified by the parameters.

The recent developments and studies in the framework of output tracking control in the field of robotics that has been studied in the Control Laboratory, are presented. An output controller based on“HardwarelntheLoop Simulation”(HILS) and“Rapid Control Prototyping”(RCP) concepts is developed using dSPACE. These new concepts are shown to be particularly beneficial for manipulator control tasks. In the Elbow manipulator design, there are two kinds of manipulators, namely the serialdrive type and the parallelogramdrive manipulator, The objective of this research is to model the two Elbow manipulators and to implement the proposed controller for manipulator applications. The control goal is to force the manipulator to follow a given trajectory in the given work space. Output controllers of the two elbow manipulators that are based on the model matching control approach have been implemented in two models that represent the robot equations of motion. To reduce the efforts in evaluating the proposed algorithm, a new system configuration method, based on HILS and RCP tools, was suggested to determine the parameters of the integrated dynamic system.

This paper presents time domain identification of an interval system. We conjectured that Markov parameters (Pulse Responses) from Kharitonov plants would envelope those of the whole interval system. The examination on interrelations between Markov parameters from Kharitonov plants of an interval system and those of the whole interval system determines the validity of the conjecture and is used to give some extremal properties of Markov parameters. The results of this paper are shown in simulations on MBC500 Magnetic Bearing System and a given interval system.

In this paper, a simple method is presented to synthesize a transfer function from experimentally obtained gain and phase data. The method we offer here is based on the previous method given by M.Hassul etc. [1], where they proposed relevant formulas in a straightforward manner so that undergraduate students could follow the development more easily. This method, however, inevitably is accompanied by a significant difference between the real and identified model especially in the low frequency region. We solve this problem by introducing a new weighting function that can be determined by using the additive uncertainty of the Identified transfer function.

The concept of dissipativity and passivity are of interest to us from a theoretical as well as a practical point of view. It is well known that the Riccati equation is derived from the dissipation inequality which expresses the fact that the system is dissipative; the energy stored inside the system doesn't exceed the amount of supply which flows into the system. The pencil model is regarded as a representation based on behavioral approach introduced by J.C. Willems. It has first order in the internal variable and zeroth order in the external variable. In general, any matrix pencil is transformed into a canonical form which is consist of several kind of subpencils, One of them has row full rank for
$^\forall S\;\in\;\mathds{C}\;\bigcup{\infty}$ , we call it underdetermined mode of the model. In our opinion, most important properties of dynamical system lay in the mode. According to the properties of canonical form for pencil, it is shown that the storage function which characterizes the dissipativity of the system can be written as a LMI for the underdetermined mode, if the system doesn't include impulse mode. 
This paper presents a scheme to actively control the vertical vibration of aerial vehicles due to the disturbances such as the sudden change of derricking angle and the external forces by using a small plunger attached to the derricking cylinder. Simulations show that the 1st mode vibration is suppressed efficiently by the proposed method without exciting the higher modes' vibration. Detailed mathematical model of the aerial vehicle, its vibration characteristics, detection method of the 1st mode vibration and the controller design based on the lagelement and the disturbance observer are described.

The report presents basic functions of Timedelay System Toolbox (for MATLAB) the generalpurpose software package for Computer Aided Design of control systems with delays. The Toolbox is a collection of algorithms, expressed mostly in mfiles for simulating and analysis of MIMO linear and nonlinear systems with discrete and distributed (timevarying) delays.

In this paper, we present new control method for robot manipulators. The design objective can be the implementation of minimax controller with H
$_{\infty}$ performance via LMI approach to guarantee the robustness and to obtain the exact tracking performance for robot manipulators with system parameter uncertainty and exogenous disturbance. We show that the Algebraic Riccati equation (ARE) which is needed for the construction of H$_{\infty}$ controller can be recast into the Algebraic Riccati Inequality (ARI) and the optimal control gain can be obtained by convex optimization method. Then, we will apply the proposed controller to rigid robot manipulators for verifying the performance of our controller. 
We argue that the combination of optimal control synthesis and QFT tuning enables design of controllers with levels of performance that surpasses what can be achieved using only a single technique. Using a constructive example, we demonstrate how the strength of each technique is utilized to arrive at a particularly desired controller in terms of tradeoffs between performance and controller complexity.

In this paper, an approach to the reduced order H
$_{\infty}$ controller synthesis is proposed. This approach employs the frequency weighted model reduction whose frequency weights are deduced from the closedloop system regarding the controller order reduction errors as uncertainties in a plant, while the resultant reduced order H$_{\infty}$ controller guarantees prescribed H$_{\infty}$ control performances. 
The physical parameters of controlled systems are uncertain and are accompanied with nonlinearity. The transfer function of the controlled system should, therefore, be expressed by interval polynomials. This paper describes the realization of robust performance for that type of control system (interval system) via model reference feedback. First, we will analyze an invariance problem of dynamic characteristics such that the dominant roots do not break away from a specified circular area, and will present a discrimination algorithm (i.e., a division algorithm) for the extreme points of the uncertain coefficients. Then, we will present a design method of control systems which have a robust performance such that the location of the dominant roots dose not vary excessively.

The strategy presented in this paper is based on modifying the past patterens and adjusting the content of the driving patterns by a new algorithm. Learning happens during the driving procedure of a mobile vehicle. The purpose of this paper is to solve the problem how to realize the hardware neurocomputer by back propagation (BP) neural network learning online.

In this paper, a new hierarchical fuzzy inference system (HFIS) using structured TakagiSugeno type fuzzy inference units(FIUs) is proposed. The proposed HFIS not only solves the rule explosion problem in conventional HFIS, but also overcomes the readability problem caused by the structure where outputs of previous level FIUs are used as input variables directly. Gradient descent algorithm is used for adaptation of fuzzy rules. The ball and beam control is performed in computer simulation to illustrate the performance of the proposed controller.

This paper addresses analysis and design of a fuzzy modelbasedcontroller for the control of uncertain SISO nonlinear systems. In the design procedure, we represent the nonlinear system by using a TakagiSugeno fuzzy model and construct a global fuzzy logic controller via parallel distributed compensation and sliding mode control. Unlike other parallel distributed controllers, this globally stable fuzzy controller is designed without finding a common positive definite matrix for a set of Lyapunov equations, and has good tracking performance. The stability analysis is conducted not for the fuzzy model but for the real underlying nonlinear system. Furthermore, the proposed method can be applied to partially known uncertain nonlinear systems. A numerical simulation is performed for the control of an inverted pendulum, to show the effectiveness and feasibility of the proposed fuzzy control method.

In this paper, we introduce the micro robot soccer playing system as a standard test bench for the study on the multiple agent system. Our method is based on following viewpoints. They are (1) any complex behavior such as cooperation among agents must be completed by sequential basic actions of concerned agents. (2) those basic actions can be well defined, but (3) how to organize those actions in current time point so as to result in a new stale beneficial to the end aim ought to be achieved by a kind of selflearning selforganization strategy.

Since the neural network was introduced, significant progress has been made on data handling and learning algorithms. Currently, the most popular learning algorithm in neural network training is feed forward error backpropagation (FFEBP) algorithm. Aside from the success of the FFEBP algorithm, polynomial neural networks (PNN) learning has been proposed as a new learning method. The PNN learning is a selforganizing process designed to determine an appropriate set of Ivakhnenko polynomials that allow the activation of many neurons to achieve a desired state of activation that mimics a given set of sampled patterns. These neurons are interconnected in such a way that the knowledge is stored in Ivakhnenko coefficients. In this paper, the PNN model has been developed using the plasma enhanced chemical vapor deposition (PECVD) experimental data. To characterize the PECVD process using PNN, SiO
$_2$ films deposited under varying conditions were analyzed using fractional factorial experimental design with three center points. Parameters varied in these experiments included substrate temperature, pressure, RF power, silane flow rate and nitrous oxide flow rate. Approximately five microns of SiO$_2$ were deposited on (100) silicon wafers in a PlasmaTherm 700 series PECVD system at 13.56 MHz. 
This paper presents a new construction method of candidate controllers using Multimodal Neural Network(MNN). To improve a control performance of multiple controller, we construct, candidate controllers which consist of MNN. MNN can learn more complicated function than multilayer neural network. MNN consists of preprocessing module and neural network module. The preprocessing module transforms input signals into spectra which are used as input of the following neural network module. We apply the proposed method to multiple control system which controls the cartpole balancing system and show the effectiveness of the proposed method.

A synthesis of fuzzy variable structure control is proposed to design a highangleofattack flight system for a modification version of the F18 aircraft. The knowledge of the proportional, integral, and derivative control is combined into the fuzzy control that addresses both the highly nonlinear aerodynamic characteristics of elevators and the control limit of thrust vectoring nozzles. A simple gain scheduling method with multilayered fuzzy rules is adopted to obtain an appropriate blend of elevator and thrust vectoring commands in the wide operating range. Improving the computational efficiency, an accelerated kernel for online fuzzy reasoning is also proposed. The resulting control system achieves the good flying quantities during a highangleof attack excursion. Thus the fuzzy logic can afford the control engineer a flexible means of deriving effective control laws in the nonlinear flight regime.

RadioControlled helicopter has superior movement abilities like as hovering or backward move. So it has been used as a sprinkler of agricultural medicines or an observer of dangerous area such as a volcano, etc. But its operation is not simple because it has many control factors and they interfere with each other. Therefore the helicopter is not controlled by simple theory in the case of automatic operation. Then fuzzy sliding mode control, which has fastness, fineness and robustness, is thought to be suitable to satisfy various requirements of the helicopter operation. In this work, the fuzzy sliding mode control was applied to the flying of RC helicopter, As tile result, it was controlled with good performance.

In order to optimize fuzzy model, we use the optimal algorithm with a hybrid type in the identification of premise parameters and standard least square method in the identification of consequence parameters of a fuzzy model. The hybrid optimal identification algorithm is carried out using a genetic algorithm and improved complex method. Also, the performance index with weighting factor is proposed to achieve a balance between the insults of performance for the training and testing data. Several numerical examples are used to evaluate the performance of the proposed model.

This paper is to present a method for recognizing an image of a tracking object by processing the image from a camera, whose attitude is controlled in inertial space with inertial coordinate system. In order to recognize an object, a pseudorandom Marray is attached on the object and it is observed by the camera which is controlled on inertial coordinate basis by inertial stabilization unit. When the attitude of the camera is changed, the observed image of Marray is transformed by use of affine transformation to the image in inertial coordinate system. Taking the crosscorrelation function between the affinetransformed image and the original image, we can recognize the object. As parameters of the attitude of the camera, we used the azimuth angle of camera, which is defected by gyroscope of an inertial sensor, and elevation an91e of camera which is calculated from the gravitational acceleration detected by servo accelerometer.

This paper describes how a priori road geometry and its estimation may be used to detect road boundaries and lane markings in road scene images. We assume flat road and road boundaries and lane markings are all Bertrand curves which have common principal normal vectors. An active contour is used for the detection of road boundary, and we reconstruct its geometric property and make use of it to detect lane markings. Our approach to detect road boundary is based on minimizing energy function including edge related term and geometric constraint term. Lane position is estimated by pixel intensity statistics along the parallel curve shifted properly from boundary of the road. We will show the validity of our algorithm by processing real road images.

This paper describes a new method for estimating timetocollision which exhibits high tolerance to noise contained in camera images. Time to collision (TTC) is one of the most important parameters available from a camera attached to a mobile machine. TTC indirectly stands far the translation speed of the camera and is usually calculated either from successive images or optical flow by using intimate relationship between TTC and flow divergence. In most cases, however, it is not easy to get accurate optical flow, which makes it difficult to calculate TTC. In this study it is proved that if the target has a smooth surface, the average of divergence over any pointsymmetric region on the image is equal to the divergence of the center of the region. It means that required divergence can be calculated by integrating optical flow vectors over a symmetric region. It is expected that in the process of the integration, accidental noise is canceled if they are independent of optical flow and the motion of the camera. Experimental results show that TTC can be estimated regardless of the surface condition. It is also shown that influence of noise is eliminated as the area of integration increases.

This paper presents a new approach to visual servoing with the stereo vision. In order to control the position and orientation of a robot with respect to an object, a new technique is proposed using a binocular stereo vision. The stereo vision enables us to calculate an exact image Jacobian not only at around a desired location but also at the other locations. The suggested technique can guide a robot manipulator to the desired location without giving such priori knowledge as the relative distance to the desired location or the model of an object even if the initial positioning error is large. This paper describes a model of stereo vision and how to generate feedback commands. The performance of the proposed visual servoing system is illustrated by the simulation and experimental results and compared with the case of conventional method fur a SCARA robot.

In this paper, the hybrid state space selftuning control technique Is studied within the framework of fuzzy systems and dualrate sampling control theory. We show that fuzzy modeling techniques can be used to formulate chaotic dynamical systems. Then, we develop the hybrid state space selftuning fuzzy control techniques with dualrate sampling for digital control of chaotic systems. An equivalent fastrate discretetime statespace model of the continuoustime system is constructed by using fuzzy inference systems. To obtain the continuoustime optimal state feedback gains, the constructed discretetime fuzzy system is converted into a continuoustime system. The developed optimal continuoustime control law is then convened into an equivalent slowrate digital control law using the proposed digital redesign method. The proposed technique enables us to systematically and effective]y carry out framework for modeling and control of chaotic systems. The proposed method has been successfully applied for controlling the chaotic trajectories of Chua's circuit.

This paper presents a multimodal neural network composed of a preprocessing module and a multilayer neural network module in order to enhance the nonlinear characteristics of neural network. The former module is based on spectral method using Chebyschev polynomials and transforms input data into spectra. The latter module identifies the system using the spectra generated by the preprocessing module. The omnibus numerical experiments show that the method is applicable to many a nonlinear dynamic system in the real world, and that preprocessing using Chebyschev polynomials reduces the number of neurons required for the multilayer neural network.

A new design methology is proposed to identify the structure and parameters of fuzzy model using PNN and a fuzzy inference method. The PNN is the extended structure of the GMDH(Group Method of Data Handling), and uses several types of polynomials such as linear, quadratic and cubic besides the biquadratic polynomial used in the GMDH. The FPNN(Fuzzy Polynomial Neural Networks) algorithm uses PNN(Polynomial Neural networks) structure and a fuzzy inference method. In the fuzzy inference method, the simplified and regression polynomial inference methods are used. Here a regression polynomial inference is based on consequence of fuzzy rules with a polynomial equations such as linear, quadratic and cubic equation. Each node of the FPNN is defined as fuzzy rules and its structure is a kind of neurofuzzy architecture. In this paper, we will consider a model that combines the advantage of both FPNN and PNN. Also we use the training and testing data set to obtain a balance between the approximation and generalization of process model. Several numerical examples are used to evaluate the performance of the our proposed model.

In this paper, we implemented the neurocomputer called MYNEUPOWER in our research to carry out the artificial neural networks (ANN) calculating. An application software was developed based on a neural network using backpropagation (BP) algorithm under the UNIX platform by the specified computer language named MYPARAL. This neural network model was used as an auxiliary controller in the temperature control of sinter cooler system in steel plant which is a nonlinear system. The neural controller was trained offline using the real inputoutput data as training pairs. We also made the system description of adaptive neural controller on the same temperature control system. We will carry out the whole system simulation to verify the suitability of neural controller in improving the system features.

In this paper we introduce a modeling of wheeled mobile robot with a differential drive derived by R.M. DeSantis and using the dynamics modeling with some disturbance term we control the wheeled mobile robot using fuzzy sliding mode control(FSMC) method. In a fuzzy control approach it is very difficult to prove the stability of the fuzzy controller. Therefore, to overcome that difficult proof of the stability in a fuzzy control method, we first propose a sliding mode controller and prove the stability of the proposed controller. Next, transforming the proposed sliding mode controller into a fuzzy sliding mode controller without changing the basic structure of the sliding mode controller, we easily obtain a fuzzy sliding mode controller(FSMC) whose stability is guaranteed without difficult stability proof procedure of the proposed FSMC.

This paper describes a new method for separation of the Volterra kernels which are identified by use of Msequence. One of the authors has proposed a method for identification of Volterra kernels of nonlinear systems using Msequence and correlation technique. When Msequence are applied to a nonlinear systems, the crosscorrelation function between the input and the output of the nonlinear systems includes crosssections of highorder Volterra kernels. However, if various order Volterra kernels exixt on the obtained crosscorrelation function, it is difficult to separate the Volterra kernels. In this paper, the authors show that the magnitude of Volterra kernels is maginified by the amplitude of Msequence according to the order of Volterra kernels. By use of this property, each order Volterra kernels is obtained by solving linear equations. Simulations are carried out for some nonlinear systems. The results show that Volterra kernels can be separated in each order successfully by the proposed method.

In this paper, the authors propose a new method for improving identification method of linear system by using Mtransform. The authors has recently proposed a new mettled for linear system identification by use of Mtransform. In this method, the input signal n(i) must have the same period N as that of the Msequence. When N becomes large, it will take a long time to compute. To overcome this difficulty, we propose a new approach of system identification by using a small size matrix. The results of simulation show a good agreement with the theoretical considerations.

Many classes of nonlinear systems can be represented by Volterra kernel expansion. Therefore, identification of Volterra kernels of nonlinear system is an important task for obtaining the nonlinear characteristics of the nonlinear system. Although one of the authors has recently proposed a new method for obtaining the Volterra kernels of a nonlinear system by use of Msequence and correlation technique, our mettled of nonlinear system identification is limited to Wienertype nonlinear system and we can not apply this method to the identification of Hammersteintype nonlinear system. This paper describes a new mettled for obtaining Volterra kernels of Hammerstein nonlinear system by adding a linear element in front of tile Hammerstein system. First we calculate the linear element of Hammerstein system by use of conventional correlation method. Secondly, we put a linear element in front of Hammerstein system. Then the total system becomes Wienertype nonlinear system. Therefore we can use our method on Volterra kernel identification by use of Msequence. Thus we get the coefficients of the approximation polynomial of nonlinear element of Hammerstein system. From the results of simulation, a good agreement with theoretical considerations is obtained, showing a wide applicability of our method.

A leastsquares identification method is studied that estimates a finite number of coefficients in the series expansion of a transfer function, where the expansion is in terms of recently introduced generalized basis functions, We will expand and generalize the orthogonal functions as basis functions for dynamical system representations. To this end, use is made of balanced realizations as inner transfer functions. The orthogonal functions can be considered as generalizations of, for example, the pulse functions, Laguerre functions, and Kautz functions, and give rise to an alternative series expansion of rational transfer functions. We show that the Laplace transform of the expansion for some sets
$\Psi_{\kappa}(Z)$ is equivalent to a series expansion . Techniques based on this result are presented for obtaining the coefficients$c_{n}$ as those of a series. One of their important properties is that, if chosen properly, they can substantially increase the speed of convergence of the series expansion. This leads to accurate approximate models with only a few coefficients to be estimated. The set of Kautz functions is discussed in detail and, using the powerseries equivalence, the truncation error is obtained. 
The distributedparameter structures expressed with the partial differential equations are considered as the infinitedimensional dynamic system. For implementation of a controller in multivariate systems, it is necessary to derive the statespace reduced order model. By the eigensystem realization algorithm, we can yield tile subspace system with the Markov parameters derived from the measured frequency response function by the inverse discrete Fourier transformation. We also review the necessary conditions for the convergence of the approximation system and the error bounds in terms of the singular values of Markovparameter matrices. To determine the natural frequencies and modal damping ratios, the modal coordinate transformation is applied to the realization system. The vibration test for a smart structure is performed to provide the records of frequency response functions used in the subspace system realization.

In this paper, a real time diagnostic algorithm fur estimating the impact location by loose parts is proposed. It is composed of two modules such as the alarm discrimination module (ADM) and the impactlocation estimation module(IEM). ADM decides whether the detected signal that triggers the alarm is the impact signal by loose parts or the noise signal. When the decision from ADM is concluded as the impact signal, the beginning time of bursttype signal, which the impact signal has usually such a form in time domain, provides the necessary data fur IEM. IEM by use of the arrival time method estimates the impact location of loose parts. The overall results of the estimated impact location are displayed on a computer monitor by the graphical mode and numerical data composed of the impact point, and thereby a plant operator can recognize easily the status of the impact event. This algorithm can perform the diagnosis process automatically and hence the operator's burden and the possible operator's error due to lack of expert knowledge of impact signals can be reduced remarkably. In order to validate the application of this method, the test experiment with a mockup (flat board and reactor) system is performed. The experimental results show the efficiency of this algorithm even under high level noise and potential application to Loose Part Monitoring System (LPMS) for improving diagnosis capability in nuclear power plants.

The goal of this paper is to describe an advanced method of the fault diagnois using Control Theory with reference to a crack detection, a new way to localize the crack position under infulence of the plant disturbance and white measurement noise on a rotating shaft. As a first step, the shaft is physically modelled with a finite element method as usual and the dynamic mathematical model is derived from it using the Hamilton  principle and in this way the system is modelled by various subsystems. The equations of motion with crack is established by adaption of the local stiffness change through breathing and gaping from the crack to the equation of motion with undamaged shaft. This is supposed to be regarded as reference for the given system. Based on the fictitious model of the time behaviour induced from vibration phenomena measured at the bearings, a nonlinear State Observer is designed in order to detect the crack on the shaft. This is elementary NL observer(EOB). Using the elementary observer, an Estimator(Observer) Bank is established and arranged at the certain position on the shaft. In case a crack is found and its position is known, the procedure for the estimation of the depth is going to begin.

Some nondestructive diagnostic methods including various types of corrosion sensors have been investigated. Under these conditions, a new structure of sensor that has a pair of electrode and magnetosupply was proposed. In order to detect the edge of the iron rust part, threepoles touchtype corrosion sensor is now proposed. The iron rust pattern where the sensor touches is estimated by means of the impedance of the sensor, and the edge of the iron rust is recognized by comparing the three measured impedances. As the result, our proposed sensor is useful to detect the initial state of iron rust.

Reliable maintenance scheduling of main equipments is a crucial problem in thermal power stations in order to skirt overall losses of power generation resulted from severe failures of the equipments. A reasonable method was proposed to determine the maintenance scheduling of whole pump system in thermal power stations in order to reduce the maintenance cost by keeping the present availability of the pump system throughout the operation. The dimensional reduction method was used to solve problems encountered due to few data which involved many operational factors in failure rate of pumps. The problem of bandlimited nature of data with time was solved by extrapolating future failures from presently available actual data with an aid of Weibull distribution. The results of the analysis identified the most suitable maintenance intervals of each pump type accordingly and hence reduce the cost of unnecessary maintenance with an acceptable range in the overall system availability.

An Electro Magnetic Suspension System, which has two floating masses connected with springs and dampers, can not keep its equilibrium when it is solved as an ordinary quartic mathematical model. So, a two dimensional controller, designed with quadratic mathematical model assuming the two mass model to be a rigid body, was used. As the result, the system floated stably. Therefore, we measured the transfer performances of this closed loop system contained this controller using the compression signal method proposed by N.Aoshima and identified the parameters of this system. Finally, we compared these parameters with the computing results of quartic mathematical model.

In this paper, a new input estimation algorithm is proposed for target tracking problem. The unknown target maneuver is approximated by a linear combination of independent time functions and the coefficients are estimated by using a weighted leastsquares estimation technique. The proposed algorithm is verified by computer simulation of a realistic twodimensional tracking problem. The proposed algorithm provides significant improvements in estimation performance over the conventional input estimation techniques based on the constantinput assumption.

This work is concerned with the problem of tracking a maneuvering target. In this paper, an error monitoring and recovery method of perception net is utilized to improve tracking performance for a highly maneuvering target. Many researches have been performed in tracking a maneuvering target. The conventional Interacting Multiple Model (IMM) filter is well known as a suboptimal hybrid filter that has been shown to be one of the most costeffective hybrid state estimation scheme. The subfilters of IMM can be considered as fusing its initial value with new measurements. This approach is also shown in this paper. Perception net based error monitoring and recovery technique, which is a kind of geometric data fusion, makes it possible to monitor errors and to calibrate possible biases involved in sensed data and extracted features. Both detecting a maneuvering target and compensating the estimated state can be achieved by employing the properly implemented error monitoring and recovery technique. The IMM filter which employing the error monitoring and recovery technique shows good tracking performance for a highly maneuvering target as well as it reduces maximum values of estimation errors when maneuvering starts and finishes. The effectiveness of the proposed method is validated through simulation by comparing it with the conventional IMM algorithm.

This paper presents a closed loop identification algorithm in time domain. This algorithm can be used for identification of unstable system and for model validation of system which is difficult to derive analytical model. In time domain, projection filter, which projects a finite number of input output data of a system into its current space, is used to relate the state space model with a finite difference model. Then recursive relations between the Markov parameters and the ARX model coefficients are derived to identify the system, controller and Kalman filter Markov parameters recursively, which are finally used to identify the system, controller and Kalman filter gains. The NASA LAMSTF is used to validate the algorithms developed.

A virtual reality system is implemented for the operator supervising a robot's operation at a remote site. For this implementation, a two D.O.F forcereflective joystick is designed to reflect the force/torque measured at the end of robotic manipulator and to generate the motion command for the robot by the operator using this joystick. In addition, the visual information that is captured by a CCD camera, is transmitted to the remote operator and is displayed on a CRT monitor. The operator who is holding the force reflective joystick and watching the CRT monitor can resolve unexpected problems that the robot confronts with. That is, the robot performs the tasks autonomously unless it confronts with unexpected events that can be resolved by only the operator. To demonstrate the feasibility of this system, a remote peginhole operation is implemented and the experimental data are shown.

Cutting process has been automated due to progress of CNC and CAD/CAM, but polishing process has been only depended on experiential knowledge of expert. Polishing work for a curved surface die demands simple and repetitive operations but requires much time for its high precision. Therefore it is operated in the handiwork by skilled worker. However the workers intend to avoid gradually polishing work because of the poor environments such as dust and noise. In order to reduce the polishing time and solve the problem of shortage of skilled workers, it has been done some research for an automation of polishing. To automate the polishing process, a 2 axes polishing robot which is attached to a 3 axes machining center has been developed by our previous research. This automatic polishing robot is able to keep the polishing tool normal on the curved surface of die. Therefore its performance of polishing is improved because of always keeping the tool normal on the surface. In this paper, the smaller sized polishing robot is developed to improve polishing performance. And the controller for 2 axes polishing robot is developed. The controller is composed of TMS320C31 with high speed which is 40ns instruction cycle time, RAM memory with 64K words, digital input with 64 bits, digital output with 32 bits, and D/A converter with 4 channels, which is 12 bits resolution. To evaluate polishing performance of this developed robot, polishing experiment for shadow mask was carried out.

A musculotendon model is investigated to show muscle fatigue under the repeated functional electrical stimulation (FES). The normalized Hilltype model can predict the decline in muscle force. It consists of nonlinear activation and contraction dynamics including physiological concepts of muscle fatigue. A muscle fatigue as a function of the intracellular acidification, pHi is inserted into contraction dynamics to estimate the force decline. The computer simulation shows that muscle force declines in stimulation time and the change in the estimate of the optimal fiber length has an effect only on muscle time constant not on the steadystate tetanic force.

Neural Networks, modeled succinctly from the real nervous system of a living body, can be categorized into two folds; artificial neural network(ANN) and biological neural network(BNN). While the former has been developed to solve practical problems using function approximation capability, pattern classification) clustering algorithm, etc, the latter has been focused on verifying the information processing capability to which brain research gives an impetus, by mimicking real biological systems. However, BNN suffers Iron severe nonlinearities dealt with. A bridge between two neural networks is chaotic neural network(CNN), which simply delineate the real norvous system and comprises almost all the ANN structures by selecting parameters. Main research theme of this area is to develop an explanation tool to clarify the information processing mechanism in biological systems and its extension to engineering applications. The CNN has a Gaussianshaped refractory function with hysteresis effect and the chaotic responses of it have been observed fur a wide range of parameter space. Through the examination of the coupling effects of excitatory and inhibitory connections, the secrets of information processing and memory structure will appear.

This paper proposes a fast H
$\sub$ $\infty$ / gain scheduled controller that stabilizes the uncertain nonlinear system with exogenous signals. The controller is constructed at a distinct and fixed value of exogenous signals using H$\sub$ $\infty$ / synthesis methodology. Then the constructed controller set is switched for the wide range of variation of exogenous signals. Using the derivative gain, the number of constructed and engaged controllers for the fast varying exogenous signal is reduced. 
In this paper, the mathematical model of a cut with an inverted flexible beam and a concentrated tip mass was derived. The characteristic equation for calculating the natural frequencies of the cartbeammass system was obtained and the motion of the system was analyzed through unconstrained modal analysis. A good positioning response of the cart without excessive vibrational motion of the tip mass could be obtained through numerical simulation using PID controller with the feedback of both the position of the cart and the deflection of the beam.

A method to construct a memoryless feedback law for systems with multiple timedelays in the states is proposed. As a plant model, a differentialdifference equation with multiple delayed terms is introduced, A stabilizability condition by memoryless feedback is presented. A feedback gain is calculated with a solution of a finite dimensional Riccati equation. It is shown that the resulting closed loop system is asymptotically stable, and moreover, it is a linear quadratic regulator for some cost functional. An alternative stabilizability condition which is easier to check is given.

The report presents an approach to constructing or control algorithms for finite dimensional dynamical systems under the deficit of information about dynamical disturbances. The approach is based on the constructions of the extremal shift strategy of the differential game theory.


A masterslave system is proposed as a teaching device for a dual arm robot. The slave robots are remotely controlled by two deltatype master arms. In order to help the operator to observe the target object from the desired position and desired direction, cameras are mounted on a specialized manipulator, Movements of two slave arms are coordinated with that of the cameras. Due to this coordinated movements, the operator needs not to care the geometrical relation between the cameras and the slave robots.

The manmachine interface Is an important factor in the computer system, and it is thought that lineofsight (LOS) detection technology will allow significant advances in this field. Techniques for detecting LOS for use in human interfaces have been studied［1]［2］. In earlier studies, however, LOS was detected with a head piece, goggles, or through fixing the position of the head. The limitations imposed by these fixed conditions render them unsuitable far use in interfaces, as they have adverse mental or physical effects on humans. Therefore. they have not been sufficiently developed for practical application. Research on noncontact LOS detection is expected to result in a usable LOS manmachine interface［3］［4］, and the current study is intended to be a step in that direction. The authors used color contact lenses for LOS detection, and applied this new method to a computer interface. The use of color contact lenses simplifies image processing. The algorithm used in this study is sufficiently accurate for practical applications. This technique can be used in input devices, in virtual reality applications, and in human engineering research.

A design of PIDA (ProportionalIntegralDerivativeAcceleration) controller for the thirdorder plant using the CDM (Coefficient Diagram Method) is presented. Using CDM, the closedloop system with the designed PIDA controller can be made stable and satisfied both transient and steady state response specifications without any adjustment. The effect of output step disturbance can also be lastly rejected. The fast step response of the controlled system can be achieved by reducing the equivalent time constant. The MATLAB's simulation results show that the performances of the designed controlled system using CDM is better than the performances of the controlled system using PIDA controller designed by its own technique.

Mizoguchi, Hiroshi;Teshiba, Masashi;Goto, Yoshiyasu;Hidai, KenIchi;Shigehara, Takaomi;Mishima, Taketoshi 401
If a mobile robot can be controlled remotely via the internet using wireless IP protocol network, it becomes much useful and convenient. However risk of illegal access is also increased. This paper discusses problems of the illegal access and proposes protection methods against the access. 
Prasit, Julseeewong;Prapart, Ukakimaparn;Thanit, Trisuwannawat;Anuchit, Jaruvanawat;Kitti, Tirasesth 407
A design technique based on the root locus approach for the SISO (SingleInput SingleOutput) systems using PID (ProportionalIntegralDerivative)${\times}$ (n1) stage PD as a controller for the n$\^$ th/ order plant is presented. The controller is designed based on transient and steady state response specifications. This controller can be used instead of a conventional PID controller. The overall system is approximated as a stable and robust second order system. The desired performances are achieved by increase the gain of the controller. In addition, the controller gain can be adjusted to obtain faster response with a little overshoot. The simulation results show the merits of this approach. 
We often use induction motor in the hard environment including vibration and high ambient temperature, or in maintenance free operation, because induction motor has durable and simple structure. However, when we use it as servo actuator or accurate speed control motor, we have to equip sensor such as encoder and tachogenerator with the motor control system. And generally those sensor's abilities against bad environment are less than the induction motor itself, So if we can remove these sensors from the system, it'll have more environmental resistance, and the cost will also be reduced. Actually this removal has been achieved in limited field. However, that needs complex calculations and a certain elapse time for data processing. In our study, we intended to estimate the rotational speed from the motor current instead of speed sensor, easily and rapidly in comparison to former methods.

This paper presents how to design speed control of wound rotor induction motors with slip energy recovery. The speed is limited at some range of subsynchronous speed of the rotating magnetic field. The problem with speed control by adjusting resistance value in the rotor circuit reduces the efficiency of power, because of the slip energy is lost when it passes through the rotor resistance. The control system is designed to maintain efficiency of motor, where it recovers loss energy by returning it to the system to improve the efficiency. A new PI control method of adaptive control [1］,［13］is applied for the system with cascade type PI controller on the main loop to keep the speed constant and the internal loop to adjust the rotor appropriated current of the load provides the good transient response without overshoot.

This paper studies the regional identifiability of spatiallyvarying parameters in distributed parameter systems of hyperbolic type. Let Ω be a bounded domain of R
$^n$ and let Ωo be a subregion of the closed domain Ω. The distributed parameter systems having unknown parameters defined on Ω are described by the second order evolution equations in the filbert space L$^2$ (Ω) and the observations are made on the subregion Ωo ⊂ Ω. The regional identifiability is formulated as the uniqueness of parameters by the identity of solutions on the subregion. Several regional identifiability results of the spatiallyvarying parameters of hyperbolic distributed parameter systems are established by means of the Riesz spectral representations. 
The previously developed control design methodology, EALQR(Eigenstructure Assignment/LQR), has better performance than that of conventional LQR or eigenstructure assignment. But it has a constraint for the weigting matrix in LQR, that is the weighting matrix could be indefinite for highorder systems. In this paper, the effects of the indefinite weighting matrix in EALQR on the Sequency domain properties are analyzed. The robustness criterion and quantitative frequency domain properties are also presented. Finally, the frequency domain properties of EALQR has been analyzed by applying to a flight control system design example.

A control strategy for flexible structure under irregular disturbance by using of
$\boxDr$ random gain$\boxUl$ is developed and implemented. System equation is transformed to stochastic domain by FPK approach from physical domain. A controller is designed in the stochastic domain, accordingly system is controlled by$\boxDr$ random gain$\boxUl$ in time domain. In the paper, a new control technique is successfully employed for flexible system under white noise, and the result is verified by MonteCarlo simulation and compared with the performance via LQR controller. 
In this paper, we consider numerical solution of a HJB (HamiltonJacobiBellman) equation of elliptic type arising from the stochastic control problem. For the numerical solution of the equation, we take an approach involving contraction mapping and finite difference approximation. We choose the It(equation omitted) type stochastic differential equation as the dynamic system concerned. The numerical method of solution is validated computationally by using the constructed test case. Map of optimal controls is obtained through the numerical solution process of the equation. We also show how the method applies by taking a simple example of nonlinear spacecraft control.

In this paper, a robust nonlinear predictiontype controller (RNPC) is developed for the continuous time nonlinear system whose control objective is composed of system output and its desired value. The basic control law of RNPC is derived such that the future response of the system is first predicted by appropriate functional expansions and the control law minimizing the difference between the predicted and desired responses is then calculated. RNPC which involves two controls, i.e., the auxiliary and robust controls into the basic control, shows the stable closed loop dynamics of nonlinear system of any relative degree and provides the robustness to the nonlinear system with parameter/modeling uncertainty. Simulation tests for the position control of a twolink rigid body manipulator confirm the performance improvement and the robustness of RNPC.

This paper proposes a new method of Model Predictive Control (MPC) of nonlinear process by using the measured Volterra kernels as the nonlinear model. A nonlinear dynamical process is usually described as Volterra kernel representation, In the authors' method, a pseudorandom Msequence is ar plied to the nonlinear process, and its output is measured. Taking the crosscorrelation between the input and output, we obtain the Volterra kernels up to 3rd order which represent the nonlinear characteristics of the process. By using the measured Volterra kernels, we can construct the nonlinear model for MPC. In applying Model Predictive Control to a nonlinear process, the most important thing is, in general, what kind of nonlinear model should be used. The authors used the measured Volterra kernels of up to 3rd order as the process model. The authors have carried out computer simulations and compared the simulation results for the linear model, the nonlinear model up to 2nd Volterra kernel, and the nonlinear model up to 3rd order Volterra kernel. The results of computer simulation show that the use of Valterra kernels of up to 3rd order is most effective for Model Predictive Control of nonlinear dynamical processes.

We propose robust MILP model for scheduling and design of multiproduct batch processes in this paper. Recent stochastic modeling approaches considering uncertainty have mainly focused on maximization of expected NPV. Robust model concept is applied to generate solution spectrum in which we can select the best solution based on tradeoff between robustness measure and expected NPV. Robustness measure is represented as penalty term in the objective function, which is Upper Partial Mean of NPV. We can quantify solution robustness by this penalty term and maintain model as MILP form to be computationally efficient. An example illustrates the effectiveness of the proposed model. In many cases sufficient robustness can be guaranteed through a little reduction of expected NPV.

In spite of significant nonlinearities even in the simplest model, some types of steadystate and dynamic behavior common for nonlinear systems have never been associated with distillation columns. In recent years, multiplicity of steady states has been a subject of much research and is now widely accepted. Subsequently, stability of steady states has been explored. Another phenomenon that. although widely observed in chemical reactors, has not been associated with models of distillation columns is the existence of periodic oscillations. In this article we study the steadystate and dynamic behavior of the azeotropic distillation of the ternary homogeneous system methanolmethyl butyratetoluene. Our simulations reveal nonlinear behavior not reported in earlier studies. Under certain conditions, the openloop distillation system shows a sustained oscillation associated with branching to periodic solutions. The limit cycles are accompanied by traveling waves inside the column. Significant underdamped oscillations are also observed over a wide range of product rates.

A mathematical model was developed for a continuous reactor in which free radical polymerization of methyl methacrylate (MMA) occurred. Elementary reactions considered in this study were initiation, propagation, termination, and chain transfers to monomer and solvent. The reactor model took into account the density change of the reactor contents and the gel effect. A control system was designed for a continuous reactor using extended Kalman filter (EKF) based nonlinear model predictive controller (NLMPC) to control the conversion and the weight average molecular weight of the polymer product. Control input variables were the jacket inlet temperature and the feed flow rate. For the purpose of validation of the control strategy, online digital control experiments were conducted with densitometer and viscometer for the measurement of the polymer properties. Despite the complex and nonlinear features of the polymerization reaction system, the EKF based NLMPC performed quite satisfactorily for the property control of the continuous polymerization reactor.

In this paper, we present a mathematical model of the piezoelectric pump and its application to the automobile brake system. The piezoelectric pump consists of a multilayered piezoelectric element a diaphragm, pumping values, resonant pipes and accumulators, and the maximum pumping power of 62W nab obtained in the previous experiments by using the piezoelectric element of 22mm diameter and 55.5mm length. A detailed mathematical model of the pump is derived here by considering the compressibility of the working oil, nonlinear characteristics of piezoelectric element, the time delay of pumping values' action and be on. The validity of the model is illustrated by comparing the experimental data and the simulation results. Using the mathematical model of the piezoelectric pump, a brake system for automobile disk brake is also simulated in this paper. The brake system consists of a piezoelectric pump as a power source, calipers and its piston to generate brake force, and a three position solenoid value to change the brake situation. It is shown that 15mm/sec of piston speed and 20kN of piston force are obtained by using the piezoelectric element of 33mm diameter and 55.5mm length.

For memory products, it is very important to develop a new production line as soon as possible. All products are inspected to get rid of defected products at the last testing stage. Those inspection data are called FCM. In this paper, based on the FCM data, Area Usage Factor (AUF) analyzing method will be proposed. Process engineers can make up their mind to which direction they should concentrate their analyzing power.

In this paper, we examine the displacement characteristics of the parallel leaf spring mechanism with largedeflective elastic hinges, and the validity of this mechanism as a translational and rotational mechanism is confirmed with multiinput system. This study is focused on the linear driving force as an input force, which is applied to the largedeflective elastic mechanism, and the displacement characteristics are discussed with theoretically and experimentally. The motions of this mechanism due to largedeflective hinges are changed by the position of loading force regardless of a single driving force. The numbers of degree of freedom are increased with the hinges, and we can be used to a multiple driving force in order to obtain many types of Output.

In this paper an asymmetric hydraulic actuator which consists of single acting cylinder and servo valve is modeled for a quarter car active suspension system. This model regards the force as an internal state rather than a control input. The control input of the model is the sum of oil flows that pass through the valve's orifices. The resulting dynamic equation in the state space appears a feedback connection of a nominal linear time invariant term with a nonlinear bounded uncertain block. Since this model makes it possible to eliminate the force control phase, analysis and controller design are made straightforward and simple. Well known LQR method is then applied. Simulation and test rig experiment show the effectiveness of this approach in modeling and control.

Not merely running at the designated constant speed as the classical cruise control, the adaptive cruise control (ACC) maintains safe headway distance when the front is blocked by other vehicles. One of the most essential part of ACC System is the range sensor which can measure the position and speed of all objects in front continuously, ignore all irrelevant objects, distinguish vehicles in different lanes and lock on to the closest vehicle in the same lane. In this paper, the hierarchical object recognition algorithm (HORA) is proposed to process raw scanning laser data and acquire valid distance to target vehicle. HORA contains two principal concepts. First, the concept of life quantifies the reliability of range data to filter off the spurious detection and preserve the missing target position. Second, the concept of conformation checks the mobility of each obstacle and tracks the position shift. To estimate and predict the vehicle position Kalman filter is used. Repeatedly updated covariance matrix determines the bound of valid data. The algorithm is emulated on computer and tested online with our ACC vehicle.

A mathematical model is developed to describe the relationship between the manipulated variables (e.g. jacket inlet temperature and feed flow rate) and the important qualities (e.g conversion and weight average molecular weight (Mw)) in a continuous polymerization reactor. The subspacebased identification method for Wiener model is used to retrieve from the discrete sample data the accurate information about both the structure and initial parameter estimates for iterative parameter optimization methods. The comparison of the output of the identified Wiener model with the outputs of a nonlinear plant model shows a fairly satisfactory degree of accordance.

It is an interesting area in the field of artificial intelligence to and an analytic model of cooperative structure for multiagent system accomplishing a given task. Usually it is difficult to design controllers for multiagent systems without a comprehensive knowledge about the system. One of the way to overcome this limitation is to implement an evolutionary approach to design the controllers. This paper introduces the use of a genetic algorithm to discover a fuzzy logic controller with rules that govern emergent cooperative behavior: A modified genetic algorithm was applied to automating the discovery of a fuzzy logic controller jot multiagents playing a pursuit game. Simulation results indicate that, given the complexity of the problem, an evolutionary approach to and the fuzzy logic controller seems to be promising.

In this paper, a robust faulttolerant control scheme for robot manipulators overcoming actuator failures is presented. The joint(or actuator) fault considered in this paper is the freeswinging joint failure and causes the loss of torque on a joint. The presented faulttolerant control framework includes a normal control with normal(nonfailed) operation, a fault detection and a faulttolerant control to achieve task completion. For both no uncertainty case and uncertainty case, a stable normal controller and an online fault detection scheme are presented. After the detection and identification of joint failures, the robot manipulator becomes the underactuated robot system with failed actuators. A robust adaptive control scheme of robot manipulators with the detected failedactuators using the brakes equipped at the failed(passive) joints is proposed in the presence of parametric uncertainty and external disturbances. To illustrate the feasibility and validity of the proposed faulttolerant control scheme, simulation results for a threelink planar robot arm with a failed joint are presented.

This paper presents a robust control scheme of freejoint manipulators to overcome actuator failures and uncertainties in Cartesian space where tasks are planned. The control scheme has the adaptation law for the upper bound on the norm of uncertainties through the Lyapunov function approach. To solve the dynamic singularity problem in the controller, the singular and nonsingular regions are investigated based on a computer simulation. Then a singularityfree Cartesian trajectory planning is achieved in order to guarantee the availability of the control scheme. To illustrate the validity of the proposed control scheme, simulation results for a threelink planar robot arm with a free joint are shown.