In this study, a design method to obtain a robust suboptimal regulator for linear multivariable system is presented. This new design method is based on the optimal regulator design method using eigen-structure assignment and it uses additional cost function which represent robustness of the closed loop system. When we design the regulator using pole assignment method for linear multivariable system we have extra degree-of-freedom after assigning desired eigenvalues of the closed loop system in determining the feedback gain. So we assign additional robust suboptimal regulator. In this study we also feedback the system output for more practical applications.

In this paper we present a method of reducing controller design problem from LQG/LTR approach to H.inf. optimization. The condition of the existance of the optimal solution is derived. In order to derive the controller equation for NMP plant we reduce the H.inf. LTR problem to Nehari's extension problem and derive the optimal controller equation which is best approximation for this problem. Furthermore, we show that the controller obtained by presented method guarantee the asymptotic LTR condition and stability of closed loop system.

The purpose of this paper is to design a robust controller for a class of time-varying systems with bounded disturbance described by the differential equation. The robust desiging method proposed in this paper, called "incentive design method" is different from developed designing methods in the past, and has following properties. The robust control law designed by this method can guarantee a certain value of the cost functional no matter how the disturbance vary within the given bounds. Here, the certain value of the cost functional may not be a saddle-point value, but is the value selected by a system designer. Therefore, the bounded disturbance has at least no bad effect on the value of the cost functional during finite interval of time. The method is based on the theory of incentive differential games. In addition, the form of control law is constructed by the system designer ahead of time. A numerical illustrative example is given in this paper. It is shown from this derivation and this numerical example that the approach developed in this paper is effective and feasible for some practical control problem.l problem.

A synthesis of feedback control-law with combined H$_{2}$/H$_{\infty}$ perfoemance criteria is proposed for discrete-time systems, under the assumption that the state is available for feedback. An auxiliary minimization problem is defined to enforce the H$_{\infty}$ disturbance attenuation constrain while minimizing the H$_{2}$ performance bound. The design equation is presented in terms of a modified Riccati equation which leads to the standard LQ solution when the H$_{\infty}$ constraint is completely relaxed. The results of the paper clarity the correspondences between H$_{2}$/H$_{\infty}$ results in discrete-time systems and their continuous-time counter-parts.rts.

In this paper, a state space solution to the discete time H$_{\infty}$ control problem is presented. It is shown that there exist LQ game problem corresponding to H$_{\infty}$ control problems and the H$_{\infty}$ controller can be obtained by solving the LQ game problem. Explicit state space formulae are given for the state feedback H$_{\infty}$ controller and output feedback H$_{\infty}$ controllers.lers. state feedback $H_{\infty}$ controller and output feedback $H_{\infty}$ controllers.

We derive an intelgral-type constraint on the complementary sensitivity function in digital control systems. Some design guidances are proposed for the pole assignment of digital controller with computation-time delay to improve the complementary sensitivity characteristics.

Robustness of a model refernece direct adaptive pole placement control for not necessarily minimum phase systems is studied subject to unmodeled dynamics and bounded disturbances. The adaptive control scheme involves two estimators for the system and the controller parameter estimation, respectively. The robustness is obtaind under some weak assumptions and by using both a normalized least-squares algorithm with dead zone and an appropriate nonlinear feedback.

There have been many approaches to solve the disturbance rejection problem in the control of LTI systems with state independent disturbances or possibly nonlinear state dependent disturbances. From the view point of each actuator, robot manipulators can be modeled as the second class of systems. With this model, M.Nakao et al. [1] introduced a decentralized control scheme based on interference estimation which is simple in its implementation and robust to the coupled dynamics and parameter variations. This paper systematically generalizes the control scheme to arbitrary finite dimensional LTI systems with disturbances. In doing so, we develop a disturbance observer theory for solving the disturbance rejection problem. We also present a discrete version of the theory with discussion of sampling and time-delay effects.

In this paper, a new hybrid position control algorithm for the direct drive arm is proposed. The proposed control is composed of discrete feedforward component and continuous feedback component. The discrete component is the nominal torque which approximately compensates the strong nonlinear coupling torques between the links, while the continuous control is a modified version of sliding mode control which is known to have a robust property to the disturbances of system. For the proposed control law, we give sufficient condition which guarantees the bounded tracking error in spite of the modeling errors, and the efficiency of the proposed algorithm is demonstrated by the numerical simulation of a three link manipulator position control with payloads and parameter errors.

A theoretical and experimental study of a single link flexible arm a tip mass is presented for the translational end-point positioning. The problem of shifting the end-point from its initial position to the commanded position by the amount of W $_{d}$ is considered for the open loop control such that the base follows up the given path function. The theoretical results are obtained by applying the method of the Laplace transform to the governing equation, and the solution is calculated by the method of numerical inversion. Experimental results are obtained and compared with the theoretical ones.s.

A simulation analysis is presented for the position control of a single-link flexible manipulator whose end-effector is subjected to an impulsive force. Arm is rotated by a d.c. servomotor at the shoulder so that the end point stays precisely at its initial position even if the end effector is thumped with the impulsive loading. A gap sensor is used to measure the tip displacement. The control torque based on the PD control law is applied to the motor through the driver circuit. The control strategy is tested by means of computer simulation for the one-link flexible-arm prototype in the authers' laboratory at Tohoku Univ.

On-line robust control based on a stability index for time-delay systems has been developed. The purpose of the proposed design algorithm is to on-line tune a filter in the control loop. The problem of robust control with an incorrect given bound on the modeling error is investigated. Illustrative examples are presented to show the promise of the proposed method.

An approach to robot force control, which allows force manipulations to be realized without overshot and overdamping while in the presence of unknown environment, is given in this paper. The main idea is to use dynamic compensation for known robot parts and fuzzy compensation for unknown environment so as to improve system performance. The fuzzy compensation is implemented by using rule based fuzzy approach to identify unknown environment. The establishment of proposed control system consists of following two stages. First, similar to the resolved acceleration control method, dynamic compensation and PID control based on known robot dynamics, kinematics and estimated environment compliance is introduced. To avoid overshoot the whole control system is constructed overdamped. In the second stage, the unknown environment stiffness is estimated by using fuzzy reasoning, where the fuzzy estimation rules are obtained priori as the expression of the relationship between environment stiffness and system response. Based on simulation result, comparisons between cases with or without fuzzy identifications are given, which illustrate the improvement achieved.

In this paper, attempts have been made to control AC synchronous servo motor used as actuators of joints of the FARA robot with high dynamic performance and precise positioning. The AC synchronous servo motors used in FARA robots have resolves as position sensors. Resolver to digital converters are used in order to obtain the information of rotor speed and position from resolver outputs. The proposed joint position control system consists of four speed controller and one position controller. Analog methods are used in the position controller, while digital methods are used in the position controller. For precise position control, PID control algorithm and interpolation functions are executed in two 16 bit microprocessors with sampling rate 2ms. Experimental results show that the proposed joint position control system can be effectively applied to industrial robots in order to obtain high dynamic performance and precise positioning. The proposed joint position control system is being used in the control of FARA robots of Samsung Electronics.

Pattern classification is an essential step in automatic robotic assembly which joins together finite number of seperated industrial parts. In this paper, a fast and systematic algorithm for classifying occlusion-free objects is proposed, using the notion of incremental circle transform which describes the boundary contour of an object as a parametric vector function of incremental elements. With similarity transform and line integral, normalized determinant curve of an object classifies each object, independent of position, orientation, scaling of an object and cyclic shift of the stating point for the boundary description.

This paper describes the design and implementation of a real-time, high-precision vision system and its application to SMT(surface mounting technology) automation. The vision system employs a 32 bit MC68030 as a main processor, and consists of image acquisition unit. DSP56001 DSP based vision processor, and several algorithmically dedicated hardware modules. The image acquisition unit provides 512*480*8 bit image for high-precision vision tasks. The DSP vision processor and hardware modules, such as histogram extractor and feature extractor, are designed for a real-time excution of vision algorithms. Especially, the implementation of multi-processing architecture based on DSP vision processors allows us to employ more sophisticated and flexible vision algorithms for real-time operation. The developed vision system is combined with an Adept Robot system to form a complete SMD system. It has been found that the vision guided SMD assembly system is able to provide a satisfactory performance for SND automation.

This paper considers the problem of patterns recognition using the artificial neural network systems. The artificial neural network systems provide an effective tool for classifying patterns and/or characters by learning them in a certain repeated hashion. The mechanism of the learning process and the structure of neural network systems used are main concerns in the accurate and fast classification of the patterns which are slightly different each other. The neural network system employed in this study has three layers structure which is composed of input, intermidiate, and output layers. Our main concern is to develope an effective learning mechanism how to learn the patterns fastly and accurately. The experimental study performed shows that there exists an effective learning method to get higher recognition ratio in classifying the several different patterns by artificial neural network system constructed.

A method of recognizing 2-dimensional polygonal object is proposed by using a concept of generalized incremental circle transform. The generalized incremental circle transform, which maps boundaries of an object into a circular disc, represents efficiently the shape of the boundaries that are obtained from digirized binary images of the objects. It is proved that the generalized incremental circle transform of an object is invariant to object translation, rotation, and size, and can be used as feature information for recognizing two dimensional polygonal object efficiently.

To recognize isomorphic transformation patterns, such as scale-change, translation and rotation transformed patterns, is an old difficult but interesting problem. Many researches have been done with a dominant approach of normalization by many eminent pioneers. However, there seems no a perfect system which can even recognize 90 .deg.-multiple rotation isomorphic transformation patterns for real needs. Here, as a new challenge, we propose a method of how to recognize 90 .deg.-multiple rotation isomorphic and symmetry isomorphic transformation patterns.

An automatic visual classification system is introduced which provides for measuring the length and diameter of coilform cores and dividing them into 5 different classed in terms of how far their length be from the desired length. This task is fully automated by controlling two STEP motors and by using image processing techniques. The classification procedure is broken into three logical parts, First, cores in the form of randomly stacked bundle are lined up one by one so as to be well captured by a camera. The second part involves capturing core image. Then, it enters the measuring process. Finally, this machine would retain all the information relating to the length. According to the final result, cores are sent to the corresponding bin. This considerably simplifies the selecting task and facilitates a greatly improved reliablity in precision. The average classifying capability is about 2 pieces per second.

The present paper describes a new technique for associating images employing a set of local constraints among pixels on an image. The technique describes the association problem in terms of consistent labeling which is an abstraction of various kinds of network constraints problems. In this particular research, a pixel and its gray value correspond to a unit and a label, respectively. Since constraints among units on an image are defined with respect to each n-tuple of pixels, performance of the present association technique largely depends on how to choose the n-tuples on an image plane. The main part of this paper is devoted to discussing this selection scheme and giving a solution to it as well as showing the algorithm of association. Also given are some results of the simulation performed on synthetic binary images to examine the performance of proposed technique, followed by the argument on further studies.

A method to determine nonlinear system observability will be introduced here. For the determination of the deterministic nonlinear system observability two conditions connectedness and univalence, are developed and used. Depending on how the conditions are satisfied, observability is classified in three categories : observability in the strict sense, wide sense, and the unobservable case. Including simple linear and nonlinear system example an underwater acoustical locaization tracking nonlinear system, the bearing-only tracking example is analyzed.

An iterative learning control scheme for tracking control of a class of uncertain nonlinear systems is presented. By introducing a model reference adaptive controller in the learning control structure, it is possible to achieve zero tracking of unknown system even when the upperbound of uncertainty in system dynamics is not known apriori. The adaptive controller pull the state of the system to the state of reference model via control gain adaptation at each iteration, while the learning controller attracts the model state to the desired one by synthesizing a suitable control input along with iteration numbers. In the controller role transition from the adaptive to the learning controller takes place in gradually as learning proceeds. Another feature of this control scheme is that robustness to bounded input disturbances is guaranteed by the linear controller in the feedback loop of the learning control scheme. In addition, since the proposed controller does not require any knowledge of the dynamic parameters of the system, it is flexible under uncertain environments. With these facts, computational easiness makes the learning scheme more feasible. Computer simulation results for the dynamic control of a two-axis robot manipulator shows a good performance of the scheme in relatively high speed operation of trajectory tracking.

This paper presents a method of suboptimal control for nonlinear systems via block pulse transformation. The adaptive optimal control scheme proposed by J.P. Matuszewski is introduced to minimize the performance index. Nonlinear systems are controlled using the obtained optimal control via block pulse transformation. The proposed method is simple and computationally advantageous. Viablity of the this method is established with simulation results for the van der Pol equation for comparision with other methods.

The system discussed in this paper is an integrated stand-alone system with the full functional capabilities required of a telerobot system. It is complete with a force-reflecting 6-DOF hand controller, driving a PUMA 560 or 762 robot, with an integrated force-torque sensing wrist sensor and servo-driven parallel jaw gripper. A mix of custom and standard electronics, distributed computers and microprocessors, with embedded and downloadable software, have been integrated into the system, giving rise to a powerful and flexible teleautonomous control system.

In order to find the optimal control law for the precise trajectory tracking of a robot manipulator, a perturbational control method is proposed based on a linearized manipulator dynamic model which can be obtained in a very compact and computationally efficient manner using the dual number algebra. Manipulator control can be decomposed into two parts: the nominal control and the corrective perturbational control. The nominal control is precomputed from the inverse dynamic model using the quantities of a desired trajectory. The perturbational control is obtained by applying the second-variational method on the linearized dynamic model. Simulation results for a PUMA-560 robot show that, by using this controller, the desired trajectory tracking performance of the robot can be achieved, even in the presence of large initial positional disturbances.

This paper presents a fundamental study of a millimeter-sized master-slave robot driven by conduit-guided wires, which is expected to be applied to the delicate surgical operations, the assembling precise and small parts and so on. This system consists of a millimeter-sized slave robot and a master manipulator of which the size is adapted to a human finger. Displacement and torque of the master side can be reduced and transferred to the slave robot by controlling the motor torque against the master torque by feeding back tension signals. The master can feel the tensions by the motor torque. In this paper, the design method and making process of the master-slave system and the dynamical characteristic of displacement and torque control are proposed.

A minimum-time trajectory planning for two robot arms with designated paths and coordination is proposed. The problem considered in this paper is a subproblem of hierarchically decomposed trajectory planning approach for multiple robots : i) path planning, ii) coordination planning, iii) velocity planning. In coordination planning stage, coordination space, a specific form of configuration space, is constructed to determine collision region and collision-free region, and a collision-free coordination curve (CFCC) passing collision-free region is selected. In velocity planning stage, normal dynamic equations of the robots, described by joint angles, velocities and accelerations, are converted into simpler forms which are described by traveling distance along collision-free coordination curve. By utilizing maximum allowable torques and joint velocity limits, admissible range of velocity and acceleration along CFCC is derived, and a minimum-time velocity planning is calculated in phase plane. Also the planning algorithm itself is converted to simple numerical iterative calculation form based on the concept of neural optimization network, which gives a feasible approximate solution to this planning problem. To show the usefulness of proposed method, an example of trajectory planning for 2 SCARA type robots in common workspace is illustrated.

We present algebraic algorithms for collision-avoidance robot motion planning problems with planar geometric models. By decomposing the collision-free space into horizontal vertex visibility cells and connecting these cells into a connectivity graph, we represent the global topological structure of collision-free space. Using the C-space obstacle boundaries and this connectivity graph we generate exact (non-heuristic) compliant and gross motion paths of planar curved objects moving with a fixed orientation amidst similar obstacles. The gross motion planning algorithm is further extended (though using approximations) to the case of objects moving with both translational and rotational degrees of freedom by taking slices of the overall orientations into finite segments.

This research focuses on developing a new and computationally efficient algorithm for free space structuring and planning collision free paths for an autonomous mobile robot working in an environment populated with polygonal obstacles. The algorithm constructs the available free space between obstacles in terms of free convex area. A collision free path can be efficiently generated based on a graph constructed using the midpoints of common free links between free convex area as passing points. These points correspond to nodes in a graph and the connection between them within each convex area as arcs in this graph. The complexity of the search for collision free path is greatly reduced by minimizing the size of the graph to be searched concerning the number of nodes and the number of arcs connecting them. The analysis of the proposed algorithm shows its efficiency in terms of computation ability, safety and optimality.

In the Orient, most references are made on the visual perception of the eye having the blue iris, even in the case where the reference should be made on the perception by the eye having the dark iris. In order to clarify the difference in the visual perceptions of a monochromatic light, an objective boundary wavelength between hues incident on the dark eyes was obtained and the correlation between the valuesalready reported as having been obtained with the blue eyes was examined. A significant difference was found in the subjective boundary wavelengths of hues depending on the difference in the visual sensitivity distribution on the dark and blue eyes.

Iterative algorithms for detecting the collision of convex objects whose motion is characterized by a path in configuration space are described. They use as an essential substep the computation of the distance between the two objects. When the objects are polytopes in either two or three dimensional space, an algorithm is given which terminates in a finite number of iterations. It determines either that no collision occurs or the first collision point on the path. Extensive numerical experiments for practical problems show that the computational time is short and grows only linearly in the total number of vertices of the two polytopes.

Accuracy improvement of a self-mixing semiconductor laser range finder is predicted by simulation, in which the laser modulation current is reshaped to give an ideal triangular waveform of the optical frequency change. The maximum range measurement error of less than 0.1% in a wide range of O.1m to 1m is expected by the reshaping of the modulation current. Experimental verification of the effect of current reshaping on the linearization of the derivative of the optical frequency change curve is given.

In this paper, we propose a model-based method of estimating the velocity of a moving object from a series of images. The proposed method utilizes Kalman filtering technique. Assuming that the motion is described by an affine transformation, we construct a discrete-time state variable model of the motion based on the dynamic motion imagery modeling technique proposed by Schalkoff. Using this state variable model, we derive a Kalman filter algorithm. Some simulation results are presented to show that the proposed Kalman filter algorithm is superior to a simple least square method without a model.

This paper describes a system which enables a real-time measurement of 2-D water flow field. One distinctive feature of our system is that velocity vectors of water flow are obtained from the movement of tracer particles at video rate. In order to enable a fast measurement a real time video processor and two Digital Signal Processor(TMS32OC25) are employed. The real-time video processor extracts contours of tracer particles in order to reduce the amount of image data to be processed. And two DSP(Digital Signal Processor) analyse the correlation of every tracer paticle in the consecutive two images to obtain the velocity distribution of water flow.

The air-fuel ratio of an internal combustion engine must be controlled with accuracy for the improvements of exhaust emission and fuel consumption. Therefore, it is necessary to measure the exact instantaneous amounts of fuel and suction air, so we carried out the experiments for measuring the air flow velocity in a suction pipe of an internal combustion engine using three types of instantaneous air flowmeter. The results obtained can be summarized as follows: (1) The laminar-flow type flowmeter is able to measure both the average and the instantaneous flow rate, but it is necessary to rectify the pulsating air flow in the suction pipe. (2) The a spark-discharge type flow velocity meter is able to measure the instantaneous air velocity, but it is necessary to choose the suitable electrode form and the spark character. (3) The tandem-type hot-wire flow velocity meter indicates the instantaneous flow velocity and its flow direction.

The degenerate case of multivariable hyperbolic distributed-parameter systems (systems of hyperbolic partial differential equations) in time coordinate t and space coordinate x is characterized by a property that all the characteristic curves of the state equations are parallel to the coordinate axes of independent variables. It is a disturbing fact, although not well known, that the so-called maximum principle as applied to these systems does not exist for the control that depend on time alone. In this paper, however, it is shown that a set of necessary conditions in the small can exist for unconstrained as well as magnitude constrained controls in a locally convex set. The necessary conditions thus derived can be used conveniently to find the optimal control for degenerate hyperbolic distributed-parameter control systems.

We formulate optimal quadratic regulator problems with trajectory sensitivity terms as a optimization problem for a fixed controller structure. Using well-known techniques for parametric LQ problems, we give an algorithm to obtain suboptimal feedback gains by iterative solutions of two Lyapunov equations. A numerical example is given to illustrate the effectiveness of the proposed algorithm.

This paper proposed a predictive tracking controller for the continuous-time systems by using the receding horizon concept in the optimal tracking control. This controller is the continuous-time version of the previous RHTC (Receding Horizon Tracking Control) for the discrete-time state space models. The problems in implementing the feedforward part of this controller is discussed and a approximate method of implementing this controller is presented. This approximate method utilizes the information of the command signals on the receding horizon and has simple constant feedback and feedforward gain. To perform the offset free control, the integral action is included in the continuous time RHTC. By simulation it is shown that the proposed method gives better performance than the conventional steady state tracking control.

In general it is difficult to determine a Liapunov function for a given asymptotically stable, nonlinear differential equations system. But, in the system with control inputs, it is feasible to make a given positive function, except for a small area, globally satisfy the conditions of the Liapunov function for the system. We call such a positive function a Liapunov-like function, and propose a method of nonlinear optimal regulator using this Liapunov-like function. We also use the periodic Liapuitov-like friction that suits the system whose equilibrium points exist periodically. The relationship between the Liapunov function and cost function which this nonlinear regulator minimizes is considered using inverse optimal method.

This paper presents a method for the optimal control of the distributed parameter systems (DPSs) by a hierarehical computational procedure. Approximate lumped parameter systems (LPSs) are derived by using the Galerkin method employing the Legendre polynomials as the basis functions. The DPSs however, are transformed into the large scale LPSs. And thus, the hierarchical control scheme is introduced to determine the optimal control inputs for the obtained LPSs. In addition, an approach to block pulse functions is applied to solve the optimal control problems of the obtained LPSs. The proposed method is simple and efficient in computation for the optimal control of DPSs.

This paper proposes a navigation strategy of multiple autonomous mobile robots with communication within a specified space. Assuming that each robot has complete detectability with finite range, simple navigation strategy is derived by introducing repulsive forces between robots and attractive force between a robot and its goal point analogous to those between electric charges. When a robot is close to its goal point, a pseudo-domain based on the distance between the closest point of the domain boundary and the goal point is proposed to enhance its convergence to the goal state. This paper concludes with the results of computer simulation studies on the dynamic behavior of multiple interacting robots with the proposed navigation strategy.

Autonomous action, which corresponds actively to the change of conditions in complicated circumstances, is a fundamental function required to an intelligent robot. To develope a control system for a robot having the ability to adapt itself to complicated circumstances, it is necessary to establish self-tracing technology, which recognizes the corresponding position between peripheral objects and itself. So we need to manipulate the moving system with flexibility. It is effective for solving problem that fuzzy theory is adapted to algorithm on a complicated circumstances. We develope a method to generate a route-map which has not only a course from the present position to the destination but also useful information on surroundings.

This paper gives an account of teleoperated mobile robot system which is intended to operate in hostile environments where human access is limited or prohibited. A prototype mobile robot equipped with manipulator was designed and initial tests were made in laboratory environment. Test results, yet preliminary, have been encouraging for further research efforts. Future plans emerging from these initial results are also summarized.

This paper describes the development and control of a quadruped walking robot, named as KAISER-II. The control system with multiprocessor based hierachical structure is developed. In order to navigate autonomously on a rough terrain, an identification algorithm for robot's position is proposed using 3-D vision and guide-mark pattern Also, a simple attitude control algorithm is included using force sensors. Through experimental results, it is shown that the robot can not only walk statically on even terrain but also cross over or go through the artificially made obstacles such as stairs, horizontal bar and tunnel-typed one.

A d.c. servomotor with pulse encoder is used to improve the movement of a hospital mobile robot along the desired line. We can achieve an improved movement of the robot by applying a PLL control. It is then shown that we can also reduce 42% of the power dissipation by the use of a PWM control. Furthermore, some simulation studies are presented to illustrate the design of PI control and optimal regulator for the control of the d.c. servomotor.

We consider a method for enhancement of speech signal degraded by additive random noise with time-variant and/or colored natures. For enhancement of speech signal with such noise, it is effective to utilize the natures of speech and noise. The objective of enhancement of speech is to improve the overall quality and the articulation of speech degraded by the time-variant and/or colored random noise. In the proposed method the distribution model of speech spectrum is given as information to noise reduction system. The proposed system can improve about lOdB in SNR when the input SNR is 0 dB.

This paper describes a new method of noise-canceling in an instrumentation using a pair of identical Potential Transformers (PT). This method allows us to get reliable signals without any noise, not only on the transmission line, but also on the sensors; even if we do not have a reference noise or a specific noise. In this case "any noise" means normal mode noise (NMN), under a quarter frequency of the switching. We proposed to call this method alternating noise-canceling (ANC). The accuracy of this new method has been verified by experimentations.entations.

Sorce noise effect in 1.5 Mach-Zehnder (MZ) interferometer is analyzed. It is shown numerically that with fine adjustments to the feedback gain and initial phase biases, the operating point of the interferometer to achives common mode compensation can be made to lie in a region where the measurand sensitivity is greater than it would be in a conventional Mach-Zehnder interferometer even if the source is less coherent.

Described is a new control signal transmission system which utilizes an optical fiber to transmit 2-bit control signals from the transmitter to receiver. In the transmitter the DC series control voltages are converted into the multiple frequency signals by voltage controlled oscillator (VCO). The multiple frequency signals can easily be transmitted by optical fiber. In the receiver the multiple frequency signals can be detected by analog or digital circuits and then be converted into 2-state control signals which can be used for a variety of applications.

This paper describes the required frequency bandwidth which is used as a measure for the quality of the PFM transmission system to transmit the control signals. First, the PFM signal which has been distorted due to the frequency bandwidth limit on the transmission line is analyzed by Fourier transform and secondly distortion which has been observed at the receiver is numerically analyzed to make the frequency bandwidth required for transmitting the PFM signals. According to the analysis heretofore, the 50% threshold level for shaping the received PFM signal, which is of 50 % in TTL level, is superior than the 10- and 90-% threshold levels.

Presnted is a new optical transmission system to transmit the FM control signal (80 MHz) through an optical fiber by the use of an inexpensive LED having a cutoff frequency of as low as 30 MHz. Studies are carried out on the division of the FM signal frequency at 80 MHz by a factor of four. The FM signal frequency is, after the transmission through the optical fiber, then multiplied by a factor of four so as to obtain the normal FM control signal.

A new controller design algorithm based on the Hessenberg form for linear control systems has een proposed. The controller is composed of the dynamic compensator and the state feedback (dynamic state feedback). The algorithm gives a simple way to assign the eigenstructure (eigenvalues and eigenvectors) of the closed loop system and it also provides a method to assign the frequency shapes near the corner frequencies of the closed loop transfer function matrix. Because of this property, the algorithm is called the independent frequency shape control (IFSC) method.

In this paper, we propose a mthod to decouple the multi-input multi-output two-dimensional system. Then, we analyze the realization dimension of the feedback, feedforward given to decouple. Moreover, we consider the possibility of the reduction of the dynamical dimension needed to decouple. Besides, in order to stabilize the decoupled two-dimensional system, we suggest a method to assign the poles of each entry of the transfer function matrix to the desired positions.

In this paper, we study the pole assignment problem for three-dimensional systems. We transform the denominator of transfer functions of the closed-loop system into the product of three stable one-dimensional polynomials, by performing two-dimensional dynamical feedback and input transformation on the given three-dimensional systems. In the next, we consider the possibility that these two-dimensional dynamic compensators are realizable, thoroughly, and propose the counter-measure in case that they are not realizable. And, we obtain the conditions so that the closed-loop three-dimensional systems are stable. Moreover, we calculate the dynamical dimension which is necessary for the pole assigntmnt, and suggest the pole assignmnt method with the lowest dynamical dimnsion.

In this paper, we consider linear periodic discrete-time control systems under periodic compensation. Such a closed-loop system generally represents a periodic time-varying system. We examine the problem of finding a compensator such that the closed-loop system is realized as LTI model (if possible) with the closed-loop stability being satisfied. We present a necessary and sufficient condition for solving such problem and also give the characterization of realizable LTI models.

This paper proposes a numerical method for redundant manipulators using predetermined optimal resolution. In order to obtain optimal joint trajectories, it is desirable to formulate redundancy resolution as an optimization problem having an integral cost criterion. We predetermine the trajectories of redundant joints in terms of the Nth partial sum of the Fourier series, which lead to the solution in the desirable homotopy class. Then optimal coefficients of the Fourier series, which yield the optimal solution within the predetermined class, are searched by the Powell's method. The proposed method is applied to a 3-link planar manipulator for cyclic tasks in Cartesian space. As the results, we can obtain the optimal solution in the desirable homotopy class without topological liftings of the solution. To show the validity of the proposed method, we analyze both optimal and extremal solutions by the Fast Fourier Transform (FFT) and discuss joint trajectories on the phase plane.

This paper proposes an optimal redundancy resolution of a kinematically redundant manipulator while considering homotopy classes. The necessary condition derived by minimizing an integral cost criterion results in a second-order differential equation. Also boundary conditions as well as the necessary condition are required to uniquely specify the solution. In the case of a cyclic task, we reformulate the periodic boundary value problem as a two point boundary value problem to find an initial joint velocity as many dimensions as the degrees of redundancy for given initial configuration. Initial conditions which provide desirable solutions are obtained by using the basis of the null projection operator. Finally, we show that the method can be used as a topological lifting method of nonhomotopic extremal solutions and also show the optimal solution with considering the manipulator dynamics.

The concept of the manipulability measure of the robotic mechanism is extended to the dual arm holding a single object. This is a measure of manipulating ability of the dual arm forming a closed kinematic chain in positioning and orienting the object. Dual arm manipulability measure is defined and compared to the single arm manipulability measure, and some properties are investigated.

We present a system to measure 3-dimensional coordinates of huge structures like ships, buildings and oil tanks. Two important units are a laser spot projector and a laser spot tracker. Employing a tactful image processing, our system has some features :e.g. compactness, cost, accuracy and robustness to hazardous emvironments.

This paper discusses a force distribution scheme which minimizes the weighted norm of the forces/torques applied on weak points of cooperating multiple robot arms. The scheme is proposed to avoid the damage or unwanted motion of any weak point of robots or object stemming from excessive forces/torques. Since the proposed scheme can be used for either the joint torque minimization or the exerted force minimization on the object, it can be regarded as a unified force minimization method for multiple robot arms. The computational complexity in this scheme is analyzed using the properties of Jarcobian. Simulation of two identical PUMA robots held an object is carried out to illustrate the proposed scheme. By the proper choice of the weighting matrix in the performance index, we show that force minimization for a weak point can be achieved, and that the exerted force minimization on the object can be changed to the joint torque minimization.

Cerebellar Model Linear Associator Net(CMLAN), a kind of neuro-net based adaptive control function generator, was applied to the problem of direct inverse calibration of three and six d.o.f. POMA 560 robot. Since CMLAN autonomously maps and generalizes a desired system function via learning on the sampled input/output pair nodes, CMLAN allows no knowledge in system modeling and other error sources. The CMLAN based direct inverse calibration avoids the complex procedure of identifying various system parameters such as geometric(kinematic) or nongeometric(dynamic) ones and generates the corresponding desired compensated joint commands directly to each joint for given target commands in the world coordinate. The generated net outputs automatically handles the effect of unknown system parameters and dynamic error sources. On-line sequential learning on the prespecified sampled nodes requires only the measurement of the corresponding tool tip locations for three d.o.f. manipulator but location and orientation for six d.o.f. manipulator. The proposed calibration procedure can be applied to any robot.

A digital signal processor(DSP) is applied to realizing a compensator of control system of active magnetic bearings, to restrict a resonance caused by the first-order bending vibration of a flexible rotor, and to run the rotor beyond the critical speed. A full-order observer is applied to the translatory rotor-motion with the first-order vibration mode. A PID control is used for the conical motion. The rotor used in the experiments is symmetric, and an electromagnet and a displacement sensor are set in collocation.

In order to investigate liquid fuel filming over the intake manifold wall, an electrode-type probe has been developed by lines of authors and this probe was employed in a single cylinder two and four-stroke cycle engine and in a four cylinder four-stroke engine operated by neat methanol fuel. The performance of the probe was dependent upon several parameters including the liquid fuel layer thickness, temperature, additive in the fuel, and electric power source (i.e., AC and voltage level) and was independent of other variables such as direction of liquid flow with respect to the probe arrangement. Several new findings from this study may be in order. The flow velocity of the fuel layer in the intake manifold of engine was about (if the air velocity in the steady state operation, the layer thickness of liquid fuel varied in both the circumferential and longitydinal directions. In the transient operation of the engine, the temporal variation of fuel thickness was determined, which clearly suggests that there was difference between fuel/air ratio in the intake manifold and that in the cylinder. The variation was greatly affected by the engine speed, fuel/air ratio and throttle opening. And the variation was also very significant from cylinder to cylinder and it was particularly strong different engine speeds and throttle opening.

A new system for automatic interpretation of the awake electroencephalogram(EEG) was developed in this work. We first clarified all the necessary items for EEG interpretation in accordance with an analysis of visual inspection of the rhythms by a qualified electroencephalographer (EEGer), and then defined each item quantitatively. Concerning the automatic interpretation, we made an effort to find out specific EEG parameters which faithfully represent the procedure of visual interpretation by the qualified EEGer. Those specific EEG parameters were calculated from periodograms of the EEG time series. By using EEG data of 14 subjects, the automatic EEG interpretation system was constructed and compared with the visual interpretation done by the EEGer. The automatic EEG interpretation thus established was proved to be in agreement with the visual interpretation by the EEGer.

The authors proposed a new two-dimensional(2D) positioning system by use of M-array suitable for noisy environment in '88KACC and its revised version in '89KACC. This 2D positioning system is further improved to be used in practice; the computation time is improved by use of vector signal processor and the focussing process is improved by use of an electrically controlled zoom lense. It is shown that this system is robust to noise and also to misalignment of devices.

Recently, the CAD/CAM system to automatically design and process are used in almost every industry world. We designed an original digitizer system to digitize a real size car drawing. We succeeded in the development of super large size Laser Digitizer System (LDS) which has input area of 2m by 6m, resolution of 0.1mm and accuracy of .+-.0.5mm. This Laser Digitizer System can use in design of cars, ships, planes and big maps. Also can use in sensing the position of nozzle head of laser processing system, and so on.

This paper offers the theory and method for regression analysis of the regression model with operational parameter variables based on the fundamentals of mathematical statistics. Regression coefficients are usually constants related to the problem of regression analysis. This paper considers that regression coefficients are not constants but the functions of some operational parameter variables. This is a kind of method of two-step fitting regression model. The second part of this paper considers the experimental step numbers as recursive variables, the recursive identification with unknown operational parameter variables, which includes two recursive variables, is deduced. Then the optimization and the recursive identification are combined to obtain the sequential experiment optimum design with operational parameter variables. This paper also offers a fast recursive algorithm for a large number of sequential experiments.

This paper deals with the fault diagnosis problem in uncertain linear systems having undermodelling, linearization errors and noise inputs. The new approach proposed in this paper uses an appropriate test variable and the difference between system parameters which are estimated by the least squares method to locate the fault. The singular value decomposion is used to decouple the correlation between the estimated system parameters and to observe the trend of parameter changes. Some simulations applied to aircraft ergines show good allocation of the fault even though the system model has significant uncertainties. The feature of the approach is to diagnose the uncertain system through simple parameter operations and not to need complex calculations in the diagnosis procedure as compared with other methods.

An algorithm for multiple fault diagnosis of linear dynamic systems is proposed. The algorithm is constructed by using of the geometric approach based on observation that, when the number of faulty units of the system is known, the set of faulty units can be differentiated from other sets by checking linear varieties in the measurement data space. It is further shown that the system with t number of faults can be diagnosed within (t+1) sample-time units if the input-output measurements are rich and that the algorithm can be used for diagnosis even when the number of faults is not known in advance.

The problem of identification of continuous systems is considered when both the discrete input and output measurements are contaminated by white noises. Using a predesigned digital low-pass filter, a discrete-time estimation model is constructed easily without direct approximations of system signal derivatives from sampled data. If the pass-band of the filter is designed so that it includes the main frequencies of both the system input and output signals in some range, the noise effects are sufficiently reduced, accurate estimates can be obtained by least squares(LS) algorithm in the presence of low measurement noises. Two classes of filters(infinite impulse response(IIR) filter and finite impulse response(FIR) filter) are employed. The former requires less computational burden and memory than the latter while the latter is suitable for the bias compensated least squares(BCLS) method, which compensates the bias of the LS estimate by the estimates of the input-output noise variances and thus yields unbiased estimates in the presence of high noises.

This paper proposes a robust on-line fault detection method for uncertain systems. It is based on the fault detection method [10] accounting for modelling errors, which is shown to have superior performance over traditional methods but has some computational problems so that it is hard to be applied to on-line problems. The proposed method in this paper is an on-line version of the fault detection method suggested in [10]. Thus the method has the same detection performance robust to model uncertainties as that of [10]. Moreover, its computational burden is shown to be considerably lessened so that it is applicable to on-line fault detection problems.

In this paper, we propose a new type of decentralized learning automata for the control finite state Markov chains with unknown transition probabilities and rewards. In our scheme a .betha.-type learning automaton is associated with each state in which two or more actions(desisions) are available. In this decentralized learning automata system, each learning automaton operates, requiring only local information, to improve its performance under local environment. From simulation results, it is shown that the decentralized learning automata will converge to the optimal policy that produces the most highly total expected reward with discounting in all initiall states.

Intelligent Control is an extended paradigm that subsumes both control and AI paradigms, each of which is limited by its own abstractions. Autonomy, as a design goal, offers an arena where both control and AI paradigms must be applied -and a challenge to the viability of both as independent entities. We discuss hierarchical event-based control architectures in which AI and Control paradigms can be integrated within a model-based approach. In a niodel-based system, knowledge is encapsulated in the form of models at the various layers to support the predefined system objectives. Concepts are illustrated with a robotmanaged space-borne chemical laboratory.

Japanese manufacturers have seen the market swing from a sellers' market to a buyers' market by dint of changes in the economic environment. Therefore, they have found the need to establish a production system in which products can be manufactured at the speed at which they are being sold and which achieve a competitive advantage. FA activity in Japan today has begun to change from local automation to totally integrated automation. This paper introduces a bottom to top based JIT-FA system and describes how to approach and introduce a CIM supported by human based systems. A fully automated CIM system is not only flexible but also rigid. However we can realize a flexible production line system by partially introducing human based systems including a decision support system.

This paper discusses an optimal cyclic scheduling problem for a FMC (Flexible Manufacturing Cell) modeled by a two-machine flowshop with two machining centers with APC's (Automated Pallet Changers), an AGV (Automated Guided Vehicle) and loading and unloading stations. Cyclic production in which similar patterns of production is repeated can significantly reduce the production lead-time and WIP (Work-In-Process) in such flexible, automated system. Thus we want to find an optimal cyclic schedule that minimizes the cycle time in each cycle. However, the existence of APC's as buffer storage for WIP makes the problem intractable (i.e., NP-complete). We propose an practical approximation algorithm that minimizes, instead of each cycle time, its upper bound. Performances of this algorithm are validated by the way of computer simulations.

A large number of trainings are requested for the artificial neural network using the backpropagation algorithm. It is shown that one dimensional search technique is effective to reduce the number of trainings through some numerical simulations.

During the past two decades, a lot of researches have been done on the synthesis of grassroot heat exchanger networks(HEN). However, few have been dedicated to retrofit of existing heat exchanger networks, which usually use more amount of utilities (i.e. steam and/or cooling water) than the minimum requirements. This excess gives motivation of trades-off between energy saving and rearranging investment. In this paper, an algorithmic-evolutionary synthesis procedure for retrofitting heat exchanger networks is proposed. It consists of two stages. First, after the amount of maximum energy recovery(MER) is computed, a grass-root network featuring minimum number of units(MNU) is synthesized. In this stage, a systematic procedure of synthesizing MNU networks is presented. It is based upon the concept of pinch, from which networks are synthesized in a logical way by the heuristics verified by the pinch technology. In the second stage, since an initial feasible network is synthesized based on the pre-analysis result of MER and must-matches, an assignment problem between new and existing units is solved to minimize total required additional areas. After the existing units are assigned, the network can be improved by switching some units. For this purpose, an improvement problem is formulated and solved to utilize the areas of existing units as much as possible. An example is used to demonstrate the effectiveness of the proposed method.

A prototype integrated system and its theories for distributed SISO control structure synthesis of complete chemical plants is developed. The scope of this work includes control structure synthesis not only of simple units with unspecified control loops but also of the complex process at preliminary and basic design stage. Hierarchical approach and dual-decomposition strategy (that is multi-layer decomposition and multi-echelon decomposition) is applied to this system. Because automatic control structure synthesis of complex plants is a problem defined as a series of knowledge-intensive tasks within multiple spaces, the established methodology is complemented by not only techniques from knowledge-based expert systems but also shortcut and rigorous control theories. This system is used for education of control designers, process engineers, operators and students as well as for operability studying, in-line and on-line process control structure synthesis.

The paper describes and presents the experimental results of a fuzzy control scheme applied to an electric power system. The fuzzy control scheme which was recently proposed by T. Hiyama and T. Nakano [1] is implemented on a digital processor with A/D and D/A converters and is tested on our laboratory system. The results show that the proposed fuzzy controller with optimal settings of its parameters gives better performance than the PSS (Power System Stabilizer) optimally adjusted. The problems encountered throughout the experiment are also discussed.

The iterative learning control synthesized in the frequency domain has been utilized for temperature control of a batch reactor. For this purpose, a feedback-assisted generalized learning control scheme was constructed first, and the convergence and robustness analyses were conducted in the frequency domain. The feedback-assisted learning operation was then implemented in a bench scale batch reactor where reaction heat is simulated using an electric heater. As a result, progressive reduction of temperature control error could be obviously observed as batch operation is repeated.

In this paper, we are going to study the stabilization of the semilinear heat equation with inhomogenous boundary conditions, whose solutions are not (in general) stable. Here, we use the discrete-time feedback inputs through the boundary of geometric domain to the semilinear system under some additional conditions and assumptions. It is shown that under these conditions, the stabilization can be realized by applying pole assignment argument to the principal linear part of the system and that the solutions exist globally in discrete-time t without any finite escape time.

When a saturating control system has integral action, reset windup can cause instability as well as make the system performance unsatisfactory. An antirset-windup (ARW) limiter has been suggested to improve the stability and performance. It has been implemented with analog circuits and tested by simulations. This paper presents the stability condition of a double-integrator plant having the state feedback plus integral-action controller with the ARW limiter by using both Liapunov's second method and graphical method together.

A new method called random sampling method has been proposed for generation of binary random sequences. In this paper, a new concept, called merit factor Fn, is proposed for evaluating the randomness of the binary random sequences generated by the random sampling method. Using this merit factor Fn, some desirable conditions are investigated for uniform random numbers used in the random sampling method.

A VSS observer-based sliding mode control is described for continuous-time systems with uncertain nonlinear elements, in which the Euclidean norm of unknown element is bounded by a known value. For a case of complete state information, we first derive a sliding mode controller consisting of three parts: a linear state feedback control, an equivalent input and a min-niax control. It is then shown that the present attractiveness condition is simpler than that for a case without using the concept of equivalent input. We next design a VSS observer as a completely dual form to the sliding mode controller. Finally, we discuss a cas of incomplete state information by applying the VSS observer.

Nanometer positioning control with high velocity and long stroke is discussed. A one-axis stage mechanism driven by an AC linear motor and guided by a rolling ball guide has been constructed. Coarse and fine position controls are designed by using nonlinear dynamics of the rolling guide. Switching from coarse positioning to fine positioning is studied.

Molten steel level information of ladle is very important for process control in steelmaking process. At secondary refining process, measuring lance and snokel have to keep constant thier depth from molten steel surfaces. But, there is much slag on the molten steel surface. Besides, not only the thickness of slag is varied with refining condition, but also molten steel level is largely affected by firebrick errosion. Then, optical measuring method and/or by human eyes cannot detect true molten steel surface, but slag surface. This slag thickness is 300mm at maximum, then huge diameter eddy current sensor will be needed if that type sensor is applied. In addition to, cooling system is necessary because the molten steel and slag temperature is high. This is not practically. To solve this problem, immersion type levelmeter is developed. This sensor is made up from primary and secondary cylindrical coils. High frequency current is applied to primary coil. Electro-motive force from secondary coil is measured, which is varied with molten steel level. This complete set is installed within stainless steel long capsule and attached to top of lance. This sensor is immersed into molten steel bath of ladle or tundish with protection of expendable paper sleeve.

In this paper, compliant motion control of a manipualator in manipulator is proposed by using the self-tuning adaptive controller. Compliant motion is needed in order to applicated to complicated and accurate fields such as assembly operation in which several parts are matched. For a control method of compliant motion hybrid control is used so forces and position control are proposed selectively through a closed feedback loop. By contacting with environment, the uncertainties higher. Self-tuning controller which adapts to variable dynamic response is applied to compliant motion control in order to satisfy the desired operation. The applicability of the suggested algorithm was confirmed by simulation of the contour tracking task of four joint manipulator.

A new dual loop controller using color LCD bar graphs with LED back lights has been developed. An optional memory card is used to load or save the controller configuration, which may be a preprogrammed standard package or a user-programmed configuration, in addition to the built-in functions ready for user selection. The bar-graph display is selectable for single-loop or dual-loop use. A high grade of self-tuning functions using a modeling technique is built-in as standard. The controller can accommodate optional plug-in modules for thermocouples, communication, etc. All the options are fully field upgradable.

This paper deals with some analytical and experimental aspects to control a multi-magnet suspended vehicle. Because the response of a multi-magnet vehicle shows mutually coupled interaction, an analytical description of the vehicle dynamics is necessary. For numerical computations, a linearized modelling of vehicle dynamics is dicussed and computer simulation is carried out. And for the experiment, a test vehicle suspended by four magnets has been made and investigated by local control of each magnet. Two algorithms by PID and state feedback control law are used and compared with each other. Some kinds of disturbance characteristics and coupling effects of the width change of the test vehicle are experimented.

This paper deals with control system design strategy for electrolmaginetic suspension (E.M.S.) system. For a successful control of E.M.S. system, the nature of E.M.S. system is deeply studied in the view point of non-linear, open-loop unstable, time-varying, non-minimum phase system. To find a special control treatment for E.M.S. system, analyses and simulations for various models are carried out. As one of the successful candidates, adaptive control concept is introduced and sample hardware system using digital signal processor is implemented.

The architecture and functions of a prototype free ranging AGV system are described in this paper. The system has single tricycle configuration - the front wheel is driven and steered simultaneously. The primary position measurement device of this system is the redundant encoder system - an absolute encoder for the steering angle measurement of the front wheel, two incremental encoders for the measurement of the rear wheel rotations. The secondary position measurement device is implemented to reduce the accumulatad error in encoder measurements. The extended Kalman filter is suggested to combine the conflict measurement data for the proper position estimation.

This paper deals with the control of autonomous vehicle in the production systems. Presently, there is a significant interest in autonomous vehicles which are capable of intelligent motion (and action) without requiring a guide track to follow. This paper describes a PI-F adaptive control algorithm, which is used to drive an experimental autonomous vehicle along a given trajectory. The simulation results characterizing the accuracy og the algorithm are presented.

This paper describes the application of the recently developed feedback linearization technique to the design of a new command to line-of-sight (CLOS) guidance law for skid-to-turn (STT) missiles. The key idea lies in converting the three dimensional CLOS guidance problem to the tracking problem of a time-varying nonlinear system. Then, using a feeedback linearizing approach to tracking in nonlinear systems, we design a three dimensional CLOS guidance law that can ensure zero miss distance for a randomly maneuvering target. Our result may shed new light on the role of the feedforward acceleration terms used in the earlier CLOS guidance laws. Furthermore, we show that the new CLOS guidance law can be computationally simplified without performance degradation. This is made possible by dropping out the terms in the new CLOS guidance law, which obey the well-known matching condition.

In order to tackle global environmental problems such as destruction of the ozone layer or climatic changes due to atmospheric temperature increase, the acquisition of plentiful and precise data is necessary. Therefore, a means of conducting long-lasting high-resolution measurements over broad areas is required. A feasibility study has been made on a high altitude (20km), super-pressured helium-filled PLTA (Powered Ligher-than-Air) vehicle as an ideal platform for environmental observation. It has a long service life and carries a larger payload than an artificial satellite. This PLTA platform uses an electric propulsion system to maintain position in space against wind currents. The thruster is driven by solar power acquired from solar cells. For night use, solar energy is stored in regenerative fuel cells. This study focuses on energy balance and structural analysis of the hull and platform. The platform is capable of conducting high resolution remote sensing as well as having the capability to serve as a telecommunications relay. The platform could replace a number of ground-based telecommunications relay facilities, guaranteeing sufficient radio frequency intensity to secure good quality telecommunication transmittal. The altitude at which the platform resides has the lowest wind flow in the lower stratosphere, and permits viewing from the ground within a 1,000km range. Because this altitude is much lower than that required of an artificial satellite, the measuring resolution is a couple of thousand times higher than with artificial satellites. The platform can also be used to chase typhoons and observe them from their sources in tropical regions.

We have developed a new fuzzy control method for paper machine basis weight profile. The conventional linear control method has not yielded good results on some machines. This new control method, however, realizes long-term stability and convergence of the profile as good or better than that achieved under manual control by an operator.

The paper presents a method of constructing simple fuzzy control rules for the determination of stabilizing signals of automatic voltage regulator and governor, which are controllers of electric power systems. Fuzzy control rules are simplified by considering a coordinate transformation with the rotation angle .theta. on the phase plane, and by expanding the range of membership functions. Also, two rotation angles .theta. $_{1}$ and .theta. $_{2}$ are selected for the linearizable region and the nonlinear one of the system, respectively. Here, .theta. $_{1}$ is chosen by the pole assignment method, and .theta. $_{2}$ by a performance index. Fuzzy inference is applied to the connection of two rotation angles .theta. $_{1}$ and .theta. $_{1}$ by regarding the distance from the desired equilibrium point as a variable of condition parts. The control effect is demonstrated by an application of the proposed method to one-machine infinite-bus power system.

In Multi-Layer Perceptron (MLP) which realizes continuous mappings, the output errors is directly affected by the weight errors which may be caused by the limited precision of digital or analog hardware in implementations. So, it is important to study the sensitivity due to the perturbation of connection weights between neurons. In this paper, we derive a sensitivity function to the statistical weight perturbations in MLP with differentiable activation functions. This sensitivity function can be regarded as an ensemble average of deterministic sensitivity measures due to the perturbations of weights. Hence, this sensitivity function can be used as the criteria for selecting weights with the minimum sensitivity among possible sets of connection weights in MLP. For the verification of the validity of the proposed sensitivity function, computer simulations have been performed and through the simulations we find good agreement between the theoretical and simulation results.

Fuzzy theory may be constructed of Fuzzy sets theory, Fuzzy measurement and Appoximate reasoning, and nowadays the reseaches on it are done widely in both theoretical and practical aspects. In this paper, the authors sought for the subject in Hirota's Fuzzy CAI: A decision making simulated fuzzy problem on the speed and accelerator control of a car driven in the high way. Interesting results on these fuzzy solutions of this problem, based on a defuzzification with 9 kinds of the methods, are reported. And the characters of these solutions, linear or nonlinear and their fuzzy control evaluations are stated.

In this paper, an iterative learning control algorithm for unknown linear discrete systems is proposed by employing a parameter estimator together with an inverse system model. Regardless of initial error and inherent parameter uncertainty, a good tracking control performance is obtained using the proposed learning control algorithm characterized by recursive operations. A sufficient condition for convergency is provided to show the effectiveness of the proposed algorithm. To investigate the performance of the algorithm a series of simulations and experiments were performed for the tracking control of a servo motor.

A new control scheme is proposed in this paper which can cope with varying environment as results of load disturbances, changes of plant dynamics, and failures of components. The objective of this paper is to blend numeric-to-symbolic conversion techniques with linear conventional controllers so as to adapt to the varying environment of the system. The control scheme is based on the parametrization of stabilizing controllers, which is called Kucera/Yula parametrization. The parametrization has been extended to the class of systems which contain numeric-to-symbolic converters. It is shown how the numeric-to-symbolic converters can be blended with the linear controllers.

There is an increasing need to apply artificial intelligence to the real application fields of industry. These include an intelligent process control, an expert machine and a diagnostic and/or maintenance machine. These applications are implemented in AI Languages. It is commonly recognized that AI Languages, such as Common Lisp or Prolog, require a workstation. This is mainly due to the fact that both languages need a large amount of memory space and disk storage space. Workstations are appropriate for a laboratory or office environment. However, they are too bulky to use in the real application fields of industry or business. Also users who apply artificial intelligence to these fields wish to have their own operating systems. We propose a new design method of an intelligent controller which is embedded within equipment and provides easy-to-use tools for artificial intelligence applications. In this paper we describe the new design method of a VMEbus based intelligent controller for artificial intelligence applications and a small operating system which supports Common Lisp and Prolog.

Generally, the required control functions for the industrial auto trimming sewing machines are the sewing speed control with pedal input, the up/down stopping-position control of the needle, the automatic sewing according to the memorized sewing pattern including the number of stitches, and etc. We developed a new type of AC servo motor controller, which suffices for all the above functions. The developped controller is working well, and the performances are very good for the practical use.

A method of speed sensorless vector control of induction motor is proposed in this paper. This method uses the slip frequency estimated by only the primary voltage and current. As this slip frequency accords with the command value of it, the commanded primary frequency is controlled. The validity of the method is confirmed by the simulation and experimental results.

A novel type of a play-back servo system with high precision is designed using an iterative learning control method by employing the model algorithmic control concept together with an inverse model. A sufficient condition is also provided for the convergency. It is shown by simulation that the proposed control algorithm yields a good performance even in the presence of a periodic load disturbance and proved by experiments using microprocessor-based play-back servo system.

DC series motor has been widely applicated in industry due to many advantageous characteristics, such as high starting torque, easy construction of its controller, and cost economy, etc.. However, by high starting current, excessive surge voltage, and so on, many problems which could make engineers relinguish to use it are induced. In this paper, various protection methods for power circuit are discussed. Particularly, a new proposed snubber circuit which consists of two diodes, one capacitor, and a resistor increases the performance with respect to the suppression of the surge voltage. Furthermore, the plugging algorithm, by checking the armature current and voltage and controlling the field coil current, is designed and implemented. Also, these methods and algorithms were applicated in electric vehicle, and we could find its stability to be considerably improved.

This paper presents an indirect adaptive control scheme for discrete linear systems whose parameters are not necessrily slowly varying. It is assumed that system parameters are modelled as linear combinations of known bounded functions with unknown constant coefficients. Unknown coefficients are estimated using a recursive least squares algorithm with a dead zone and a forgetting factor. A control law which makes the estimated model exponentially stable is constructed. With this control law and a state observer, all based on the parameter estimates, it is shown that the resulting closed-loop system is globally stable and robust to bounded external disturbances and small unmodelled plant uncertainties.

Using eye mark recorder and polygraph, basic laboratory experiments have been conducted to investigate the validity of psycho-physiological measures to the analysis of on-line cognitive information characteristics at man-machine interface (MMI). It is concluded that various psycho-physiological measures are useful to estimate various aspects of human cognitive characteristics and thus to the evaluation of MMI designing from the viewpoint of its harmony to the human cognitives.

Even though the concept "Adaptive" was introduced in the late 50's, the main contribution to adaptive and/or self-tuning control has been made since late 70's. This paper describes the feature of adaptive control simulation package KERICON(I) developed in KERI. Informations on hardware environments, install and testing of a new algorithm and user interfacings are also summarized. The package is written in C language and currently being updated for expert-type adaptive control package (KERICON II).RICON II).

The system with softfware packages for control system design unifying and encompassing rule based control and conventional control based on numerical models were developed. Users who are not familiar with control theory, numerical computing, and artificial intelligence (AI) can perform system analysis, control design and development of AI control system without difficulty.