A circular holographic sonar utilizing an inverse problem solution is proposed in this paper. First, a parallel image reconstruction algorithm is developed, which has block-circulant characteristics. Second, through the several experiments, we show this holographic sonar can reconstruct an exact image with high resolution. Finally, a circular systolic array processor executing the image reconstruction algorithm is proposed.
This holographic sonar is based on three techniques, circular holography, inverse problem solution and systolic array processing. It has following features: (1) A circular array construction enables to obtain a complete azimuthal holographic data and to reconstruct an exact azimuthal image. (2) An inverse problem solution produces the much higher resolution than one which a conventional back propagation produces. (3) An isoazimuthal image reconstruction procedure enables to design a circular systolic array processor, which can reconstruct an image in real-time.
The estimated time for image reconstruction of 65, 536 points (256 sectors×256 points) is about 0.4s using this circular systolic array processor. From this estimation, this holographic sonar has an effectiveness for a robotic sensor system.

View full abstract

An imaging tactile sensor for the measurement of pressure distributions and a real-time tactile image processing system using the sensor are described.
The sensor has a 64×64 array of sensitive sites using pressure-conductive rubber and electrodes on PC board. The scanning circuit of the sensor realizes precise and high speed scanning, 16.7ms per one frame, using a FET at every site for switching. As the output of this sensor is a video signal, pressure distributions can be visualized by a monitor TV and can be recorded by a video tape recorder.
By connecting the sensor with a high speed image processor, tactile image processing has been realized. In this paper, showing some examples such as smoothing, contour, trace, and adaptation of tactile images, features of tactile image processing are definitely shown by comparing with visual image processing. This system is used as a developing system of tactile processing algorisms as well as for the measurement of pressure distributions of human body with beds or chairs in the field of human engineering.
In this paper, the structure of the sensor and the architacture of the system are described and experimantal results are shown.

View full abstract

A new high speed photometric ellipsometer consisting of simple optical elements, electronics, and personal computer is developed. Data acquisition time of the new ellipsometer is only limited by the response time of the photodetector's circuitry. This ellipsometer, having no moving parts, is suitable for measuring film thickness moving fast in the process line, where no conventional ellipsometer can be applied.
In this proposed apparatus, a certain polarized light (ex. linearly or circularly) is used as an incident beam and the reflected beam from the object is split into 3 beams through the beam splitter which consists of 3 or 4 optical pararells. The intensity of each beam transmitted through the analyzer with a certain azimuth angle is measured by the photodetector.
Each intensity signal multiplied by the gain factor for each channel, denoted by I₁, I₂, I₃, can be used to derive the two ellipsometric parameters. The analysis with Jone's vector representation shows that the ellipsometric parameters cosΔ, tanψ are expressed as follows; cosΔ=I₃-I₂/2I₁√I₁/I₂+I₃-I₁, tanψ=σ₁σ₂√I₁/I₂+I₃-I₁where incident beam is linearly polarized (azimuth -45°), the azimuth angle of each analyzer is 0° (ch. 1), 45° (ch. 2), -45° (ch. 3) respectively, and σ₁σ₂ is a constant determined by the refractive index and the angle between the reflected beam and the beam splitter.
The precision of cosΔ and tanψ by this ellipsometer is on the order of 5.0×10^-4.
This new ellipsometer is applied to measure oil film with thickness of order 0-100Å coated on tinned steel sheet moving at a speed of 300m/min. In this case, the following considerations are further taken into account.
1. Linear relation between ellipsometric parameter cosΔ and film thickness d is used.
2. Two lasers (He-Ne laser, Ar laser) with different wavelength λ₁, λ₂ are used to eliminated the ellipsometric parameter cosΔ₀ of substrate which changes continuously.
Oil film thickness measured by the new ellipsometer and those by conventional off-line method (hydrophil balance method) are agreed within the uncertainly of 1mg/m² (corresponding thickness is about 11Å).

View full abstract

A new method for measuring the delay time is proposed. If the observed signal is regarded as an output signal of the delay plus minimum phase system, its delay time can be estimated by use of the Hilbert transform relationship which holds between the log magnitude and phase of the transfer function of the minimum phase system. Discussions are mainly focused on the practical case where the observed signal is band limited. A procedure for extrapolating the frequency characteristics of the signal is presented to minimize the estimation error due to the bandwidth limitation. The method is applied to the measurement of the boundary location of acoustic materials whose reflection characteristics are minimum phase. The case of non-minimum phase system is also discussed and the boundary location of a turf-covered ground is estimated. Using this result, the acoustic impedance of the turf-covered ground is computed.

View full abstract

A new design method of simple discrete time adaptive servo system for positioning system is presented. It is assumed that the controlled system has a D.C. servo motor like characteristics with unknown inertia load and varying friction.
The proposed method is based on an adaptive pole assignment system designed for the simplified model of the D.C. servo moter.
In this paper, first, a simplified discrete time model of the D.C. servo motor is introduced.
Next, a stable adaptive pole assignment system with tracking compensator is designed for the model. Moreover, the robustness of the proposed adaptive control system is discussed for the case where the modeling error is exist.
Finally, the proposed method is successfully applied to a two-axis positioning system and the advantages of using adaptive controller are demonstrated by experimental results.

View full abstract

Landau and Lozano proposed a design scheme of discrete-time model reference adaptive control system (MRACS) using low-pass filters to alleviate the influence of noises. We also presented an alternative scheme which is obtained by modifying their scheme regarding to the synthesis of control input in order to improve the robustness to parasitics.
This paper discusses the robustness of above two schemes in the presence of parasitics and makes clear the following properties. The scheme by Landau and Lozano does not always contribute to improvement of the robustness to parasitics, because it needs inverse operation of low-pass filters to generate the control input. On the other hand, since our scheme does not require such inverse filter operation, it is superior to the one by Landau and Lozano with respect to the robustness to parasitics. These results are verified by the use of linearization technique and also confirmed by simple examples and simulation studies.

View full abstract

RH^∞-Optimization problem of minimizing cost function ||T-UV||_∞ is considered in the case of T and U being given square transfer matrices and V to be found. The result to the problem is already known in the literatures. The present derivation is, however quite elementary. Only simple matrix manipulations and results in linear system theory are used.
The problem is solved in state space, based on the definition of norm ||·||_∞. Firstly, with the state space realization of U^-1 and U^-1T, a lower bound σ of cost ||T-UV||_∞ is found. Then it is shown that there exists a V_s∈RH^∞, parameterized as V_s=D⁺_sN_s with free parameter matrix S, satisfying inequality ||T-UV||_∞≤σ, which, therefore, implies that σ is the minimum. Finally, the fact that all V'_s∈RH^∞ minimizing ||T-UV||_∞ take the form V_s will be proved.

View full abstract

The problem of optimal smoothing for noncausal two-dimensional (2-D) systems is formulated and solved based on the 2-D descriptor (generalized state-space) representation.
The problem is to obtain the linear least square estimate of the random signal corrupted by observation noise. The random signal to be estimated is the output of a noncausal 2-D shaping filter driven by white noise.
It has been shown by the authors that a noncausal 2-D system whose two-sided (z, s)-transform is rational can be represented by a 2-D descriptor system.
In the paper, after formulating the best estimation problem by describing the random signal as the output of a 2-D descriptor system driven by white noise, we obtain a set of linear equations of the best estimate.
Next, we convert the set of linear equations into the linear equations over a commutative algebra of two-sided Z-transforms.
We show that under some assumptions on the 2-D descriptor realization of the random signal shaping filter, the above equations can be solved iteratively for the best estimate of that signal.
Finally, a numerical example and simulation results for design of optimal smoothers based on the proposed method are presented.

View full abstract

The approach of singular perturbation is useful for the problems of analysis and design of control systems model by using the small parameter. For example they are appeared in systems which have parasitics elements or high gain feedback. Kokotovic and his co-workers studied the convergence of the solution and behavior of the poles, and proposed the pole assignment algorithm. But their methods are restricted in the meaning that the reduced systems, which is obtained by ignoring their parasitics, must be proper. But recently descriptor form, a generalized representation of state space form, are widely treated by many authors, and analysis and design tools for improper system have been proposed.
The purpose of this paper is to extend the traditional results of singular perturbation method on the base of descriptor form. The restriction that the reduced system must be proper is removed, and a new algorithm to specify the pole location of singular perturbed systems is proposed. Two associated descriptor systems, which are named as the reduced system and the boundary-layer system, and the impulse index play an important role. The main results of this paper are as follows.
1) When the impulsive index is zero, the behavior of the singular perturbed system is separated into those of the two associated descriptor systems.
2) A sufficient condition to make the impulse index of closed loop system zero is that the reduced system of the open-loop system is impulse controllable.
3) The necessary and sufficient condition of the arbitrary pole assignablility is both the reduced system and the boundary-layer system are exponential mode controllable.
The proposed algorithm is consisted in numerical oriented routines and applicable in many areas.

View full abstract

A position servo system which operates at a constant speed is often used in a mechatronics equipment. In this paper, a sliding mode control scheme for constant speed operation is presented.
The scheme needs only the encoder signal to obtain not only the position but also the velocity with very small delay. Thus it enables to construct a fully digital sliding mode servo system with reduced chattering. Sufficient conditions to assure the existence of the sliding mode and the stable operation regardless of the initial state is derived to give the design method.
The scheme is applied to a fully digital control paper feeding servo system of a laser beam printer. The servo system has a DC servo motor and only an encoder as the transducer. The results of numerical analyses, an operation experiment, and a print experiment are given.

View full abstract

This paper presents an on-line optimal control system of particles cooling process in fluidised bed. First, the characteristics of on-line electric-resistent type moisture sensor has been investigated for the direct feedback of controlled variables. Second, the multivariable control systems were constructed for the case that measured variables are coincident with the output ones to be controlled and the sampling interval is long. LQI and decoupling PID control approaches are applied to both temperature and moisture control of the foundry sand discharged from fluidised bed. Thirdly, the observer theory was applied to multivariable control systems for the case that measured variables are not coincident with the controlled variables and the sampling interval is short. Finally, a series of control experiments have been conducted to compare the usefulness of several control system used in this paper, and an optimal on-line control system has been proposed.

View full abstract

Fuzzy decision method deals with the ambiguous decision problems applying fuzzy sets theory. It is an important issue for fuzzy decision making to define the preference relation between outcomes represented by fuzzy sets.
In this paper, based on the model concept such as possibility and necessity, fuzzy preference relation between elements is extended to six relations between fuzzy sets. The preference relation extended by Orlovsky and Baldwin and Guild is a special case of these relations. The proposed relations are corresponding to the rankings of fuzzy numbers proposed by Dubois and Prade when the fuzzy preference relation is the large and small relation. The properties of these six preference relations are investigated. These relations preserve the properties between possibility and necessity, which is satisfied in modal logic, for the fuzzy sets whose membership functions attain 1. Since the preference relations extended in this paper are all based on the modal concept, it is possible to obtain the decision procedures reflecting the various modalities of decision maker.

View full abstract

This paper describes a new optical coupling device for an airborne digital redundant system.
For the on-board system, it is very important to increase reliability and safety, then a redundant system is applied. In the redundant system, in order to verify if each operation is correct or not, it is necessary to compare with the same time data of the other systems.
We developed a new characteristic coupling device and verified its features experimentally.
This system has the following characteristics.
(1) it has a simple hardware architecture in order to increase reliability and safety.
(2) synchronization and data exchange are separated in time, so this system utilizes the same hardware for these tasks.
(3) for the increase of transfer speed, a broadcast method is used and the number of receivers is (redundancy-1), then each system can transmit and receive simultaneously.
(4) an optical signal is used in order to increase transfer speed and this is effective for the electric separation among systems.
At last it is discussed for a digital computer how to be able to know the transfer end without the interrupt signal.

View full abstract

Small ball which has several nanometers in diameter is generated at the tip of fine gold wire. The ball size is measured in process just after formation in bonding process of IC fabrication.

View full abstract

In this paper, we investigate solvability, stability, and inverse systems of noncausal 2-D systems. A new definition of stability for noncausal 2-D systems is introduced using the notion of two sided (z, s)-transformation. Solvability and inverse construction are described based on 2-D descriptor (generarized state-space) systems.

View full abstract

This paper derives the general solutions to the deadbeat problem in terms of Λ-generalized polynomials, and necessary and sufficient conditions for the existence of deadbeat controllers are directly established from the solvability of the polynomial matrix equations.

View full abstract

This short paper describes a design method and its application result of a low order compensator.
This method is replaced the state feedback system designed by optimal regulator theory with a low order compensator.

View full abstract

This paper proposes a design method of control systems based on Laguerre function model representation. This method makes it easy to determine the parameters of two-degree-of-freedom compensator which satisfy stability and partial model matching conditions.

View full abstract

In this paper, we propose a new control method of the switching devices in the inverter system. The simulation results reveal that the output waveform is close to a sinusoidal input waveform for its fundamental part is corresponding to the sinusoidal input and its harmonics parts are completely eliminated. Furthermore the influences of disturbance can be eliminated.

View full abstract