All issues
Volume 36, Issue 7
Displaying 1-11 of 11 articles from this issue
- |<
- <
- 1
- >
- >|
-
Kenji KAWASHIMA, Tatsuya FUNAKI, Toshinori FUJITA, Toshiharu KAGAWA, K ...2000 Volume 36 Issue 7 Pages 549-554
Published: July 31, 2000
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSRecently, a new mass flow meter for gases becomes commercially available which installed a newly developed micro flow sensor on the velocity boundary layer in the pipeline. As the response of the sensor is high, this sensor is expected to use also on the unsteady flow measurement. Although, the velocity profile of the unsteady flow is different from that of the steady flow, it is not sure that the unsteady flow can measure correctly.
In this paper, the sinusoidal oscillatory flow is measured using the micro flow sensor from a velocity at a point in the pipeline with the calibrated results at the steady flow. The effect of the position of the sensor to the measurement is investigated on the laminar flow both theoretically and experimentally. Standard flow rates are generated by an unsteady flow generator which are proposed by us.
It becomes clear that when the oscillatory flow rate is measured by the flow sensor on the proper position on the velocity boundary layer in the inlet region, the sinusoidal flow up to 10Hz could be measured using the calibrated result on the steady flow.View full abstractDownload PDF (2135K) -
Xin XIN, Taek-Kun NAM, Tsutomu MITA2000 Volume 36 Issue 7 Pages 555-562
Published: July 31, 2000
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe construction of an interactor canceling the infinite zeros of a non-square proper transfer matrix is discussed in this paper. Some new properties of the infinite eigenstructure of the system matrix pencil of the transfer function are presented by investigating the numerically stable algorithm in Dooren (1979). Then, a direct and numerically stable state-space design of the interactor is proposed. Unlike Copeland and Safonov (1992), this paper does not use the relation between the infinite zero structure of the transfer function in the frequency domain and the infinite eigenstructure of the system matrix pencil of the transfer function for the derivation of the interactor.View full abstractDownload PDF (1577K) -
Naoto ABE, Hiroyuki ICHIHARA2000 Volume 36 Issue 7 Pages 563-568
Published: July 31, 2000
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper deals with an iterative approach of closed-loop identification and IMC (internal model control) design for input time delay systems. IMC is one of simple design methods for systems with time delay, however, IMC requires system parameters exactly. In this paper, a joint design method of closed-loop identification and IMC structure for input time delay systems is proposed. The control purpose, which is similar to the windsurfer approach, is to achieve the increase in the bandwidth of the closed-loop system. A multiplicative uncertainty of input time delay systems becomes smaller, when an IMC filter time constant is chosen smaller. A simulation example is shown to illustrate these results.View full abstractDownload PDF (2065K) -
Toshiyuki OHTSUKA2000 Volume 36 Issue 7 Pages 569-574
Published: July 31, 2000
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSState equations that are at most quadratic with respect to the states are considered as a universal model structure for various classes of nonlinear systems. A necessary and sufficient condition is shown for existence of such quadratic-in-the-state representation for a given nonlinear system. Furthermore, the constructive proof of that necessary and sufficient condition gives an algorithm to obtain the quadratic-in-the-state representation. Illustrative examples demonstrate the algorithm to construct the quadratic-in-the-state representation. The quadratic-in-the-state representation is expected to be useful in identification and control system design.View full abstractDownload PDF (1544K) -
Shigenori SANO, Kazuya OGATA, Yoshikazu HAYAKAWA2000 Volume 36 Issue 7 Pages 575-582
Published: July 31, 2000
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSFor the control design, it is important to build a good model. To identify the model parameter exactly, it is necessary to use data which are less affected by the disturbance or modelling uncertainty. We propose that it is important to evaluate the upper bound of the parameter estimation error when the disturbance and the unstructured model uncertainty exist. It can be used to decide the optimal identification input and frequency weight for least square criterion.View full abstractDownload PDF (1395K) -
Satoru GOTO, Masatoshi NAKAMURA, Shinji YAMANAKA, Nobuhiro KYURA2000 Volume 36 Issue 7 Pages 583-588
Published: July 31, 2000
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSAccurate contour control performance is required for industrial robot arms such as sealing process, laser cutting and so on. This paper describes contour control of industrial articulated robot arms by introducing master-slave synchronous positioning based on nonlinear separation model. The proposed method consists of nonlinear statics compensation of the inverse kinematics and kinematics, and master-slave synchronous positioning in the joint coordinates. The effectiveness of the proposed method was assured by simulation results and experimental results using an actual industrial robot arm.View full abstractDownload PDF (1043K) -
Baiquan LU, Junichi MURATA, Kotaro HIRASAWA2000 Volume 36 Issue 7 Pages 589-598
Published: July 31, 2000
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA method is proposed for designing three layer neural networks that gives relevant network structures which assure global minimization of learning errors for small training sets and small learning errors for big training sets both irrespective of the initial values. A condition on network structure is considered to achieve the above purpose, and a number of possible structures are provided together with their learning algorithms. Also, the generalization abilities of the network structures are discussed to guide the choice of structures in practice. All of the proposed structures for small training sets have zero errors after learning by a gradient-based algorithm and thus solve the local minima problem. The difference between them is in the level of locality and generalization abilities. For a big training set, first, the structure with zero learning errors for part of training data is obtained, then all of training data are used to train the network of the given structure, which improves the generalization abilities. Numerical examples are provided that support the present approach.View full abstractDownload PDF (1871K) -
Hisayoshi SAKURABA, Rieko OSU, Eri NAKANO, Yasuhiro WADA, Mitsuo KAWAT ...2000 Volume 36 Issue 7 Pages 599-608
Published: July 31, 2000
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn multi-joint planar reaching movement, hand trajectories tend to be gently curved depending on the hand position in extracorporal space. Following two explanations are possible for the observed curvature; (i) planned trajectories themselves are curved, (ii) planned trajectories are straight but realized trajectories are curved because of the reasons other than planning. We examined one of the possible reasons that belongs to the second explanation. That is, the realized trajectories are curved because the internal model in the central nervous system is imperfect. We represented imperfect internal model by (1) parametoric method, (2) neural metwork method (3) table look up method, and compared the trajectory predicted by imperfect internal model with the measured trajectories. The results suggest that the imperfect internal model cannot explain the observed curvature better than the minimum commanded torque change model that plans curved trajectories.View full abstractDownload PDF (2414K) -
Yasufumi NAGAI, Shigeru TSUCHIYA, Takashi IIDA, Shinichi KIMURA2000 Volume 36 Issue 7 Pages 609-618
Published: July 31, 2000
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSWhen we operate space robots, safety and reliability are the most important factors, even at the expense of dexterity and swiftness. Therefore it is very stressful for operators to teleoperate space robots. Especially since space robots-unlike ground robots-can not be repaired during the operation. Furthermore, operators have to quickly analyze a large amount of information about the targets, and immediately make decisions about how to proceed with the operation. The majority of information is presented by visual cues, such as digital values, status display, 3-D computer simulation, and camera images. Therefore, we developed an audio feedback system that can convey some of this information in order to improve the operator's decision making and avoid miss-recognition and miss-operation accidents. We used the Eye-Mark-Recorder (EMR) to assess the performance of this system. Until now audio systems for information presentation have only been proposed for ground-based operation systems. This is the first case where sound is utilized for space robotic operation even though sound does not actually exist in space.View full abstractDownload PDF (2803K) -
Toru KUMAGAI, Akio UTSUGI2000 Volume 36 Issue 7 Pages 619-621
Published: July 31, 2000
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSWe proposed to classify magnetoencephalogram (MEG) data by mixture of factor analyzers (MFA) for excluding artifacts. We showed the effectiveness of proposed method by applying it to an analysis of visually evoked MEG data.View full abstractDownload PDF (1922K) -
Issei SHINOHARA, Syunichirou KIKUCHI, Yasuaki OISHI, Hidenori KIMURA2000 Volume 36 Issue 7 Pages 622-624
Published: July 31, 2000
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper provides an algorithm to compute the admissible parameter region in the learning by a single-layered perceptron. This algorithm is carried out by solving linear programming problems, the number of which is proportional to that of given samples.View full abstractDownload PDF (467K)
- |<
- <
- 1
- >
- >|