All issues
Volume 30, Issue 2
Displaying 1-18 of 18 articles from this issue
- |<
- <
- 1
- >
- >|
-
Kazuyoshi KUBO, Nobuharu AOSHIMA1994 Volume 30 Issue 2 Pages 127-135
Published: February 28, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA function, by which reverberation decay curves are calculated from an impulse response of certain system, is re-defined using intuitive concept. Class and its formula of generalized reverberation decay function are derived. The re-definition is performed by 1) definition of instantaneous energy spectra by window operation in time-frequency domain, and 2) definition of time domain marginal function of the spectra as a reverberation function.
It is shown that four known functions of reverberation belong to the generalized class. And also it is shown that the Schroeder method with band passed impulse responses is the sum of the reverberation function and complementary function, namely attack curve function.
Transferring the characteristics of instantaneous power spectra to the class of reverberation by proposed re-definition, the characteristics and differences of known reverberation functions are revealed. Possibility of existence of unkown reverberation function is also described.View full abstractDownload PDF (1788K) -
Shigeki TSUCHITANI, Seiko SUZUKI, Satoshi SHIMADA, Masayuki MIKI, Masa ...1994 Volume 30 Issue 2 Pages 136-142
Published: February 28, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn microstructures such as microsensors and microactuators, a small surface force can occur between two contacting surfaces which affects operations. Therefore, it is very important to clarify the generation mechanism of this surface force and to find a way to decrease it in the devices.
In this study, a microstructure used in evaluating surface force was fabricated by a silicon micromachining technique. This device has a movable plate supported by cantilever beams and two opposing electrodes positioned above and below the movable plate. The surface force acting at the contact area of the movable plate and the opposing electrodes was evaluated as follows: The movable plate made contact with one of the electrodes when an electrostatic force was applied. Then another electrostatic force was applied to remove the movable plate by applying a voltage between it and the other electrode. The surface force was calculated from the voltage needed to separate them. This method was applied to measure the surface force under various conditions in which the amount of water adsorbed on the surface was different.
It was confirmed that the surface force could be lowered by such measures as thermal treatment at about 200°C and hydrophobic treatment of the surface, which decreased the amount of water adsorbed on the surface from the atmosphere. This meant that interaction between the water molecules adsorbed on the surface was one of the main causes of the surface force.
Another way to lower surface force was to make the contact area of the movable plate and the electrodes samller. For example, pyramid type stoppers, formed on the movable plate, were very effective to decrease the surface force.View full abstractDownload PDF (1825K) -
Shigeki TSUCHITANI, Seiko SUZUKI, Satoshi SHIMADA, Masayuki MIKI, Masa ...1994 Volume 30 Issue 2 Pages 143-149
Published: February 28, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn microsensors or microactuators which use an electrostatic capacitance change or electrostatic force as a driving force, a structure with an insulation layer on a movable electrode or its opposing electrode is often used in order to prevent short circuiting of the electrodes when they contact each other.
Model samples for evaluating surface force in such micro devices were fabricated by a silicon micromachining technique. Each sample was composed of a movable electrode supported by cantilever beam and opposing electrodes. Insulation films of thermal oxide were formed on the surface of the movable silicon electrode. Surface force between the insulation film and the opposing electrode which arise after applying voltages between the electrodes was evaluated.
The surface force depended on magnitude, applied time and especially polarity of the applied voltage. Such a surface force is considered to originate in the electrostatic force due to the electric charge stored in the insulation film or the residual polarization of the insulation film arising from the large applied electric field.
The surface force of the model sample was significantly decreased by forming shield electrodes, which have the same electric potential as the movable electrode, at opposed portions of the opposing electrode to the insulation film. In such a structure, the electric field applied to the insulation film was much reduced compared with that without the shield electrode. This structure can be applied to various kind of microsensors or microactuators to reduce the surface force which lowers performance of such devices.View full abstractDownload PDF (940K) -
Yasumichi HASEGAWA, Tsuyoshi MATSUO1994 Volume 30 Issue 2 Pages 150-157
Published: February 28, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper the discrete-time system is treated. And it is discussed the realization problem of almost-linear systems to be the special sub-class of pseudo-linear systems. Almost-linear systems are the pseudo-linear systems whose state transition equation are represented as the sum of free motion and linear action by input, where the one of pseudo-linear systems are represented as sum of free-motion and non-linear action by input.
The established realization theorem of pseudo-linear systems being based on, the realization problem of almost-linear systems is discussed.
Generally speaking, the realization problem is stated as follows. Let I/O be the set of input/output map to be a black-box and CD be the category of dynamical systems which have the same behavior as the given black-box, it is intended to obtain the following realization theorem. Realization theorem: ⌈For any given input/output map α∈I/O, there exist at least one dynamical systems σ∈CD which have the behavior α. Let any σ1, σ2∈CD have the same behavior, then σ1 is isomorphic to σ2 in sense of category CD.⌋
In this paper, let I/O be the set of any input/output map with causality, time-invarince and affinity (this map is called a time-invariant, affine input-response map) and let CD be the category of canonical almost-linear systems, then the realization theorem of almost-linear systems is obtained.
Modified impulse response being introduced, it is shown that any time-invariant, affine input-response map can be characterized by it. “So-called” linear systems can be given as specific examples of almost-linear systems. Where “so-called” linear systems mean linear systems with a non-zero initial state.View full abstractDownload PDF (1544K) -
Yoshihiko MIYASATO1994 Volume 30 Issue 2 Pages 158-164
Published: February 28, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn most of the studies of model reference adaptive control, it is assumed that an upper bound on the degree of the controlled system is known. It makes the scope of application of model reference adaptive control too restrictive, since the reasonable upper bound on the degree cannot be specified a priori in many practical cases.
In the present paper, we propose a design method of model reference adaptive control systems for nonlinear systems with unknown degrees and with unknown relative degrees not exceeding 2. Even if the degree of the controlled system varies, and even if the relative degree also varies between 1 and 2, the structure of the proposed adaptive controller does not change. It is shown that the resulting control system is uniformly bounded, and that the tracking error converges to an arbitrarily small residual region. Finally, several simulation studies also show the effectiveness of the proposed method.View full abstractDownload PDF (1019K) -
Eiichi MIYAZAKI, Takeaki YOSHIMURA, Takumi MINEMOTO1994 Volume 30 Issue 2 Pages 165-171
Published: February 28, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe A/D-transformed signal of the integrated speckle intensity detected by a TV camera yields a quantization error, and the 2nd order moment of the signal deviates from that of the integrated intensity. In this paper we have estimated the conditions to minimize the deviation. The deviation depends strongly on the dynamic range L and the quantizing number q of the A/D converter to be used. If the condition 5<L<7 is selected for q=8 bit, it has been found that the 2nd order moment can be determined within an accuracy of 3%. This result admits the image processing of the speckle pattern with enough accuracy, and it has been shown that the surface roughness of an object can be determined from the accurately measured 2nd order moment.View full abstractDownload PDF (1505K) -
Yukio HASHIMOTO, Takeshi TSUCHIYA, Toshihiko MATSUDA, Ichiro SUGIOKA1994 Volume 30 Issue 2 Pages 172-180
Published: February 28, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper considers a control of travelling cranes which dissipates the physical energy of the oscillating load and drive the trolley to a desired position. The dynamics of the system is modelled by using Hamilton's-equations to describe the energy change of the system and to derive an energy dissipative feedback control law. The obtained control needs neither feedback of the swing angle of the load nor that of its time derivative. For the transportations of the load with heavy weight, the control shows high performance even under large changes of the rope length without using any information of the load-swing. On the other hand, the oscillation suppression effect degrades for the case of light load transportations. To avoid the degradation, we propose an improved partial state feedback control to follow the state trajectories of the closed loop system with heavy load. The validity of these two controls were verified by experiments on a model crane.View full abstractDownload PDF (1802K) -
Hiroshi NAKAMURA1994 Volume 30 Issue 2 Pages 181-189
Published: February 28, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn our previous paper, a discrete-time preview-repetitive control method was presented. The experimental results on motor position control systems demonstrated the applicability of the method. In this paper, a more generalized discrete-time preview-repetitive control method is presented. It is shown that the control systems include phase lead compensations that depend on the prediction horizons. The sufficient conditions for the stability of the systems are derived and the relations between the design parameters and the stability are studied. For motor position control systems, the results of simulations and experiments show that the robust stability against the modeling errors due to the inertial load's changes can be improved by increasing the prediction horizon, which is the dominant parameter.View full abstractDownload PDF (1106K) -
Hajime KANADA1994 Volume 30 Issue 2 Pages 190-197
Published: February 28, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe number of sensors increases in instrumentation systems. And the advancements of computers make possible the collection of large quantities of sensor data. This paper considers the computational complexity of analyzing the sensor data in a distributed instrumentation network. The network is composed of many sensors, preprocessors and processors. The processors are connected by a local area network containing token rings and buses. The lower and upper bounds for some distributed algolithms of data analysis are derived for the network where each of loops makes one ring. And the time complexity and communication complexity required for processing all the data of the processors are given.View full abstractDownload PDF (1325K) -
Hiromitsu KURISU, Takushi NISHIYA, Nihei TACHI, Hiromu ADACHI1994 Volume 30 Issue 2 Pages 198-207
Published: February 28, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper describes a two-step method for a dynamic water distribution considering storage at reserviors. The dynamic distribution is reduced to a minimum cost flow problem by an extended network model. At first a minimum cost flow which satisfies preservation conditions at each node and limit conditions at each arc is found very rapidly with a kind of split simplex method. This is because the method, which we call multi stage primal method, described by only integral operations. Next, the flow is modified along the loops which have zero cost and the minimum cost does not increase. This heuristic method make the flow satisfy conditions for convenient operation of pumps and vulves.
Waterworks bureaus in large cities want a good method for dynamic water distribution to make effective use of limited water resource. However, the scale of the problem is too large to solve under limited machine emvironment. Moreover, it is very difficult to get the solution that satisfies conditions for the operation of pumps and vulves. The proposed method can get over such trouble. Wide-area water management system applied this method has already realized in a certain large city and its merit has been verified.View full abstractDownload PDF (2336K) -
Kajiro WATANABE, Kazumasa SERIZAWA1994 Volume 30 Issue 2 Pages 208-215
Published: February 28, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSPeople sometimes recognize a system which is of a high grade by both the feel and/or sound of it's keyboard switches when pushed. A keyboard whose switches when pressed, have a feeling that is neither too nor too light, and which also convey a clear confirmation of switching makes a computer.
We focus on the man-machine interface of the keyboard and investigate the human feelings of touch and push as the engineering matter in question in order to design switches comfortable to operate.
First, the categories (words) that can quantitatively characterize the switch pushing feeling were selected. Next, the sensory tests by questionnaire using these categories were carried out to acquire sensory data that relate both the feeling and the switch's physical characteristic. Finally, the artificial neural networks to store the sensory data were trained. The pattern of the touching feeling degrees expressed by the categories is used as the input data for training the patterns given by physical characteristics of switches are used as the output data.
The trained artificial neural networks have the potential to select the most preferable switch and provides the physical quantities necessary to design, just by grading each item in the switch selection sheet very similar to that used in the questionnaire.View full abstractDownload PDF (1405K) -
Akira HIRAIWA, Noriyoshi UCHIDA, Katsunori SHIMOHARA, Noboru SONEHARA1994 Volume 30 Issue 2 Pages 216-224
Published: February 28, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe cybernetic interface through which users can communicate with computers “as we may think” is the dream of human-computer interactions. Aiming at interfaces where machines adapt themselves to users' intention instead of users' adaptation to machines, we have been applying a neural network to realize electromyographic (EMG)-controlled slave hand. This paper proposes that EMG patterns can be analyzed and classified by a neural network. Through experiments and simulations, it is shown that recognition of finger movement and joint angles in dynamic finger movement can be successfully accomplished.
A 3-layred back-propagation network is used for finger action recognition from 1 or 2ch surface EMG. In the case of static fingers' motions recognition, 5 categories were classified by the neural network and the recognition rate was 86%. In the case of joint angles estimation in continuous finger motion, the root mean square error was under 25 degrees for 5 fingers 10 joints angles' estimations.
Cyber Finger with 5 fingers 10 joint angles was realized to be controlled by 2ch surface EMG. The slave hand was controlled smoothly and voluntarily by a neural network.View full abstractDownload PDF (3086K) -
Yukihiko IWATA, Haruo SAITO1994 Volume 30 Issue 2 Pages 225-227
Published: February 28, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, the tetrapolar circuit method using a cross electromagnet is proposed. Experiments show that using the proposed method, we can detect where a impedance change occured in cross divided four pieces and impedance change in an one piece sensitively.View full abstractDownload PDF (387K) -
Ikurou MIZUMOTO, Zenta IWAI, Hirofumi OHTSUKA1994 Volume 30 Issue 2 Pages 228-230
Published: February 28, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper deals with the problem of improving the robustness of simple adaptive control (SAC) method. It is derived that the addition of robust adaptive control term to SAC algolithm significantly improves the tracking performance of SAC system.View full abstractDownload PDF (309K) -
Masaru KOBAYASHI, Yuzo OHTA, Hiromasa HANEDA1994 Volume 30 Issue 2 Pages 231-233
Published: February 28, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper proposes applying Polygon Interval Arithmetic to design the loop gain via robust root locus method. By dealing s as a parameter belonging to the boundary of Γ-region, we can compute the value set of the characteristic polynomial effectively.View full abstractDownload PDF (414K) -
Hiroshi TSUJI, Shouyu WANG, Takeshi TSUCHIYA1994 Volume 30 Issue 2 Pages 234-236
Published: February 28, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, we propose a new attractive digital acceleration control method for robot manipulator based on the instantaneous relation between the movement and the acceleration.View full abstractDownload PDF (327K) -
Ken SHIMONAKA, Tetsuya MANABE, Takayuki YAMADA, Tetsuro YABUTA1994 Volume 30 Issue 2 Pages 237-239
Published: February 28, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper presents comparison between a frequency domain learning control algorithm and a time domain learning control algorithm using a force control servo system with one degree of freedom. Results show that the frequency domain learning control guarantee a uniform convergence.View full abstractDownload PDF (452K) -
Koichiro DEGUCHI, Kyoko SHINOZAWA, Iwao MORISHITA1994 Volume 30 Issue 2 Pages 240-242
Published: February 28, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA use of pattern spectrums to evaluate the optimality of the structuring elements for morphological shape decompositions is proposed. Their “average roughness” is shown to be a good measure for the selection for the simplest decomposition.View full abstractDownload PDF (364K)
- |<
- <
- 1
- >
- >|