All issues
Volume 29, Issue 3
Displaying 1-19 of 19 articles from this issue
- |<
- <
- 1
- >
- >|
-
Susumu IKEDA, Takuma DOI, Keizaburo UCHIKAWA, Yoshiharu KURITA1993 Volume 29 Issue 3 Pages 245-252
Published: March 31, 1993
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe existence of intrinsic vertical effects of spinning gyroscopes was examined systematically from the viewpoint of precise measurement, using three types of gyroscopes with largely different characteristics. The most precise weighing techniques, such as a reference tare system and various countermeasures, were applied. Slight and symmetrial mass changes observed in the experiment were explained as the parasitic effect of the balance due to the torque of the spinning gyroscopes, utilizing the attenuation constant change under atmospheric conditions. Consequently, the intrinsic vertical effects were found to be virtually zero in this experiment, and even if existent, were estimated to be neither greater than +0.7mg nor less than -0.5mg at a 99% confidence level for the angular momentum of 0.936kg·m2/s.View full abstractDownload PDF (2510K) -
Mehmet Arif ADLI, Hideo HANAFUSA1993 Volume 29 Issue 3 Pages 253-262
Published: March 31, 1993
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn closed chain mechanisms, internal forces can be generated and controlled by adding redundant active joints on the links. The internal forces do no work but have effect on the total stiffness of the mechanisms. The stiffness of a closed chain mechanism can be changed by internal forces without changing its inherent stiffness. The effect of the internal forces depends on the magnitude and mode of the internal forces as well as the change in the configuration of the mechanism. In this study, we explain how to achieve the compliance control in certain directions of motion and change the compliance center by properly generating internal forces for the selected configurations of a planar closed chain mechanism with one redundant actuation.View full abstractDownload PDF (1733K) -
Riyanto BAMBANG, Etsujiro SHIMEMURA, Kenko UCHIDA1993 Volume 29 Issue 3 Pages 263-271
Published: March 31, 1993
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSWe consider the problem of synthesizing controllers which minimize a quadratic performance subject to the constraints on H∞ disturbance attenuation as well as on location of the closed-loop poles. This synthesis problem is treated within mixed H2/H∞ design framework by clustering the eigenvalues of the closed-loop system in subregions of the complex plane. To achieve root clustering, a generalized Lyapunov equation is employed. The quadratic (H2) cost is then evaluated in terms of solutions to this generalized Lyapunov and an additional linear matrix equations. H∞-norm constraint is enforced in terms of solution to a Riccati equation which leads to an upperbound on the actual H2 cost. Based on the evaluation of the actual H2 cost, along with the upperbound, an equivalent optimization problem is formulated. Necessary condition for the optimization problem is derived. This condition involves a set of highly coupled equations. Based on descent technique, an iterative algorithm is proposed for solving such coupled equations, and a numerical example is presented.View full abstractDownload PDF (1877K) -
Julio TANOMARU, Sigeru OMATU1993 Volume 29 Issue 3 Pages 272-280
Published: March 31, 1993
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper addresses the question of how to perform on-line training of multilayer neural controllers in an efficient way. At first, a plant emulator and a feedforward controller based on multilayer neural networks are described. Only a little qualitative knowledge about the plant is required. The controller must learn the inverse dynamics of the plant from randomly chosen initial weights. Basic control configurations are briefly presented. New on-line training methods, based on efficient use of memory-stored data and distinction between sampling and learning frequencies are proposed. One method, called direct inverse control error approach, is effective for small adjustments of the neural controller when it is already reasonably trained; another one, dubbed predicted output error approach, directly minimizes the control error and greatly improves convergence of the controller. Simulation results show the effectiveness of the proposed neuromorphic control structures and training methods.View full abstractDownload PDF (1465K) -
Goutam CHAKRABORTY, Jun-ichi KUDOH, Norio SHIRATORI, Shoichi NOGUCHI1993 Volume 29 Issue 3 Pages 281-287
Published: March 31, 1993
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe advent of space-technology and multispectral sensors on satellites made available enormous amount of remote sensing data, the supervised classification of which is rather impossible, and is equally unrealistic to obtain a suitable training set. Application of Unsupervised Neural Network (NN) in this field is till now unexplored. In this paper we made possible the use of Unsupervised Neural Network algorithm for the analysis of NOAA satellite multispectral data. We have pointed out the difficulties for using unsupervised NN and deviced our method to overcome that by introducing proper preprocessing of data and adding an extra dimension to the input data and using special techniques for NN learning. It is elaborated with simulation results that without using the special technique we introduced, the NN fails to perform any worthwhile classification. The correctness of classification is confirmed by quantitative analysis of our simulation result.View full abstractDownload PDF (2576K) -
Shigeki TSUCHITANI, Seiko SUZUKI, Ben KLOECK, Masayuki MIKI, Masahiro ...1993 Volume 29 Issue 3 Pages 288-294
Published: March 31, 1993
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe mechanical stability of the seismic mass (movable electrode) in a PWM electrostatic servo accelerometer which was fabricated by using a silicon micro-machining technique was studied theoretically and experimentally.
The movable electrode is controlled so as to be parallel to the fixed electrodes in the usual servo-operation. From an analysis using an energy model considering electrostatic energy of the electric field between the movable and the fixed electrodes and elastic energy of the cantilever beam which sustains the movable electrode, the occurrence of rotation of the movable electrode around a certain axis was expected at pulse voltages higher than a certain threshold voltage. This threshold voltage VT is proportional to the square root of the rotational spring constant of the cantilever beam when the movable electrode rotates around a certain axis.
This rotation phenomenon of the movable electrode was confirmed by optical observation of the sensor and the measured threshold voltages VT were in close agreement with those calculated by the above energy model.
The maximum measurable acceleration αmax in a PWM electrostatic servo accelerometer is proportional to the square of the pulse voltage which is restricted by the threshold voltage VT and is larger in the sensor structure with larger rotational spring constant of the cantilever beam. for example, αmax for a 2-beam sensor is larger than that for a 1-beam sensor.View full abstractDownload PDF (2379K) -
Hiroshi KOYAMA, Kajiro WATANABE1993 Volume 29 Issue 3 Pages 295-301
Published: March 31, 1993
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper describes an acoustic method to monitor the state of the inside of gas pipeline based on the acoustic propagation in the pipelie. The proposed method can monitor the inside of the pipeline from only one terminal of the pipeline.
Echoes of the sound in the pipeline occur at the structural changes like as the terminal end, the partial plugging and the connecting branches, when white noise sound is fed at the open terminal. The location and the degree of the structural changes can be detected from the delay time and amplitude of the echoes.
The proposed method can monitor the following items:
1. State of the other end of a single pipe (Closed or Open) and length of the pipe.
2. Location and degree of partial plugging in a single pipe.
3. Location of partial plugging and, branches and the length of the branches.
Experiments carried out by a laboratory pipe and real field pipeline demonstrated the validity of the method and the monitoring accuracy was reasonably good.View full abstractDownload PDF (1404K) -
When Measured Variables are not Coincident with Controlled VariablesNobuhide SUDA, Qiuling WANG1993 Volume 29 Issue 3 Pages 302-311
Published: March 31, 1993
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper analyzes a linear time-invariant two-degree-of-freedom decentralized control system. A general solution of the robust servo problem in the decentralized control system when measured variables are not coincident with controlled variables is derived.View full abstractDownload PDF (1529K) -
Kang-Zhi LIU, Tsutomu MITA1993 Volume 29 Issue 3 Pages 312-318
Published: March 31, 1993
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper deals with the steady-state performance of lumped LTI feedback systems in the face of plant uncertainty. The plant uncertainty considered here is the multiplicative perturbation at the output port. The necessary and sufficient conditions for various robust steady-state performances are obtained. Internal model principle like results on the structure of controllers achieving various robust steady-state performances are derived from these necessary and sufficient conditions. More importantly, these conditions are transformed into the measure of μ, thus enables the synthesis of controllers achieving robust steady-state performance in the framework of μ synthesis and the generalized H∞ control theory.View full abstractDownload PDF (891K) -
Akira KOJIMA, Kenko UCHIDA, Etsujiro SHIMEMURA1993 Volume 29 Issue 3 Pages 319-325
Published: March 31, 1993
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSFor the control of time delay systems, a synthesis based on predictive controllers is well known as a remedy to cancel the effect of time delay. And further, it is also indicated that the control law derived via the pole assignment problem or the LQ-problem has a feature of state prediction. This fact implies that the argument of state prediction forms an essential part for the control of time delay systems. However in the context we are faced with the plant uncertainties, the feature of required robust stabilizing law is no longer obvious.
In this paper, a robust stabilization problem is discussed for a system with delays in control. We derive a robust stabilizing law against additive perturbations. Then, motivated by the structure of predictive controllers, we provide some interpretations on the structure of the required robust stabilizing law. The key point in the derivation is to transform the time delay system into a lumped parameter system such that both systems share same poles and provide equivalent input/output mappings. This approach first enables us to characterize the solvability of the problem based on finite-dimensional Riccati equations.
Finally, it is also clarified that a robust stabilizing law which guarantees the stability degree of the closed loop system can be derived in like fashion.View full abstractDownload PDF (1558K) -
Masanobu KOGA, Katsuhisa FURUTA1993 Volume 29 Issue 3 Pages 326-333
Published: March 31, 1993
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSWe define complex-systems of degree n, corresponding to real-systems of degree 2n, whose inputs, outputs, states, and parameters are all complex values. The introduction of such complex-systems makes it more suitable to treat not only the robust control but also the pole assignment in separate regions. The relation between the feedback-control law of the real-system and that of the corresponding complex-system is derived. A feedback-control law to assign all the poles in the prespecified separate regions is determined by using the complex-system.View full abstractDownload PDF (967K) -
Shigeru YAMAMOTO, Qi SHEN, Hidenori KIMURA1993 Volume 29 Issue 3 Pages 334-339
Published: March 31, 1993
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper considers the quadratic stabilization for a class of uncertain linear systems. The system under consideration contains norm-bounded time-varying uncertainties which linearly depends on scalar parameters. The quadratic stability problem for the system is reduced to finding a common positive definite solution of 2m Lyapunov inequalities, where m is the number of uncertain scalar parameters. We derive a sufficient condition for the quadratic stabilizability of the uncertain system in terms of a Riccati equation containing 2(m+1) free parameters. The results are applied to the stabilizing control of a magnetic levitation system. The effectiveness of our methods is illustrated both in simulations and experiments.View full abstractDownload PDF (1002K) -
Yasushi YOKOYAMA, Takehisa KOHDA, Koichi INOUE1993 Volume 29 Issue 3 Pages 340-346
Published: March 31, 1993
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, firstly we propose a neural network with a mixed structure, which consists of multilayer and recurrent structure, for learning the dynamics of a nonlinear plant. A neural network with a mixed structure can learn time series, therefore, it can learn the plant dynamics without knowning the plant order.
Next, we consider the optimal control synthesis problem using the neural network with a mixed structure, which has learned the plant dynamics completely, as a plant model. Procedures are as follows: (1) the neural network is expanded into an equivalent feedforward multilayer network, (2) it is shown that the gradient of criterion functional to be optimized can be easily obtained from this multilayer network, and then (3) the optimal control is generated by applying any of the existing nonlinear programming algorithms based on this gradient information.
The proposed method is sucessfully applied to the optimal control synthesis problem of a nonlinear coupled vibratory plant with a linear quadratic criterion functional.View full abstractDownload PDF (1400K) -
Tetsuo KOTOKU, Kazuo TANIE1993 Volume 29 Issue 3 Pages 347-355
Published: March 31, 1993
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSGraphic simulators are currently used in application such as off-line teaching tasks on industrial robots and CAD systems. While recent advances in computer graphics have led to the development of high powered computers with high quality static and dynamic graphics, the display of kinesthetic information is still lacking. This display of kinesthetic information is critical for the development of simulators that can more accurately and realistically present simulated objects to the user. With this problem in mind, this paper proposes a method of calculating interaction forces produced in the simulation world. These forces are effectively used in a simulator where contact forces are felt by the user through its force-reflecting handle. In this method, two fundamental contacts, Face-Vertex and Edge-Edge, will be introduced, and the interaction force calculations will be defined. It will be demonstrated that every contact between two convex polyhedral objects can be described using this set of the fundamental contacts. The interaction forces will be determined from the sum of forces calculated at each fundamental contact. Using a simulator system which displays forces, certain preliminary experiments are carried out and the effectiveness of the method confirmed.View full abstractDownload PDF (3109K) -
Toshio FUKUDA, Shigetoshi SHIOTANI, Fumihito ARAI, Takanori SHIBATA, K ...1993 Volume 29 Issue 3 Pages 356-364
Published: March 31, 1993
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSAritificial Neural Network (NN) is applied to control, recognition and so on. The multi-layered Neural Network with sigmoid function (NNS) is often used in these fields because the NNS has two abilities of the interpolation and generalization. The generalization is, for example, for NN to recognize unlearned patterns to a certain extent. After this, we discuss recognition problems using NN. The NNS must learn both unlearned patterns and patterns given before to memorize unlearned patterns additionally, because the NNS cannot learn the patterns additionally. ART (Adaptive Resonance Theory) model can memorize the patterns additionally. However the ART model cannot classify patterns which have same pattern vectors.
A new neuron model called a Neural Network based on the distance between patterns (NDP) is proposed in this paper. The NDP has similar response functions to the Radial Basis Function. The NDP learns patterns by varying regions of neurons with the BP algorithm and adding new neurons. It is shown from experimental results on image recognition that the NDP can memorize patterns additionally and recognize unlearned patterns to some degree.View full abstractDownload PDF (2188K) -
Koichi SUYAMA, Noboru SEBE1993 Volume 29 Issue 3 Pages 365-367
Published: March 31, 1993
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper studies the existence of a supplementary compensator which can simultaneously achieve both reliable stability with passive redundancy and an arbitrary transfer function matrix of the resulting total compensator.View full abstractDownload PDF (344K) -
Kazuyuki JOYA, Katsuhisa FURUTA1993 Volume 29 Issue 3 Pages 368-370
Published: March 31, 1993
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper treats D-stability of a family of polynomials where D is an arbitrary domain or the union of domains which do not intersect each other. The main result of this paper is a criterion with which we can easily check robust D-stability of a family of polynomials whose coefficients are constrained with lp-norm. A numerical example is given to show usefulness of the criterion.View full abstractDownload PDF (339K) -
Toshiro NORITSUGU, Hui YANG1993 Volume 29 Issue 3 Pages 371-373
Published: March 31, 1993
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper proposes a master-slave control method for a cooperative control of human and robot. A selection matrix is introduced to share control variables between a master control and an autonomous slave control. This method is applied to the object following control of a robot manipulator with a visual sensor. Some experiments show the effect of selection matrix and the availability of this cooperative control to fuse a human skill into a robot control ability.View full abstractDownload PDF (369K) -
Toshio TSUJI, Tohru YAMANAKA1993 Volume 29 Issue 3 Pages 374-376
Published: March 31, 1993
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe present paper proposes a trajectory generation method for multiple manipulators. Firstly, kinematics of the manipulators connected each other is divided into a set of closed-chains formed by two manipulators. Then, the instantaneous inverse kinematics of the multiple manipulators is derived using virtual arms. The method can generate trajectory of the manipulators utilizing kinematic redundancy.View full abstractDownload PDF (756K)
- |<
- <
- 1
- >
- >|