All issues
Volume 32, Issue 10
Displaying 1-15 of 15 articles from this issue
- |<
- <
- 1
- >
- >|
-
Hirohisa SAKURAI1996 Volume 32 Issue 10 Pages 1395-1398
Published: October 31, 1996
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe triple point of oxygen is realized by using the adiabatic calorimetric technique with an open cell and four samples including three different sources. The molar impurity concentrations of the samples and the triple point values of ideally pure oxygen are estimated from the relations between the melting temperature and the inverse of fraction of melt by a thermal analytical method. The estimated triple point values of the samples are agreed within the measurement uncertainty of ±0.035mK (one standard deviation level). The values are pointed out to be about 1.25mK lower than the triple point of the sealed cell, NRLM-7801, which is one of the transfer cells of the international intercomparison organized by the Consultative Committee of Thermometry.View full abstractDownload PDF (1596K) -
Kenji IKEDA, Seiichi SHIN1996 Volume 32 Issue 10 Pages 1399-1406
Published: October 31, 1996
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSWe propose a decentralized high gain adaptive control for a stabilization of interconnected subsystems. Backstepping is applied in order to relax the relative degree constraint. The assumptions for the plant are that each subsystem is minimum-phase and its relative degree is known, and that the interconnection satisfies the so-call matching condition. Because a certain failure of a local controller can be modelled by a change of interconnections, we propose a decentralized adaptive control system which is robust for the uncertainty of interconnections. In order to avoid an unnecessary high gain feedback, we adopt the σ-modified adaptive law. Fault tolerance of the proposing system is proved by showing that a certain class of failure does not change the form of a mathematical model of a decentralized system and that the proposing controller stabilizes the system represented by the mathematical model. Finally, we show a numerical result in order to illustrate the fault tolerance of the system.View full abstractDownload PDF (1098K) -
Extension of the Algebraic Riccati EquationYoshiaki KAWAMURA, Wei PENG1996 Volume 32 Issue 10 Pages 1407-1415
Published: October 31, 1996
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSGenerally the algebraic Riccati equation does not have a useful solution if the system has uncontrollable modes (jω modes) on the imaginary axis. It is shown that the necessary and sufficient relation of this case is not given by the standard Riccati equation but by a certain non-symmetric part of the Riccati equation. General solutions of the LQ and H∞ control problems with uncontrollable and/or unobservable jω modes are given by this non-symmetric Riccati equation.View full abstractDownload PDF (1219K) -
Kentaro HIRATA, Yutaka YAMAMOTO, Allen TANNENBAUM, Tohru KATAYAMA1996 Volume 32 Issue 10 Pages 1416-1424
Published: October 31, 1996
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA complete solution for the two-block H∞ problem for stable distributed plants is derived. The solution consists of one Riccati equation and a finite-dimensional determinantal criterion in terms of a Hamiltonian, which is a natural extension of the well-known formula due to Zhou, Khargonekar, Lypchuk and others. An example is given to illustrate the result.View full abstractDownload PDF (1454K) -
Xiaofeng YANG, Tielong SHEN, Katsutoshi TAMURA1996 Volume 32 Issue 10 Pages 1425-1431
Published: October 31, 1996
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper discusses an approximate realization of nonlinear H∞ state feedback controller based on neural networks. A three-layer neural network is constructed and is trained to satisfy Hamilton-Jacobi inequality, then the state feedback H∞ controller is designed using the network solution. It will be shown that the network learning problem for the solution of Hamilton-Jacobi inequality can be formulated as a maximum value function optimization problem. A nondifferentiable optimization technique is used to develop a learning algorithm. The efficiency of the proposed method is demonstrated by a numerical example.View full abstractDownload PDF (962K) -
Toshiharu SUGIE, Michihiro KAWANISHI1996 Volume 32 Issue 10 Pages 1432-1438
Published: October 31, 1996
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper is concerned with design and stability analysis for a class of nonlinear systems which are exactly linearizable via nonlinear state-feedback only. First, a modeling method for linearized systems is given, which reflects the uncertainties of both the physical parameters of the original nonlinear systems and the state-dependent nonlinear terms in a non-conservative way by using linear fractional representation. Second, a sufficient condition for the stability is given for the closed-loop system, which consists of linearized system and linear compensator, in the presence of parameter variations. Third, the effectiveness of the proposed method is evaluated by experiments on tracking control using a three-link DD manipulator.View full abstractDownload PDF (1426K) -
Takashi NOGUCHI, Koichiro DEGUCHI1996 Volume 32 Issue 10 Pages 1439-1446
Published: October 31, 1996
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper describes a new approach to visual servoing. The visual servoing system has a camera mounted on a manipulator, which is included in a control loop of the manipulator as a visual sensor observing the objects. The key problem in visual servoing is a reduction of dimensionality of image data without a loss of sensitivity to change of object appearance. We use eigenspace method based on Karhunen-Loève expansion which reduces the dimensionality and characterizes a given image data set. The proposed method does not require feature extraction, feature correspondences or object models for obtained images. It relates the camera pose and the reduced image data directly, and leads the robot arm mounting the camera to a goal position.
We construct a surface (manifold) which constraints the camera position and the dimension reduced image in their product space. Using tangential properties, we control the camera to a given goal position which is specified as where the goal image is obtained. Experiments using a real robot arm and a TV camera show feasibility of this proposed method.View full abstractDownload PDF (2262K) -
Fuminori IKEMOTO, Hiroshi KAWAKAMI, Tadataka KONISHI1996 Volume 32 Issue 10 Pages 1447-1453
Published: October 31, 1996
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSWe propose a method for supporting conceptual design by using computer. This method generates creative candidates for novel design of physical systems. First, physical causalities are generated by exchanging variable-relations of normal physical causalities which are acquired by analyzing existing objects. New physical causalities are organized into candidates for design solution. Although some of the candidates yielded by proposed system may not be able to materialize, others have the potential for obtaining novel designs. Moreover, the candidates that have some problems to materialize by the present technology can focus designers' intentions on the problems to work out in order to attain a novel design.View full abstractDownload PDF (1583K) -
Yutaka MATSUTANI, Susumu HASHIZUME, Katsuaki ONOGI, Yoshiyuki NISHIMUR ...1996 Volume 32 Issue 10 Pages 1454-1460
Published: October 31, 1996
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper deals with the condition/event net (C/E net) construction problem PT(X), in which the objective is to synthesize a C/E net to exhibit the desired behavior specified as a partial language X using the only events in X. Since the solution of PT(X) is not always unique, we must choose the desirable ones among all solutions. This paper first shows that the set of solutions of PT(X) can be found by solving a system of inequalities with 0-1 variables and 0-1 coefficients. It next presents a method of solving inequalities of this type.View full abstractDownload PDF (1369K) -
Kazuhiro OHKURA, Kanji UEDA1996 Volume 32 Issue 10 Pages 1461-1469
Published: October 31, 1996
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSMost deceptive problems are difficult for genetic algorithms (GAs). This is because a GA changes its search direction based on the building block hypothesis. This paper shows that a GA with neutral mutations, which is proposed for nonstationary optimization problems, is also an effective approach to solving deceptive problems. Since the GA with neutral mutations adopts a redundant string representation, most genetic operations become to be neutral for an objective function. Another novel characteristic point is that the GA has various genetic operations for operons, which are defined as the clusters of genes, in order to bring about adaptive genetic changes, according to the state of the genetic search. This paper modifies the GA with neutral mutations on the following three points: (1) The value lists in the genes have the same fixed lengths. (2) In order to maintain the consistency to the above rule, the tail values in the lists are deleted in the case of applying a duplication or inserted in the case of applying deletion. The inserted values are determined by a predefined rule for avoiding the fixations of genes on specific values. (3) The population is divided into subpopulations only at the stage of selection and reproduction for maintaining the appropriate diversity. The optimization ability of the proposed method is examined with four well-known deceptive problems. It is observed that it overcomes deception by the emergent property of finding deceptive hyperplanes and escaping the population from them by using the large genetic transitions to their complements.View full abstractDownload PDF (2016K) -
Hiroshi HASHIMOTO, Tomio YAMAURA, Minoru HIGASHIGUCHI1996 Volume 32 Issue 10 Pages 1470-1472
Published: October 31, 1996
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper proposes a method to measure the position of the view point of single camera on a straight passage with constraint conditions of the inclination of the boundary lines. Taking account of the conditions, this method obtains the vanishing point and the perspective translated boundary lines from the 2D-image directly without the binalization processing, and shows good results for the vehicle navigation in the case that the disturbances exist on the passage.View full abstractDownload PDF (895K) -
Wataru KASE1996 Volume 32 Issue 10 Pages 1473-1475
Published: October 31, 1996
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe problem discussed in this note is of finding polynomial matrices N and Y satisfying XN+YD=I for a given pair of skew prime polynomial matrices D and X. A new necessary and sufficient condition for this problem is shown. A solution is given using state space parameters.View full abstractDownload PDF (285K) -
Kinya FUJITA1996 Volume 32 Issue 10 Pages 1476-1478
Published: October 31, 1996
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe existence of the self-directed stiffness regulation mechanism was experimentally demonstrated. The amount of the electromyogram induced by a mechanical perturbation was measured at three ankle positions in four normal volunteers. The reflex loop gain was decreased by the ankle dorsiflexion, in contrast to the increase of the joint elasticity. It is concluded that the change of the reflex gain acts as a joint stiffness regulator which compensates the change of the joint stiffness.View full abstractDownload PDF (384K) -
Fuminori KOBAYASHI1996 Volume 32 Issue 10 Pages 1479-1481
Published: October 31, 1996
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSWhen an element in a bond graph becomes derivatively causal, a resister, of 0Ω in series or of ∞Ω in parallel, with the element changes its causality into integral without any other effect. The resulting system can be analyzed by treating these virtual R's first as symbolic and then as numeric.View full abstractDownload PDF (285K) -
Shigeru KUROSU, Motoyuki ADACHI, Kazuyuki KAMIMURA, Hisanori UCHINO1996 Volume 32 Issue 10 Pages 1482-1484
Published: October 31, 1996
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper describes analytical and experimental studies about a gyroscopic force measuring apparatus (simply called “GFMA”) which provides precise direct digital output proportional to the load applied. The principal dynamical characteristics of the GFMA are analyzed. The GFMA is designed on the basis of theoretical results and constructed. To evaluate the performance, a comparison is carried out between the GFMA produced in this study and the original one in Germany under the same environmental situation. Experimental results obtained are promising for practical applications.View full abstractDownload PDF (379K)
- |<
- <
- 1
- >
- >|