This paper describes the development of temperature sensor based on a temperature-sensitive quartz “tuning fork” resonator. The resonator was fabricated using photolithography and anisotropic etching. The optimum cut of the crystal and optimum resonator orientation on the wafer were determined-taking into account the anisotropic etching properties of the crystal-to give a large temperature coefficient of frequency consistent with small equivalent series resistance.
The tuning fork temperature sensor operates over a wide range of temperature (from 4.2K to 250°C). After the temperature sensor was cycled between 0°C and 4.2K, hysteresis in the temperature characteristic was only within 0.005K at 0°C. The time constant of thermal response is about 0.9s. The quartz temperature sensor is quite compact-it is mounted in a hermetic case only 3mm in diameter.

View full abstract

When two-color pyrometers are used in order to determine the true temperature of a subject, the difference between a calculated true temperature (T'), which is calculated from the measured two-color temperature by using two effective wavelengths, and the true temperature (T), in other words, “a range of the absolute value in error calculated (ΔT)”, is very important in discussing the practicality of the two-color pyrometer.
The effects which the spectral characteristic of two-color pyrometer i.e. the measured band width of wavelength has on ΔT above mentioned were discussed.
The measured band width of wavelength determined by the spectral characteristic of the detector and the filters is classified into a narrow, a medium, and a broad bands. In spite of the change of the temperature and the spectral emissivity of the object, the narrower is the measured band width of wavelength, the smaller are the shift of the effective wavelengths and ΔT. But the medium band two-color pyrometer proves to be fully practical in accuracy to determine the true temperature. If there is the absorption band attenuating the measured light of a certain wavelength in the measuring light path having consideration for the practical measuring condition, it directly causes ΔT to be larger.
On the other hand, the broader is the measured band width of wavelength, the larger is ΔT. The examinations of the general expression to reduce ΔT bring us only a too troublesome one for practicality. ΔT, however, isnot liable to be effected by the absorption band of the measuring light path in using the broad band two-color pyrometer. Then the intense signal can be obtained.
Consequently, it is advantageous to broaden as much the measured band width of wavelength as possible within range where ΔT can be practically neglected; and the band width is proper when the characteristic obtained by composing the spectral characteristic of the filters and of the detector is approximated to an equilateral triangle, and its half band width is set up about 0.08μm and then the measured band width of wavelength is about 0.16∼0.20μm (medium band).

View full abstract

In big cities today, there are many tall buildings which can accommodate a large number of people, and it has become an important issue to know how many people are currently in a building for safety reasons, for example. This has yielded a need for accurate counting of pedestrians who enter and leave such a building.
For accurate measurement of the number and the direction of pedestrians, a two-dimensional image pattern recognition method using an image pickup system such as a TVcamera system has been known. This method, however, has a problem in the real time processing if the quantity of image information obtained is too much, which is likely in many cases.
Basically, the number and the direction of pedestrians can be known by monitoring a pair of parallel lines across the pedestrian way. The authors have developed a method for bidirectional pedestrian traffic measurement which uses a pair of linear charge coupled devices. In this method, the CCDs monitor two parallel lines from above the pedestrian way and the number and the direction of pedestrians are measured based on the two-valued one-dimensional CCD image outputs.
This paper discusses the basic measurement conditions required for real time processing and for accurate measurement of multiple pedestrians passing in line, that is, the interval between the two monitor lines, the data sampling period, and the pedestrians recognition criteria.
The results showed that the interval between the monitor lines should be 8.5cm, that the data sampling period must not exceed 19.3ms, that the width along the monitor line to be recognized as a pedestrian should be 20.0 to 71.1cm when only one pedestrian passes and 23.7 to 71.1cm when two or more pedestrians pass side by side, and that the chest thickness should be 8.5 to 36.9cm when only one pedestrian passes and 12.3 to 36.9cm when two or more pedestrians pass one behined another. With these settings, the theoretical measurement error can be 0.3% or less.

View full abstract

For the production of high quality images from synthetic aperture radar (SAR) received signal data, a method of suppressing and removing the degradation of SAR image range resolution caused by range migration is proposed and verified.
In spaceborne SAR systems, the response signal from each point on the earth surface spreads over several hundred range bins due to a spatial distortion known as range migration. This leads to the degradation of range resolution of SAR images produced by appling one dimensional matched filters in range and azimuth directions in the frequency domain.
A conventional method to solve the above problem is to apply a time invariant inverse filter to range compression in the frequency domain by neglecting that the degradation depends upon Doppler frequency and range distance of responce signal from each point on the earth surface. This can suppress only the bias term of the degradation of the SAR image resolution in the range direction.
In this paper, in order to remove the higher order terms of the degradation, the time variant inverse filter is additionally applied to azimuth compression in the time domain. This filtering is realized as the resampling to correct range migration in the azimuth compression. The filter function can be expressed as a three dimensional look up table. In this way the time variant filtering requires little additional computation load.
It is verified by the experiment using simulated SEASAT SAR data that for a look with maximum range migration the proposed method improves 2.0% of range resolution of SAR image against 1.3% which is obtained by the conventional method.

View full abstract

When we make a sequence by sampling every s digit of an M-sequence, the properties of the obtained sequence are already known. But the properties of randomly sampled M-sequence are not yet known. This paper describes the properties of the randomly sampled M-sequence.
The random sampling method is as follows. First, successive k-tuple of an M-sequence is generated and then using uniform random number X_i(0≤X_i<1), ([k·X_i]+1)'th digit of the k-tuple is chosen.
Autocorrelation function of the original M-sequence has a sharp peak at delay N (N: period of M-sequence), while ensemble averaged autocorrelation functions of the randomly sampled sequences have small peaks at several delays other than N. Mean values of the autocorrelation functions of the sequences are obtained theoretically as a function of the period of M-sequence and tuple length.
Maximum values of the ensemble averaged crosscorrelation functions between the randomly sampled sequences are also obtained theoretically and it is shown that these values are less than the maximum values of crosscorrelation functions between the original M-sequences.

View full abstract

In determination of kernels of a volterra series model for an unknown nonlinear system from input/output experiments, the amount of computation required is generally excessive. If the input is a white Gaussian signal, however, it is greatly reduced by using G-functionals, which is a set of orthogonal functionals on such input.
The purpose of the present paper is an extension of the theory of G-functionals, and it is shown that, a set of orthogonal functionals on non-white Gaussian signals can be derived by applying Karhunen-Loève expansion theorem, and by use of such an orthogonal set of functionals for a basis of functional series model of a nonlinear system, the same amount of computational reduction as for a white Gaussian input can be obtained for a non-white Gaussian input.

View full abstract

This paper is concerned with the design problem of the high-gain regulator system by using the singular perturbation approach.
The synthesis procedure is summarized as follows; at first the controlled system is changed into decoupled systems, and divided into several subsystems which have simple structures. For such systems multiple time scale singular perturbation method is applied and then we obtain multiple parameter high-gain feed-back systems.

View full abstract

When the plant transfer matrix is given by T(s), the interactor L(s) is the polynomial matrix which makes L(s)T(s) a full rank constant matrix as s tends to infinity. The interactor matrix has an important role for the design of the exact model matching control and the adaptive control for the linear multivariable systems. It was first defined by Wolovich and Falb as a lower left triangular polynomial matrix whose diagonal elements are monomials.
In this paper, the structure and the derivation algorithm of the generalized lower left interactor matrix are discussed in detail using the new concepts of d_i-column properness and d_i-improperness index of a polynomial matrix. The algebraic relation of the coefficient matrices between the generalized interactor and the plant transfer matrix is obtained. Based on this relation, it is shown that the diagonal elements of the generalized interactor can be chosen as arbitrary polynomials of adequate degrees which are given by d_i-improperness indices calculated in each stage of the derivation algorithm, and that once the diagonal elements are chosen, the off diagonal elements (lower left part) are determined uniquely except the constant terms. This generalized interactor contains the interactor matrix proposed by Wolovich and Falb as a special case.

View full abstract

Since the numerator polynomial of the plant is adaptively canceled in model reference adaptive control, it can be applied only to minimum phase plants. If model reference adaptive control would be unjustly applied to a nonminimum phase plant, the closed loop system will always be led into instability. While nonminimum phase plants are found more often in discrete time systems, the plant should be regarded as nonminimum phase from the general point of view. Unless the plant is certainly of minimum plase, in order to obtain a closed loop system with satisfactory dynamics, we need adaptive pole assignment where only the denominator polynomial is changed to a desired stable polynomial, leaving the numerator unchanged. There have been many research works on adaptive pole assignment, but most of them are concerned with indirect method. The unallowable defect of indirect method is that stability is not guaranteed unless the outer reference input contains sufficiently many frequency components. Elliott presented a direct method for constructing adaptive pole assignment systems, but he did not demonstrate the motivation of introducing Bezout identity. Moreover, stability was not discussed at all. This paper is also concerned with direct method. Not only the motivation of introducing Bezout identity is clearly demonstrated, but stability is also discussed in detail. The important result is that stability is guranteed even if the outer reference input is not sufficiently rich. Of course, otherwise pole assignment is adaptively achieved. The key concept in the discussionof stability is comparison of orders of two divergent time functions.

View full abstract

In a design of a multi-input multi-output (MIMO) adaptive control system, an interactor plays an important role. The derivation of the interactor requires the system parameter values unless the interactor is in diagonal form. This is a considerable restriction of the application of the interactor to the adaptive control system.
In this paper an alternative approach is presented, where a diagonal pre-compensator is designed and almost of all compensated plants are able to have diagonal interactors. Since the design of the pre-compensator needs only relative degrees of the elements of the transfer matrix and does not need the parameter values, this approach is more applicable to the adaptive control system design.
For the design of the compensator, the structure of the zero position of a matrix A is investigated, where A is the high frequency gain matrix of a MIMO plant. And the characteristics of the structure are given. Using the zero structure of A, an algorithm to obtain the compensator is given. A design of adaptive control system with the compensator is shown.

View full abstract

The adaptive Robbins-Monro stochastic approximation procedure is extended. Parameter convergence rate of the stochastic approximation procedure crucially depends on the gain coefficient which determines the increment of the parameter at each step. The adaptive Robbins-Monro procedure contains an algorithm to estimate the optimal gain. In this paper, an extension of this algorithm for the gain is proposed. We clarify the conditions which guarantee the convergence of the gain to the optimal one. Under these conditions, it becomes possible to improve the parameter convergence rate of the stochastic approximation procedure. An illustrative example is shown.

View full abstract

Collision avoidance problem of a robot manipulator has been studied quite extensivey by many researchers. In this paper, a new approach for this problem is proposed:
“Some of the collision avoidance problem can be formulated as a multiobjective programming problem which has the following two kinds of the objective functions (A) & (B).”
(A) The objective function which forces the end effector of the manipulator trace the ideal path.
(B) The objective function which keeps each link of the manipulator at a distance from obstacles in the workspace in order to avoid collisions.
The satisficing trade-off method is utilized in order to solve the multiobjective problem with the two kinds of objective functions (A) & (B). Several computer simulation results confirm our new approach.

View full abstract

This paper proposes the way which effectively searches the near minimum-time joint trajectories by iterative improvement. The proposed method consists of two procedures: the first one iteratively changes the joint trajectories described by B-spline and the second one minimizes the travelling time of them using the idea of the time scaling. The joint trajectories are iteratively and locally changeable based on the local controllability of B-spline, so that the second procedure is applied only to the range of the changed trajectory. For this reason, the proposed method becomes an efficient algorithm as a whole. It is also capable of treating the trajectories which have any initial and terminal conditions and pass specified points. Examples of planned trajectories of the manipulator with two links and two degrees of freedom are shown and the validity of the obtained trajectories is discussed.

View full abstract

In this paper, the manipulating force and grasping force are defined explicitly, and the relation between fingertip force and these two forces for articulated multi-fingered hands is made clear. Firstly the definitions of manipulating and grasping forces for two fingered hands are given such that they agree with our intuitive image of these forces. Then a new representation of the internal force among the fingers is given for three fingered hands. Based on this representation, the grasping force is defined as a fingertip force which produces no resultant force and satisfies the static friction constraint. The grasp mode is also introduced. The manipulating force is then defined as a fingertip force which produces the specified resultant force, is not in the inverse direction of the grasping force, and does not contain any grasping force component. Finally an algorithm for decomposing a given fingertip force into manipulating and grasping forces is presented, and validity of this decomposition are shown by a numerical example.

View full abstract

This paper describes a calculation method of reachability matrix considering three types of information between components, that is 1) relate, 2) not relate, and 3) unknown. A renewal method of reachability matrix when a relation is modified is also presented. The purpose of these method is to support an incremental development of the structural model which considers the types of information, such as the intensity of relation or logical relation.
The max-min operation is proposed to calculate a reachability matrix including unknown relations. This operation is also employed, when other types of information is considered. So, the same process is used to calculate a reachability matrix. From this matrix, we propose six forms of practical question to judge unknown relations. These questions are based on the number of unknown entires in the reachability matrix. And they are arranged to judge easily using a hierarchical relation which has already known.
The renewal method can reduce a calculating time of reachablity matrix, when a relation is modified. By this method, the renewal process can be employed to confirm the effect of modification easily and visually on a structural model.

View full abstract

Cyclotomic phase is a useful concept in studying the phase properties of an M-sequence. This paper shows that the resemblance of the partial properties of an M-sequence at cyclotomic phases is clearly explained by use of sampling properties of M-sequence.

View full abstract

In this short paper, an FTSC (Finite Time Settling Control) algorithm with an L-th order softening filter is proposed for multiple-input multiple-output processes. This algorithm is very simple, but it is practical and effective in improving the responses of the FTSC systems.

View full abstract

Precise dynamic models are necessary for various advanced control of manipulators. Several identification methods have been studied so far, but there has not been any quantitative evaluation method of identified parameter values or identification tests. In this paper, a criterion for evaluation of the identification tests is proposed. According to this criterion, it is one of the desirable identification tests to take data for manipulator motions which are similar to those used during the actual application of the manipulator.

View full abstract