計測自動制御学会論文集
Online ISSN : 1883-8189
Print ISSN : 0453-4654
ISSN-L : 0453-4654
28 巻, 3 号
選択された号の論文の18件中1~18を表示しています
  • 北山 正文
    1992 年 28 巻 3 号 p. 297-300
    発行日: 1992/03/31
    公開日: 2009/03/27
    ジャーナル フリー
    Recently, it has been found that there is a constant difference between the results of the particle size distribution measured with the sieve method and the optical microscope image processing method.
    The causes for this are due to the particles being destroyed with the sieve and the dependence on the varied shape of the particles and how they flow through the mesh of the sieve.
    In this paper, the relation between the mesh size and the shape of the particle is modelled and the formula which expresses the relation between the equivalent mesh diameter Dm of the sieve and the equivalent volume diameter Dp is established.
    The propriety of this model is shown by the experimental results.
  • 寺尾 吉哉, 高本 正樹, 片桐 拓朗
    1992 年 28 巻 3 号 p. 301-305
    発行日: 1992/03/31
    公開日: 2009/03/27
    ジャーナル フリー
    The calibration results of 61 commercially available air flow velocimeters over the velocity range of 0.05 to 1m/s are presented. The calibration was performed with the towing carriage system which is the national standard for low range air velocity. The effect of the residual flow along the rail on the calibration result was estimated experimentally. It is proved that the calibration uncertainty was sufficiently minimal.
    The error of the thermal type velocimeters showed vast scattering against the standard velocity. It was found that the error was due to the following facts. (1) The velocity standards of the manufacturers are inconsistent with the national standard. (2) The secular change is fairly large. (3) The sensitivity significantly changes below 0.2m/s.
    When the flow velocity changed from zero to 0.5m/s, most of the thermal type velocimeters showed longer response time than those stated in their specification.
  • 小林 一行, 渡辺 嘉二郎, 大川 洋児, 松本 茂
    1992 年 28 巻 3 号 p. 306-312
    発行日: 1992/03/31
    公開日: 2009/03/27
    ジャーナル フリー
    In this paper we consider how to realize the radio wave speedometer that uses the spatial pattern of electric field intensity of radio waves propagated from a broadcasting station as the spatial measure. The radio wave speedometer measures the speed relative to the space in which radio waves are standing, whereas the conventional methods like as the Doppler method, the spatial filtering method and the correlation method measure the speed relative to the ground. The prior research verified the validity of the radio wave speedometer under restrictive circumstances such as on the roof of a building.
    Here we investigate if the radio wave method can work correctly under real situations and circumstances where many vehicles are running. We developed a two channel electric field intensity meter with the wide dynamic range and quick response. The properties of the electric field pattern i.e., (1) the stability of the pattern, (2) the difference of the pattern when moving to different directions, and (3) the influences by the moving reflectors were investigated by using the meter developed. Further we investigated if the radio wave speedometer can measure the accurate speed of a vehicle in the variety of real roads by experiments.
    It was verified that the patterns were stable at most places where we carried out experiments and little influences were observed by the moving reflectors like as automobiles. Thus the patterns can be used as the spatial measure for the speed measurement in real situations. The speed measured by the radio wave speedometer showed the good reproducibility and linearity.
  • 高田 等
    1992 年 28 巻 3 号 p. 313-319
    発行日: 1992/03/31
    公開日: 2009/03/27
    ジャーナル フリー
    This paper presents a numerical computer method of nonlinear two-point boundary value problems using a formal linearization method. We have considered a nonlinear system with two-point boundary conditions. The nonlinear system is transformed into a linear system by the formal linearization method. The transforming function is given as a solution of a partial differential equation and is solved by Taylor or orthogonal expansions. The missing values at the boundary points are evaluated using the Newton method. As a result, the nonlinear two-point boundary value problem becomes a nonlinear initial value problem which can be easily solved. The error bounds of the presented method are also investigated.
    Two examples are included to demonstrate the effectiveness of this method. One of them is the case of a transforming function of polynomials, the other being the case of a nonlinear optimal control problem using the maximum principle.
  • 大塚 尚久
    1992 年 28 巻 3 号 p. 320-325
    発行日: 1992/03/31
    公開日: 2009/03/27
    ジャーナル フリー
    A simultaneous block triangular-decoupling and disturbance-rejection problem with incomplete-state feedback for linear multivariable systems is studied in the framework of geometric approach. Some solvability conditions for this problem are presented. Further, an illustrating example is given.
  • Suthee PHOOJARUENCHANACHAI, Katsuhisa FURUTA
    1992 年 28 巻 3 号 p. 326-333
    発行日: 1992/03/31
    公開日: 2009/03/27
    ジャーナル フリー
    This paper presents a finite dimensional stabilizing scheme for a class of time-delay systems having some imprecisely known parameters. The technique is based on a new stability criterion simply derived by the classical method of Lyapunov functionals. Frequency domain interpretation of the criterion in H-norm sense is also given. A sufficient condition for stabilizability via dynamic output feedback follows by a simple application of the stability result. It is shown that solutions to two parametrized Riccati equations play central role in synthesis of a suitable stabilizing controller. The resulting controller is of finite order, and its formula is expressed in terms of system parameters. Besides these results, some relations between the theory presented here and the well-known quadratic stability theory are explored.
  • 石原 正, 鄭 良安, 猪岡 光
    1992 年 28 巻 3 号 p. 334-342
    発行日: 1992/03/31
    公開日: 2009/03/27
    ジャーナル フリー
    Taking account of feedback delays, we discuss a state-space representation of the discrete-time doubly coprime factorization. First, we give eight proper and stable matrices satisfying the relation of the doubly coprime factorization based on the predictor-based LQG controller which consists of a filtering type Kalman filter and a state predictor. It is found that the factorizations for different delays are related to each other by a simple free parameter in the YJB parametrization. Then we propose a simple and convenient parametrization for all stabilizing controllers under feedback delays. The free parameter in this parametrization can be an arbitrary proper and stable rational matrix. Using this parametrization, we can easily apply various new design techniques, which are usually proposed for continuous-time systems, to discrete-time systems under feedback delays. Finally, as an application, we discuss loop transfer recovery using the free parameter in the proposed parametrization.
  • 参照入力の一般化
    藤崎 泰正, 池田 雅夫
    1992 年 28 巻 3 号 p. 343-350
    発行日: 1992/03/31
    公開日: 2009/03/27
    ジャーナル フリー
    A two-degree-of-freedom optimal servosystem for reference signals generated by a linear differential equation is proposed. First, an optimal tracking system with no servocompensator is designed for a given model of the plant and a specified performance index. The obtained control system is augmented with a servocompensator. Then, a state feedback from the plant, a feedforward from the generator of reference signals, and an external signal having the modes of a servocompensator, are introduced to cancel the output of the servocompensator, if there is no modeling error of the plant and no disturbance input to the plant. When the modeling error or the disturbance input exists, the effect of servocompensation appears.
  • チャタリングを低減させる仮想目標点の提案
    坂本 憲昭, 増淵 正美, 川田 誠一
    1992 年 28 巻 3 号 p. 351-357
    発行日: 1992/03/31
    公開日: 2009/03/27
    ジャーナル フリー
    As the robust control strategy which assures stability for linear systems having uncertainties caused by parameter variations, the one using linear and nonlinear laws by G. Leitmann and others based on the second method of Lyapunov has been known in these ten years.
    For this control strategy, however, saturating function approximating discontinuous function is used and it has been pointed out when implemented to existing plant that chattering may be generated at the manipulating variable due to the input delay caused by the sampling time in the computer system.
    This paper proposes a new algorithm of Moving Attractor to reduce this chattering effect. The control deviation is not evaluated by using the existing deviation but by using the deviation between the Moving Attractor and controlled variables. Position of Moving Attractor is first given at a prescribed place and then is moved to the position of real command under some given rule. The rule of Moving Attractor which assures uniform boundedness and uniform ultimate boundedness is derived.
    It is shown by an example of a robotic arm that the chattering at the manipulating variable is remarkably reduced under this proposed system.
  • 廣木 富士男, 山本 圭治郎, 増田 泰二
    1992 年 28 巻 3 号 p. 358-365
    発行日: 1992/03/31
    公開日: 2009/03/27
    ジャーナル フリー
    A new type fluidic oscillator had been developed and the fatigue testing machine to which this oscillator is applied is now used to study the mechanism of impact fatigue fracture of reed valves for compressors and engines. Since the output flow of this oscillator is used directly as working fluid of the machine, the oscillator must be operated under high supply pressure range above the critical one. The oscillator was realized with the construction that two outlets of a supersonic bistable device were bifurcated and the load capacitor was connected to each bifurcated channels. Thereby, the output jet can keep high power without losing energy through the load capacitor.
    The fluidic circuit of the oscillator has been analyzed by a lumped parameter equivalent circuit model based on some parameters of a load capacitor, then the calculated values of oscillation frequencies have been compared with the experimental values. It was appeared that the inductive and thermal conductive effects in the load capacitor must be considered especially in the case of low aspect ratio of the load capacitor.
    Switching condition was clarified that the pressure in the load capacitor rises up to 80% of the total pressure of the flow in the outlet. With this analytical model, the oscillation frequency can be estimated for various configurations of the load capacitor.
  • 辻 敏夫, 三木 義民, 伊藤 宏司
    1992 年 28 巻 3 号 p. 366-373
    発行日: 1992/03/31
    公開日: 2009/03/27
    ジャーナル フリー
    In the present paper, we examined the nature of the human's hand forces which were exerted on the rigid objects during voluntary constrained tasks. Experimental observations found that when the force was exerted on the objects arbitrarily, subjects tended to generate force patterns. the derivatives of which had roughly single-peaked, bell-shaped profiles. This was held independent of the stage of skill, the arm postures, and the force directions.
    Force coordination was modeled mathematically by defining a virtual trajectory and limb elasticity. The virtual trajectory is the sequence of equilibrium points which the limbs are intended to follow due to the modulable elastic properties of the muscles. Then, it is specified through an optimality criterion that the mean square jerk (rate of change of acceleration) integrated over the virtual trajectory is minimal for all possible choices of trajectories. This is equivalent to assuming that a major goal of motor coordination is the production of the smoothest possible force of the hand. The measured force profiles confirmed the predictions of the mathematical model quantitatively.
  • 山田 克彦, 土屋 和雄
    1992 年 28 巻 3 号 p. 374-382
    発行日: 1992/03/31
    公開日: 2009/03/27
    ジャーナル フリー
    The subject of this paper is a space robot that has a manipulator arm mounted on a satellite body. As the attitude of the space robot is varied due to the arm motion, it is necessary to understand the relation of the arm trajectory and the attitude variation. In this paper, we evaluate the attitude variation and obtain the hand trajectory to minimize it for the simple case where the arm mass is concentrated on the hand. We study here two coordinate frames of the hand: an inertial frame and a satellite-fixed frame. For the inertial frame, the attitude variation is proportional to the area bounded by the hand trajectory and the origin. For the satellite-fixed frame, after the coordinate frame is transformed, the attitude variation is also proportional to the area bounded by the hand trajectory and the origin. In both cases, the hand trajectory to minimize the attitude variation is approximated by a hyperbola. Numerical simulation is executed to examine these results.
  • 梶原 逸朗, 長松 昭男, 稲垣 亨
    1992 年 28 巻 3 号 p. 383-391
    発行日: 1992/03/31
    公開日: 2009/03/27
    ジャーナル フリー
    The authors propose a method to design the optimum control system toward flexible structural systems. The mechanical structures used to be modelled by FEM. As a result, these models have large degrees of freedom. The small model can be obtained by identifying the spatial matrices of the structure from the frequency response functions calculated by FEM. The optimum control system is easily designed by applying the optimum control theory to the identified system. However, the stability of the control system is not always guaranteed because the machanical system is generally identified with small degree of freedom in which high order modes are ignored. An optimization method of the control system which guarantees the stability of the actual system of large degrees of freedom is proposed in this paper. The characteristics of both the identified and the actual systems are simultaneously optimized by the proposed method based on sensitivity analysis and nonlinear programming. An application to a FEM model and experimental result verify the effectiveness of the proposed method.
  • 阪口 豊, 中野 馨
    1992 年 28 巻 3 号 p. 392-400
    発行日: 1992/03/31
    公開日: 2009/03/27
    ジャーナル フリー
    When perceiving an object, human beings can construct its internal image from its fragmental features and recognize it based on the image, without observing it all over. In this article, the authors propose a neural network model which constructs internal representation of a given figure and recognizes it based on the representation.
    The model consists of a sensory unit, a figure representation field and a recognition layer. The model moves the sensory unit and observes a local feature of the figure. The figure representation field consists of many “feature units”, whose activation represents the figure. These units interact through mutual connections which reflect topological structure of the external world. Through the interaction, the field constructs global representation of the figure from the local features observed by the sensory unit.
    The recognition layer makes a recognition output using the field activation. It means that the recognition is based not on sensory information itself but on internal representation modulated by the field dynamics. This “internal-representation-based recognition” bears an aspect of human recognition. By imposing external input reflecting “context” to the field, for instance, “contextdependent recognition” is realized.
    In addition, the authors propose an active searching algorithm which helps representation construction to advance smoothly. In the course of construction, competition often arises between feature units, especially around edges or corners of the figure, which have significant meaning for constructing the representation. The model can observe “intentionally” such essential parts by moving the sensory unit to the corresponding position. Such competition-invoked searching is a new algorithm for active perception.
    The model behavior is investigated through computer simulation. Some related matters on representation in the brain are discussed.
  • 橋爪 進, 小野木 克明, 西村 義行
    1992 年 28 巻 3 号 p. 401-408
    発行日: 1992/03/31
    公開日: 2009/03/27
    ジャーナル フリー
    Petri nets often contain unnecessary places whose existence does not affect the behavior of net. To detect such redundant places is important to analysis and design of nets. This paper deals with such reundancy of condition/event-nets (C/E-nets).
    Every C/E-net is composed of simple nets, containing only one condition, called atoms. Therefore the construction problem of C/E-nets with desired behavior is reduced to the search for constituent atoms of nets. This paper first clarifies the relation between the individual behavior of each constituent atom and that of the over-all net. It next defines a redundant atom and discusses its properties. On this discussion, the paper then proposes a method of finding redundant conditions in nets. It last shows a sufficient condition for a net constituent atom to be redundant.
  • 明見 幸泰, 太田 裕文
    1992 年 28 巻 3 号 p. 409-411
    発行日: 1992/03/31
    公開日: 2009/03/27
    ジャーナル フリー
    Design of a robust controller for linear time invariant plants with both input and output uncertainties is considered. Using structured singular value, less conservative robust stability condition for the system is obtained, and robust controllers are derived analytically from μ-synthesis.
  • 伊藤 博, 大森 浩充, 佐野 昭
    1992 年 28 巻 3 号 p. 412-414
    発行日: 1992/03/31
    公開日: 2009/03/27
    ジャーナル フリー
    An algorithm for designing stable compensators which internally stabilize a linear time-invariant SISO system and attain low sensitivity property is presented in this paper, based on Nevanlinna-Pick interpolation. Stable compensators are practically desirable when we consider the integrity of the system.
  • 陶山 貢市
    1992 年 28 巻 3 号 p. 415-417
    発行日: 1992/03/31
    公開日: 2009/03/27
    ジャーナル フリー
    From the viewpoint of pole-zero cancellation, this paper studies sampled-data PID control systems designed for a certain class of plant models by the partial model matching method.
feedback
Top