All issues
Volume 24, Issue 3
Displaying 1-17 of 17 articles from this issue
- |<
- <
- 1
- >
- >|
-
Yasutaka TAMURA, Hitoshi ABE, Kazuyoshi ENDO, Takao AKATSUKA1988 Volume 24 Issue 3 Pages 213-219
Published: March 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA new digital beam-forming technique is proposed for the reflective-mode ultrasound imaging. This technique is based on the holographic beam-forming using uncorrelated transmitting pulses, and it permits high speed data acquisition with simple hardware.
In the proposed method, a carrier signal is modulated with uncorrelated pulse sequences, and ultrasound pulses emitted from transmitters construct a wavefront of non-directional power propagation. Beam-forming on the transmitting and receiving stages can be accomplished numerically using delay-and-sum operation and correlation.
Point spread function (P.S.F.) of the imaging system is derived as a function of the range and azimuth differences. From the analysis of the P.S.F., it is shown that both the range and azimuth resolution limit is sufficiently small when the waveforms are uncorrelated to each other and the sum of their auto-correlation functions has a single sharp peak, even if the band-width of each waveform is narrow.
We consider the utilization of Walsh functions as the modulating sequences. An experimental system utilizing a 50kHz airborne sensor array is constructed. This system has a linear array which consists of 16 transmitters and a single receiver. The experimental results indicate the feasibility of the method. Computer simulation shows that the additional receivers improve the P.S.F. of the system.View full abstractDownload PDF (2117K) -
Yoshikazu IIKURA, Yoshifumi YASUOKA1988 Volume 24 Issue 3 Pages 220-227
Published: March 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSAmong automatic supervised classification methods, the best linear discrimination (BLD) is superior in the theoretical sense because it can utilize statistical features (covariance matrix) extracted from each category. However, as pairwise comparisons were needed, processing speed of BLD became very slow when the number of categories increased.
In this paper, we proposed following two efficient algorithms: (a) adaptive ordering of pairwise comparisons by utilizing classified result of neighbor pixels (BLD/adp), and (b) utilization of the BLD as the decision rule of the binary decision tree (BDT/bld). As a theoretical approximation of processing speed of these algorithms, we showed the number of multiplication required for classifying a pixel, which depended on the number of categories (N) and variables (M). It was shown that the number of multiplication ranges from O (logN×M) to O(N×M) in the case of BDT/bld, while the number was not less than O(N×M) in the case of BLD/adp.
In order to see the actual performance, we applied these algorithms to multispectral data of LANDSAT 5. The conventional methods (maximum likelyhood, Fisher's linear discrimination, and binary decision tree) were also applied to the same data and were compared with proposed methods. It was shown that the performance of BDT/bld was better than that of Fisher's linear discrimination when the number of categories was large.View full abstractDownload PDF (2809K) -
Masatoshi ISHIKAWA1988 Volume 24 Issue 3 Pages 228-235
Published: March 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA parallel processing LSI for the purpose of realization of intelligent sensors and its application to a tactile sensor are described.
The LSI, which is called “SPE-8”, is located as a special purpose processor for sensors in hierarchical sensing and control systems. The parallel processing of the SPE-8 has special features which is different from normal parallel processing. Each processing element has a direct input from a sensor. As the arithmetic and logical unit of the SPE-8 uses bit serial processing, the circuit of processing elements is so compact that eight processing elements including functions of scanning and transmission are integrated in one chip.
A tactile sensor using the SPE-8 is also described. The sensor has 8×8 detective matrix using pressure-conductive rubber and realizes high speed pattern processing and data transmission. The sensor can be called an integrated parallel processing tactile sensor or a high performance intelligent sensor.
In this paper, the design concept, functional features, and the structure of the SPE-8 are shown. In addition, the structure of the tactile sensor using the SPE-8 and basic sensing and processing results of the sensor are shown.View full abstractDownload PDF (2763K) -
Ken-ichi HAYAMIZU, Tohru IDOGAWA1988 Volume 24 Issue 3 Pages 236-242
Published: March 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA novel method for the measurement of velocity in a plane using crosscorrelation functions is presented. Signals are obtained from a random surface of an object by a circular array sensor of 64 photodiodes. Crosscorrelation functions R(m, n) (n=1, 2, …, 32) are computed from 32 pairs of photodiodes; each pair consists of 2 photodiodes placed at the both ends of a diameter of the sensor. Magnitude and direction of a velocity of the object is determined from m=m0 and n=n0 where a maximum of R(m, n) is found.
It took less than 1/25 s to compute the correlation functions R(m, n) by a correlator made for our experiments. The present method was found to be of much practical use by experiments: (1) A rotating disk with random surface was observed by the circular array sensor; the point observed by the sensor, as seen from the disk, traveled along a circle on the disk surface. A travel distance of the point was measured to an accuracy of 0.1%. A center and angular velocity of its circular motion were also determined accurately. (2) Flow velocity distribution of a water surface was measured by displacing the sensor and, for example, a position of a vortex center was measured to an accuracy of 0.4mm.View full abstractDownload PDF (1749K) -
Eiji KONDO, Takanori KIYOTA, Teruo SUNAGA1988 Volume 24 Issue 3 Pages 243-248
Published: March 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper discusses a canonical form of a deadbeat state observer for a discrete-time, linear, time-invariant system. A general form of the deadbeat state observer which shows the freedom of design has been obtained; however the design parameters have redundancy. To avoid the redundancy, a canonical form which is expressed by the linear combination of independent design parameters is derived for some subsets of the general form. Finally, illustrative examples are given to show the effectiveness of the proposed procedure.View full abstractDownload PDF (780K) -
Michio NAKANO, Toshio OGUCHI, Takashi HIROMATSU, Takashi KOGA1988 Volume 24 Issue 3 Pages 249-254
Published: March 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn a positioning servo-system, the ideal response can be obtained through Posicast control or Bang-bang velocity feed-back control. But the perfection of that responce depend entirely upon stability of the magnitude of the input signal.
Then, in this work, a new control algorithm (we call hysteresis control) is proposed to eliminate the influence of the input signal.
In this algorithm, a velocity feedback through a hysteresis element is introduced in a servo system modeled by a “second order system” to accomplish the Posicast control principle without influence of the input signal.
Several simulations were carried out to evaluate the effectiveness of the hysteresis control in real positioning servo-systems, and the following points were noted:
-For servo-systems with higher order-lag system which can be approximated by a “second order-lag system with deadtime”, it is also possible to accomplish the Posicast control principle without influence of the input signal (finite time settling response without oscillations is obtained).
-Even in systems with saturation features, it is possible to accomplish the Posicast control principle without influence of the input signal.
Finally, some experiments were done in a servo-system control for DC servo-motor to confirm that the proposed method generates a simple control structure with good response features.View full abstractDownload PDF (792K) -
Hideo ISHIHARA, Seiji KOBAYASHI1988 Volume 24 Issue 3 Pages 255-262
Published: March 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper proposes an adaptive control scheme for unknown nonlinear systems. Approximating plant dynamics around a suitably chosen point (named operating point) by a finite power series model with unknown coefficients, an adaptive algorithm is derived, which contains estimation of parameters, determinations of control inputs, and change of operating points.
The latter is performed when the plant behavior is apart from a current operating point. Since this causes parameter jumps in the model, new parameters must be estimated again, the initial values of which are predicted from the old. These procedures are termed “reset”.
Conditions for the stability of the algorithm is investigated, and effect of “reset” is discussed. Simulation results show the effectivenss of the proposed algorithm.View full abstractDownload PDF (1061K) -
Atsushi FUJIMORI, Hirobumi OHTA1988 Volume 24 Issue 3 Pages 263-269
Published: March 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA synthesis method of linear optimal regulators for multivarible systems is presented using singular values of the loop transfer function evaluated at the input point. To this objective, performance specification and condition of stability robustness are first examined to obtain magnitude constraints of the singular values. The weighting matrices of a quadratic performance index are then tuned iteratively so that the resulting loop shape may satisfy these constratints. This tuning is done by the quasi-Newton method so as to minimize the criterion which consists of a weighted sum of the differences between desired and actual singular values at some discrete frequencies. To avoid numerical difficulties, the weighting matrices are assumed to be diagonal, which leads to a nonsingular matrix in terms of directional derivatives. Two numerical examples are given to illustrate and to substantiate the proposed method. The resulting optimal requlators are shown to improve the characteristics of the controlled system effectively while keeping their inherent stability property.View full abstractDownload PDF (829K) -
Toshiharu SUGIE, Toshiro ONO1988 Volume 24 Issue 3 Pages 270-275
Published: March 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper is concerned with a doubly coprime factorization technique and the class of all stabilizing compensators. First, a doubly coprime factorization technique is proposed in a transfer function approach. We can obtain all stabilizing compensators directly from one stabilizing compensator via this technique. Second, it is shown that the well-known doubly coprime factorization technique developed by Nett and coworkers can be derived naturally from our method. In addition, a relation between observer-based compensators and the Bezout identity is clarified. Based on these results, a few extended forms of previonus results are given. Finally, a structure of all stabilizing compensators is shown.View full abstractDownload PDF (689K) -
Tomoyuki AKASHI, Yohji KOBAYASHI, Masayoshi TANIMURA1988 Volume 24 Issue 3 Pages 276-283
Published: March 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe pattern of molding conditions and state variables, which consists of injection speed, pressure, specific volume and temperature of resin, determines the quality of moldings in an injection molding process. The demand for the pattern control is increasing in the field of the precise moldings. On the other hand, a desired value of the control system is too complicated to select automatically for several moldings to realize the precise pattern control.
The purpose of the study is to design a manmachine system in the injection molding process using a polygon graph, which is transformed from the desired value and the control value. The polygon graph is displayed as an operator can easily set and correct the value. The characteristics of resin and the dynamical model of the control system is also obtained for the transformation from the desired value to the polygon graph. The desired value is set and corrected, using the polygon graph through intuition and experience of an operator, and the method is effective for such a complicated control system as the molding process.View full abstractDownload PDF (1158K) -
Minoru HASHIMOTO, Richard P. PAUL1988 Volume 24 Issue 3 Pages 284-291
Published: March 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSRobot manipulator Joints with only a single state sensor (either position or velocity) are limited in their ability in determining both state variables (position and velocity). Also, if the single sensor fails, the robot becomes uncontrollable. In the present paper, we propose a method which efficiently uses multi-sensor information to obtain a more precise description of joint position and velocity and allows for robust control system free from the affects of spurious sensor observation and impervious to single sensor failure. We employ an encoder and a tachometer for each joint as sensors. A technique is described for combining both uncertain sensor observations with an observation model to provide an optimal estimation of both state variables. The estimation method does not result in more than one sample period of time delay as the estimator does not use a time series analysis. In order to detect a sensor failure, we apply the cross-checking hypothesis test for redundant sensor information. In addition we can detect a control failure by including the command information of the control in the hypothesis test. The algorithm for implementing the estimation and the hypothesis test is presented. Simulation results demonstrate that the techniques works well.View full abstractDownload PDF (930K) -
Toshio TSUJI, Koji ITO, Mitsuo NAGAMACHI, Yoshihiro TAKAHASHI, Yoshihi ...1988 Volume 24 Issue 3 Pages 292-299
Published: March 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe human-prosthesis systems for force control may be regarded as the man-machine control systems, in which the manipulated variables are driven by EMG signals and the display information is given by electrocutaneous stimulations. Progress in prostheses requires a clearer understanding of the human operator's characteristics and developments of the more intimate interface between amputee and prosthesis based on his characteristics.
This paper analyzes the human operator's dynamic characteristics in human-prosthesis systems by a series of simulated experiments. First, it is shown that the operator's phase lag caused by the electrocutaneous stimulation is longer than the one in regular man-machine systems and that the cross-over model does not hold. From such results, a human-prosthesis interface with the Smith controller is proposed. The Smith controller can compensate the operator's dead time and improve system stability. The experimental results demonstrate that the system performance can be improved largely and a operator's load could be reduced.View full abstractDownload PDF (1173K) -
Isao TSUSHIMA, Norihisa KOMODA, Kenichi OKASAKA1988 Volume 24 Issue 3 Pages 300-307
Published: March 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn production system, layout and operation of the process are frequently modified because of rapid changes in consumer demand. Therefore, as flexible material handling equipment, automated guided vehicle (AGV) has been used instead of conventional conveyor. However, AGVs operation software is incomprehensible and hard to maintain. Also, the efficient traffic control has not been realized by the conventional software.
In this paper, AGVs operation problem is decomposed to condition satisfaction subproblem and combinatorial subproblem. Various modification occurrs in operation conditions such as transportation priority of the first subproblem. And operation know-how is also included in the conditions of this subproblem. Therefore, the IF (condition), THEN (conclusion) rule style is adopted to describe the above conditions, because it is well suited to express operation conditions in non-procedural manner and makes it simple to describe and to modify them. On the other hand, the second subproblem is regard as a kind of combinatorial problem. For this subproblem, the efficient and standardized traffic control algorithm is proposed.
Based on the above concept, the condition and data description method and the configuration in the AGVs operation system are presented. Also, the flexibility and efficiency of the proposed method is demonstrated through an example. The operation method has been successfully installed in a real process.View full abstractDownload PDF (1216K) -
Fumio MATSUDA, Youici YAMAKAWA, Masafumi ITOH, Shuzou HATTORI1988 Volume 24 Issue 3 Pages 308-310
Published: March 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper describes a improved discrimination method of the interior and exterior of a closed curve. This method recognizes the interior and exterior using a improved control signs, without drawing the control signs and using the interior and exterior discrimination algorithm.View full abstractDownload PDF (323K) -
Toshimitsu NISHIMURA1988 Volume 24 Issue 3 Pages 311-313
Published: March 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (380K) -
Tsutomu NOHARA, Shigeyuki HOSOE, Masami ITO1988 Volume 24 Issue 3 Pages 314-316
Published: March 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper the problem of obtaining a reduced order model is considered for an interconnected high order system which is composed of several subsystems.
For such systems retaining the internal structure of the original system in a reduced order model is quite important in the engineering view point.
The most desirable solution to the problem would be, of course, to derive an algorithm for constructing a reduced order model under a single criterion, but unfortunately such an approach is not found yet. In stead, here we proceed to approximate each subsystems by reduced order models and combine the results.
In this case there happen the following questions.
1. Does the reduced order model obtained as above preserve the stability property under the assumption that the original system is stable? If not always, under what conditions?
2. How does the reduced order model approximate the input output characteristics of the original system?
This paper answers these questions.View full abstractDownload PDF (352K) -
Hiroshi YAMADA1988 Volume 24 Issue 3 Pages 317-318
Published: March 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSDownload PDF (228K)
- |<
- <
- 1
- >
- >|