巻号一覧
12 巻, 1 号
選択された号の論文の20件中1～20を表示しています
- |<
- <
- 1
- >
- >|
-
有本 卓, ガーファル モハメッドＹ．, 中井 隆洋, 宮崎 文夫1976 年 12 巻 1 号 p. 1-6
発行日: 1976/02/28
公開日: 2009/03/27
ジャーナル フリーThis paper is concerned with the horizontal guidance of an aircraft in and near the terminal area when no flying is permitted in a half plane of the space. The problem of guiding an aircraft in minimum time from arbitrary initial position and heading to fixed terminal position and heading is considered, under the condition of constant speed and with the constraint on bank angle. It is found that an optimal path cannot have a boundary segment on the boundary of the constrained space, more than one junction point and more than three switching times, while if the flight path remains interior then the maximum number of the switching times is two. Furthermore, in view of these results it is found that there are at most twelve extremal controls and their control regions are not disjoint. The optimal control is chosen to minimize the performance index among extremal controls.抄録全体を表示PDF形式でダウンロード (846K) -
富塚 誠義1976 年 12 巻 1 号 p. 7-12
発行日: 1976/02/28
公開日: 2009/03/27
ジャーナル フリーThe optimal finite preview control problem is formulated for the continuous time system. It is assumed in the present problem that in deciding control the controller can make use of preview information with respect to a command signal and/or a disturbance from the present time t up to tla time units in the future beyond time t. The time interval tla is called the preview time and usually tla<tf-t0, where t0 and tf are the initial time and the final time respectively. Using the local future information obtained by a finite preview, the optimal control is determined so that it minimizes the cost functional evaluated over the entire time interval [t0, tf]. The solution is obtained by dynamic programming. The steady state solution is obtained for the case when the duration of problem is infinite and the quantities that appear in the problem are all time invariant. For the trajectory following type problem, the transfer function between the command signal and the output of the optimally controlled plant is derived. An example shows the structural properties of the optimal finite preview control system.抄録全体を表示PDF形式でダウンロード (708K) -
迷路内ほふく推進の解析と実験梅谷 陽二, 広瀬 茂男1976 年 12 巻 1 号 p. 13-17
発行日: 1976/02/28
公開日: 2009/03/27
ジャーナル フリーThe snake makes the most of its locomotive capability by skillfully adapting its flexible slender body to such environments as desert, swampy places, rough stony ground or dense bush where othere animals with limbs as well as artificial terrain vehicles can not easily traverse.
The authors previously studied the snake's serpentine movement and obtained some fundamental knowledge concerning the snake's gliding movement. Their studies so far were limited to the case that a snake moves steadily on an even surface such as on the lawn. A more generalized kinematical theory is needed to analyse the serpentine movement in terrestrial environments other than an idealized even surface treated hitherto.
In this paper, the fundamental kinematical relations, which can be applied for every type of serpentine movement, are derived in the first place by generalizing the relations previously used to analyse a steady straightforward serpentine movement.
Next, among various terrestrial environments, narrow labyrinth composed of two walls facing at the same distance is chosen as the typical and the most fundamental model, and theoretical gliding shape of the snake at the bent corner in this labyrinth is discussed from the kinematical standpoint.
Finally these derived theoretical relations are verified by the zoological experiment using snakes, Elaphe Quadrivirgata. The feasibility of applying these knowledges to the controlling of artificial snake-like vehicle is also discussed.抄録全体を表示PDF形式でダウンロード (1261K) -
田村 捷利, 神谷 正紀1976 年 12 巻 1 号 p. 18-24
発行日: 1976/02/28
公開日: 2009/03/27
ジャーナル フリーThis paper is concerned with a solution of an optimal control problem with a steady state on an intermediate period of the control interval. On the remaining periods the system is assumed to be in a transient state. The problem consists of three successive phases in which the system equations and the criterion indexes are assumed to be different from each other.
In order to develop an algorithm the trajectory decomposition method is employed and then a multilevel calculation system is considered. At the first level of the calculation system, three independent optimization subproblem, which correspond to the three phase, are solved. At the second level, inference variables between phases are chosen so as to decrease the cost, functional of the problem and are fed to the first level. The algorithm is of a min-min type.
The optimal train operation curves are obtained by the algorithm.抄録全体を表示PDF形式でダウンロード (846K) -
荒木 光彦1976 年 12 巻 1 号 p. 25-28
発行日: 1976/02/28
公開日: 2009/03/27
ジャーナル フリーIt is shown that the input-output stability criterion of composite systems given in terms of the M-matrix can be derived as a special case of the small gain theorem by Zames.抄録全体を表示PDF形式でダウンロード (578K) -
野崎 武敏1976 年 12 巻 1 号 p. 29-34
発行日: 1976/02/28
公開日: 2009/03/27
ジャーナル フリーThis paper describes a course deviation detecting method for the automatic driving of an automobile. The remarkable feature in this method is that the course deviation at some distance ahead of an automobile is to be detected. For this purpose, the light reflector is installed on the road surface as the reference for the automatic driving, and a TV camera is used as the sensor which detects the course deviation. The fundamental of the information processing is that the light reflector is sufficiently brighter than any other part of the road surface. Practical tests of the detector equipped on the automobile gave good results of its operating stably at a running speed of about 40km/h.抄録全体を表示PDF形式でダウンロード (1999K) -
単てこの動力学(理論)西山 卓, 増尾 龍一, 前田 親良, 篠崎 順一郎1976 年 12 巻 1 号 p. 35-39
発行日: 1976/02/28
公開日: 2009/03/27
ジャーナル フリーIntending to design shipboard scales, the authors have discovered two ways: the first involves determining the difference in the angles of deviation of two independent levers which swing in a similar manner when the base moves; the second involves designing the levers so as to cancel the influence of the motion of the base as a whole.
In the both cases, it is basically important to understand the behavior of the simple lever suspended by an elastic fulcrum which is fixed on the moving base.
In this paper, the authors derive the general equation of motion for the simple lever, assuming that the base swings according to θ=θ0 cos ωt, where θ is the reclining angle of the base.
Solving the equation, the authors characterize the behavior of the simple lever.
In the following paper, experimental results which agree well with the above theory will be presented.抄録全体を表示PDF形式でダウンロード (497K) -
浜野 文男, 古田 勝久1976 年 12 巻 1 号 p. 40-45
発行日: 1976/02/28
公開日: 2009/03/27
ジャーナル フリーThe purpose of this paper is to investigate the structural properties of the decentralized time-invariant linear multivariable systems in connection with the problems of disturbance localization and output decomposition. In the first place, the necessary and sufficient conditions, under which the disturbance localization is possible by applying local state feedback laws, are given. The conditions are also stated for ordinary (centralized) systems with imperfect state measurement and for those with output measurement. Finally, the problems of output decomposition in decentralized linear multivariable systems are proposed, and the necessary and sufficient conditions for the solvability of the problem are derived.抄録全体を表示PDF形式でダウンロード (702K) -
宮川 清孝1976 年 12 巻 1 号 p. 46-50
発行日: 1976/02/28
公開日: 2009/03/27
ジャーナル フリーThe study aimed at realizing a stable freezing point of tin and to utilize the freezing point in the practical appratus for calibration of a thermometer.
The apparatus features a stable super-cool which takes place in tin filled in a graphite crucible without an excessive falling rate of temperature by means of moderate cooling effect of circulating air in the copper pipe surrounding the crucible.
Two measuring wells of the crucible allow the accurate measurement of freezing point or the concurrent inspection of two resistance thermometers.
In case of 1.5kg of tin molten in the crucible, the temperature of freezing point of 99.9999% tin is 231.9681±0.001°C, which is higher than that of 99.999% tin by 1mK, and has a good agreement with that in IPTS-68.抄録全体を表示PDF形式でダウンロード (742K) -
服部 晋1976 年 12 巻 1 号 p. 51-56
発行日: 1976/02/28
公開日: 2009/03/27
ジャーナル フリーThis blackbody at the gold point was newly designed and constructed to establish the temperature scale above the gold point (1064.43°) by the definition of the International Practical Temperature Scale of 1968 (IPTS-68).
The blackbody is composed of a graphite crucible and a furnace. The graphite crucible contained a 0.9kg gold ingot with purity of 0.99999 in which opens a reentrant graphite cavity forming a blackbody. The effective emissivity of the cavity was evaluated to be 0.99998. The furnace can maintain the temperature of the crucible at 1064.5° with a power supply as low as about 800W by the aid of the radiation shield. The temperature along the furnace axis is maintained within 0.1K by two control blocks.
At the blackbody of the gold point, the slope of the straight lines on the observed freezing plateau and the dispersion of the replication gave estimations that the desired blackbody radiation was correctly realized within the precision of the multiplicative comparator of the spectral radiance which was used to compare the spectral radiance of the blackbody with that of a tungsten strip lamp.
Thus the accuracy of the blackbody at the gold point (ΔLλ[T68(Au)]/Lλ[T68(Au)]) is expected to be better than 0.0001.抄録全体を表示PDF形式でダウンロード (929K) -
神谷 祐二1976 年 12 巻 1 号 p. 57-62
発行日: 1976/02/28
公開日: 2009/03/27
ジャーナル フリーMany methods to design the control system which are known up to now are formed under the assumption that there are no variation in the parameters and no noises in the controlled object. The variation means the measurment error and the change with time.
But, it is apparent that this assumption is not satisfied in many practical controlled objects. Therefore, some means are necessary to approximate the practical controlled object by an idealized model which satisfies this assumption.
This paper proposes one way to compose the model following system. This system can be applied to the practical controlled objects in which there are multi inputs and outputs and only the outputs can be observed directly. Its structure is such that the outputs of the practical controlled object follow the outputs of the model which is constructed with the measured values of the parameters of the practical controlled object.
In this system, the following error can be controlled at will. Consequently, we can consider the model following system as the new controlled object which has not the influence of the variation of parameters and noises. And in addition, all state variables can be observed directly, for they can be obtained from the model.抄録全体を表示PDF形式でダウンロード (696K) -
奥田 徹示, 田中 英夫, 浅居 喜代治1976 年 12 巻 1 号 p. 63-68
発行日: 1976/02/28
公開日: 2009/03/27
ジャーナル フリーMuch of decision-making in the real world takes place in an environment in which states of nature, feasible actions and available information are fuzzy. Therefore it is important to study decision problems in such an environment. An example of fuzzy states may be described by such expressions as “It will be warm, ” “It will be cool, ” etc. This fuzziness stems from such adjectives as “warm, ” “cool, ” etc. Since fuzzy states are defined as fuzzy events which are fuzzy sets on a probability space, the decision problem is related to fuzziness and randomness.
The present paper deals with a fuzzy decision problem in this environment.
The application of the fuzzy sets theory and the statistical decision theory to decision problems with fuzzy events leads to a specific formulation of fuzzy decision problems and to the definitions of entropy, worth of information and quantity of information. On the basis of these definitions, this paper gives some analytical results concerning perfect, probabilistic and fuzzy information, which are analogous to those in the statistical decision theory, and shows the advantages for considering the decision problem with fuzzy events.抄録全体を表示PDF形式でダウンロード (840K) -
田村 坦之, 青谷 年永1976 年 12 巻 1 号 p. 69-75
発行日: 1976/02/28
公開日: 2009/03/27
ジャーナル フリーAs a new application of GMDH, a largespatial pattern identification problem is solved where space is the independent variable of the system. The interest to solve this problem has been stimulated by the practical situation of monitoring air pollution concentration. Selection of input variables is one of the most important tasks in the GMDH. Two kinds of input variables are tested for our purpose:
Type 1: (x, y) coordinates
Type 2: (x, y) coordinates and the output of pointwise information at the neighboring points of a prediction point.
Although Type 2 involves more computational tasks than Type 1, we can obtain better accuracy of pattern identification from Type 2.
GMDH algorithm is implemented in the Time Sharing System (TSS), therefore we could use the computer program by the man-computer interactive way which enables us to change some heuristics in the GMDH easily.抄録全体を表示PDF形式でダウンロード (855K) -
高橋 三起夫, 久村 富持1976 年 12 巻 1 号 p. 76-83
発行日: 1976/02/28
公開日: 2009/03/27
ジャーナル フリーThis paper investigates an optimization problem of a class of distributed parameter systems with spatially concentrated controls whose positions, as well as control magnitudes, are assumed to be changeable in the spatial domain.
The following two cases are discussed.
(1) In a control period, the control positions can be movable with time in the defined domain, and also the control magnitudes can be changeable.
(2) The control positions must be selected optimally before the controls are applied.
For the both cases, the necessary conditions for optimality are derived by the use of a variational method. An example is shown to illustrate the effect of moving control positions and the result shows an improvement in the performance index compared with the system with fixed control positions.抄録全体を表示PDF形式でダウンロード (766K) -
藤川 英司, 示村 悦二郎1976 年 12 巻 1 号 p. 84-88
発行日: 1976/02/28
公開日: 2009/03/27
ジャーナル フリーIn this paper, an approximate method of solving the optimal control problems of the system with time-delay is discussed. The mathematical model of the system with time-delay is formulated as the special case of the coupled systems of the lumped parameter subsystem and the distributed parameter subsystem. The state variables of this coupled system are infinite dimensional and some of the control variables are distributed in space. Therefore it is necessary in practice to implement the control system with a finite number of observers and a finite number of controllers. That is, one should design the practical control system for the approximated finite dimensional system. In this paper the Galerkin method is applied to obtain the approximated system. The characteristic features of this method and the results of the simulation are described.
From the computational experience, it is shown that the proposed method requires in many cases a smaller number of observers and controllers compared with the conventional finite difference method to achieve the same degree of control performance.抄録全体を表示PDF形式でダウンロード (710K) -
Certainty Equivalenceについて内田 健康, 示村 悦二郎1976 年 12 巻 1 号 p. 89-95
発行日: 1976/02/28
公開日: 2009/03/27
ジャーナル フリーIn this paper optimal control problems of linear stochastic systems with a quadratic criterion under a classical information structure are discussed. The emphasis is put on an effort to derive the condition under which the certainty equivalence property holds, and it is shown the most essential sufficient conditions are (I) that the system is linear, (II) that the performance criterion is quadratic and (III) that the information structure is classical. These conditions do not require any assumption on the statistical property of noise processes and an initial state. That is, even if the additive noise processes are arbitrary stochastic processes and the initial state is an arbitrary stochastic variable, the certainty equivalence property holds under the conditions (I), (II) and (III). Next, filtering problems are discussed. In the case where the noise process of dynamics and the initial state are nongaussian, the Kalman-Bucy like filter is derived. Problems are formulated and discussed with the continuous time model. To clarify the sufficiency of the conditions (I), (II) and (III), the discrete time case is also discussed.抄録全体を表示PDF形式でダウンロード (955K) -
衣笠 晨策1976 年 12 巻 1 号 p. 96-101
発行日: 1976/02/28
公開日: 2009/03/27
ジャーナル フリーThe general technics to analyze the plastic working of the pressure sensing elements were already reported in previous transactions. In this paper the processes and the results of the analysis of the metal diaphragm based on the general technics are reported.
There are several conditions in the press working of the diaphragm. The analysis is easy under the conditions that one or both of the inside or outside edges of the diaphragm are free. The circular elongation εc(r) and the radial elongation εr(r) are connected with the inclination angle θ of the wave, the indideto-outside radiuses ratio ρ. The maximum calculated elongation εcm is shown in the graphs.
The problem about the buckling on the outside of the diaphragm is complicated under the condition that the inside edge is fixed and the outside edge is free. The analysis concerning this problem is described.
The method of transformation by means of centrifugal force is adopted in order to solve the non-linear problem under the condition that the both inside and outside edges are fixed. The result of the analysis is clear and elegant.The experiments to measure the elongation εc were done. The result of the experiments shows the mixture of the conditions that the edges are fixed and free.抄録全体を表示PDF形式でダウンロード (749K) -
河野 通夫, 杉浦 一郎1976 年 12 巻 1 号 p. 102-106
発行日: 1976/02/28
公開日: 2009/03/27
ジャーナル フリーA non-interacting control of an r-input m-output linear discrete time system over the lattice group via state variable feedback is considered. In a linear discrete time system over the lattice group, components of the input vector, the output vector and the state vector take only integral values.
This paper defines the sequential reproducibility which corresponds to the functional reproducibility in a linear system over the vector space, and gives a necessary and sufficient condition for the sequential reproducibility in a single-variable linear discrete time system over the lattice group.
This paper defines the non-interacting control system which is sequentially reproducible. A sufficient condition for non-interaction is derived In particular, when r=m, a necessary and sufficient condition for non-interaction is obtained.
General decoupling control laws are given in order to improve responses. An example is given illustrating the result.抄録全体を表示PDF形式でダウンロード (599K) -
林 靖享1976 年 12 巻 1 号 p. 107-112
発行日: 1976/02/28
公開日: 2009/03/27
ジャーナル フリーThis report describes a computing method using the gradient method, in an optimum control of the system including the stop mechanism. The system equations are written in a vector form as x=f(x, u). Only one equation among them, xl=fl(xl, u), includes u in the right hand side, and xl is limited because the movement in the xl-coordinate is limited by means of the stop mechanism.
In the nominal control, a small deviation of u is given in the admissible range of xl, and an imaginary small deviation of xl is given in the stop range of xl. As a result, the small deviation of each state variable is written by using the transition matrix. The deviation of a criterion function, which is contributed by a small deviation of u or xl, is obtained as a function of the small deviation of state variable at the end point. That transition matrix of this equation is eliminated by introducing a costate vector, and the gradient function of each range is written with the costate vector. According to the value of the gradient function, small correction of nominal control is applied repeatedly in order to bring the criterion function toward the minimum value.
As an application, it is shown in the computation to countermeasure a vibration which is caused at the time when a hydraulic actuator controlled with a valve turns its direction.
Partial modifications may be possibly applied to on-off control systems.抄録全体を表示PDF形式でダウンロード (636K) -
室井 伸一, 井上 昭, 得丸 英勝1976 年 12 巻 1 号 p. 113-118
発行日: 1976/02/28
公開日: 2009/03/27
ジャーナル フリーThis paper considers the optimal design problem of a discrete-time time-invariant linear compensator for a linear control system with inaccessible states. It is assumed that the control system is also discrete-time and timeinvariant and that its initial states are unknown, but only their mean and covariance are known. The performance index for this problem is a quadratic cost averaged over initial state values.
To obtain the optimal compensator, this problem (Problem 1) is changed to two problems: problem 2 of designing an optimal compensator for a linear stochastic control system minimzing a performance index in steady-state, and Problem 3 of minimizing an average performance index per unit of time. Problem 3 is solved at first.
A separation theorem is shown. The optimal compensator is optimal in all linear control laws, and is constructed by using the optimal state-estimate observer and the feedback gain which is optimal in the complete state feedback.抄録全体を表示PDF形式でダウンロード (834K)
- |<
- <
- 1
- >
- >|