All issues
Volume 24, Issue 11
Displaying 1-20 of 20 articles from this issue
- |<
- <
- 1
- >
- >|
-
Wataru MITSUHASHI, Kazuhiko OKA, Hiro YAMASAKI1988 Volume 24 Issue 11 Pages 1111-1117
Published: November 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA measurement system of dynamic image motions is developed by the use of spatial filters which are implemented with electronic circuits and video signals. The conventional spatial filter method, developed to measure the translating velocity of object motions, is not suitable to measure object displacement precisely. A pair of electronic spatial filters, of which phase are orthogonal to each other, forms a Hilbert transform pair and, therefore, allows object displacement to be measured as well as velocity. In addition, a complex logarithmic mapping of dynamic images will transform scaling and rotating image motions to translating ones independent of one another. These motion parameters, therefore, can be measured by utilizing the spatial filter method on the transformed domain. Originally developed for translation velocity measurement applications, the spatial filter method, implemented with electronic devices, now has capabilities to measure the global parameters of object motions.View full abstractDownload PDF (1140K) -
Shigeyasu KAWAJI1988 Volume 24 Issue 11 Pages 1118-1123
Published: November 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSLinear descriptor system of the form Ex=Ax+Bu, y=Cx with E singular is a mixture of dynamic and static equations and can be served as a useful method in analysis of many real and practical situations.
This paper deals with the problem of designing an observer for descriptor systems. The traditional approach to such an observer is to separate the dynamic equations from the algebraic equations, and to estimate the unobservable parts from the outputs. Contrary to this, the observer is presupposed to have the same structure as Luenberger observer. At first, the fundamental equations that the plant and the observer must satisfy are derived. Secondly, based on these equations, sufficient conditions are given for the existence of the observer and methods of designing an identity observer and low-order observers are presented by utilizing the generalized matrix inverse.View full abstractDownload PDF (724K) -
Tomohiro KUBO, Etsujiro SHIMEMURA, Kenko UCHIDA1988 Volume 24 Issue 11 Pages 1124-1128
Published: November 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, a semigroup model of a flexible structure with an actuator is presented. The flexible structure is modeled as a cantilever beam and the actuator is modeled as a lumped-parameter subsystem. First, a set of elementary equations which describe the dynamics of the plant is shown. Then, an appropriate Hilbert space is chosen as the state space, and an operator is defined in the space. It is shown that i) the domain of the operator is dense in the state space, ii) the operator is a closed operator, iii) the operator is an infinitesimal generator and thus a differential equation with the generator is a semigroup model.View full abstractDownload PDF (584K) -
Kazuo AIDA, Keiji YOKOSE, Sadatoshi HORI, Toshiyuki KITAMORI1988 Volume 24 Issue 11 Pages 1129-1136
Published: November 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis study is concerned with a design method of digital preview controllers for the robust servo-mechanism with predetermined feedback controllers.
It is assumed that a reference input or a disturbance is output of linear free system and the future values of them can be utilized upto ‘l’ time units in the future.
We design the preview control so as to make a quadratic cost criterion defined in the infinite period minimize under the conditions that the overall system is, initially at equilibrium.
This method is, first, derived for state feedback control systems, then extended for systems with D-action such as I-PD control by adding some another states.
Simulation studies show that the preview action improves the performance of feedback control systems.View full abstractDownload PDF (867K) -
Katsuji UOSAKI, Toshiharu HATANAKA1988 Volume 24 Issue 11 Pages 1137-1142
Published: November 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSExperiment design for dynamic system identification has attracted considerable attention to obtain the maximal information from the observed input/output data. Most studies in this area have been devoted to accurate parameter estimation within a specified model structure. However, the problem of primary importance in system identification might be to determine the model structure itself. From this point of view, the optimal input design problem is discussed in this paper for discriminating two rival autoregressive models efficiently. An optimal input is derived, which maximizes the time increment of the Kullback's discrimination information measure to make the distance between two models as big as possible. The applicability of the proposed input for efficient autoregressive model discrimination is exemplified by simulation studies.View full abstractDownload PDF (694K) -
Nobuaki KOBAYASHI, Takayoshi NAKAMIZO, Sigeru UCHIKADO, Kimio KANAI1988 Volume 24 Issue 11 Pages 1143-1149
Published: November 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper describes a new approach to the exact model matching problem for multivariable linear time invariant systems.
The structural exact model matching problem is formulated to structural systems, and the solvability condition is derived by using the graph-representation of systems. The structure of the interactor which plays an important role in a design of model matching systems and model reference adaptive control sytems is also investigated. It is shown that a simple and intuitive procedure for determining the structure of the interactor can be obtained by the help of graph-representation of systems. An example is given to illustlate these algorithms.View full abstractDownload PDF (986K) -
Tetsuzo TANINO, Naohiko TAKAHASHI1988 Volume 24 Issue 11 Pages 1150-1157
Published: November 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe concept of fixed modes plays an important role in stabilization and pole assignment for linear time-invariant systems. Several characterizations of fixed modes have been obtained.
In this paper, a concept of degrees of fixed modes is introduced for linear time-invariant control systems with constrained control structures, typically including decentralized control systems. The degree of a fixed mode corresponds to its geometric multiplicity, and is defined by the minimal dimension of all eigen spaces of feedback system matrices for the fixed mode. It coincides with minimal multiplicity of the fixed mode as the transmission zero for regular subsystems. Moreover, in decentralized control systems, the degree has an equivalent expression related to the algebraic characterization of fixed modes by partitions of stations.
Moreover, a concept of minimal effective subsystems corresponding to each fixed mode is introduced by considering the Jordan canonical form of the linear system. It is proved that at least one minimal effective subsystem exists for every fixed mode. This fact is important in applications such as elimination of fixed modes.View full abstractDownload PDF (964K) -
Yasuhiko MUTOH, Kunihiko ICHIKAWA1988 Volume 24 Issue 11 Pages 1158-1164
Published: November 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe continuous model reference adaptive control system with an amplitude constrained input signal is discussed in this paper. The plant to be controlled is assumed to have stable poles and zeros. The adaptive algorithm stated here is based on what Monopoli presented. The purpose of this paper is to analize the behavior of such an adaptive system when t tends to infinity. The sufficient condition for the adaptive input to become free from the amplitude constraint and to converge to the desired input signal as t tends to infinity is considered. The modified continuous adaptive algorithm is also presented to improve the trangent adaptive response.View full abstractDownload PDF (840K) -
Constraints on Sensitivity Characteristics in Multi-Input Multi-Output Discrete-Time Control SystemsShinji HARA, Hak-kyung SUNG1988 Volume 24 Issue 11 Pages 1165-1172
Published: November 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper is concerned with the limitations on the sensitivity characteristics for discretetime MIMO control systems. By the Poisson inequality for the subharmonic functions, some integral-type constraints and the lower bounds of the weighted H∞-norm imposed by unstable poles, unstable zeros, and their directions of the open-loop system or the given plant are developed by a factorization approach. These constraints and bounds are also characterized by the state-space forms, which are closely related to a special type of the algebraic Riccati equation.View full abstractDownload PDF (1085K) -
Masatoshi NAKAMURA, Tsukasa ABE1988 Volume 24 Issue 11 Pages 1173-1180
Published: November 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper describes a controller design for a water temperature system of two tanks, aiming at an accurate control of experiments. Points of discussion consist of the following three items: characterization of detectors and actuators, a control method in the transient period, and a control method in the steady state period. We concentrated mainly on the control method in the steady state period. Oscillation phenomena of the tanked water temperature around a desired value were found to be caused by a drive delay of heaters in the water temperature system. We then proposed a compensation control method for the oscillation. The features of the proposed compensation method are as follows: the compensating control signals were calculated by using a prediction of a real time simulation of the actual system, and errors of a model mismatch were recovered in real time at each step. In an experimental result, the oscillation around a desired value was greatly diminished by using the compensation method. Next, we discussed a control method in the transient period. We showed that switching use of a maximum output control and a pole placement control brought a quick response of the water temperature to a desired value. Finally, we argued that the characteristics of detectors and actuators should be determined under a similar condition of the actual experiment. Those characteristic curves brought an accurate control performance. The results obtained in this study will be used effectively in general problems of fluid temperature control.View full abstractDownload PDF (1198K) -
Masakazu HARADA, Osamu OYAMA, Satoki FUJII1988 Volume 24 Issue 11 Pages 1181-1186
Published: November 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA precise force control system is designed by using a pneumatic piston-cylinder, membrane type electro-pneumatic converter and microprocessor, and a favorable result is obtained. The pneumatic piston-cylinder has relatively large frictional force and its irregularity even if the stroke is short. Then a closed loop control circuit with microprocessor so called software servo-system are employed to eliminate the affection. To obtain a higher accuracy in operating force, the effect of PD and PI controls are examined with a simple feedback system.
A favorable result is obtained by changing the sensitivity of integral action with the change of error. The level of reproducibility in controlled force is obtainable better than 0.095% of desired value 20.3N in standard deviation. The settling time can be reduced by changing control mode. At the start, derivative action is added to proportional action, but PI combination is supposed to be best for the settling and to reduce the difference between desired value and set value. The stability of the control system is assured by Hurwitz's criterion and experiment.View full abstractDownload PDF (686K) -
Toshiro NORITSUGU, Tsutomu WADA, Masami YANOSAKA1988 Volume 24 Issue 11 Pages 1187-1194
Published: November 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA pneumatic actuator tends to be applied to the more rigorous motion control in addition to conventional simple works, and some PID or optimal control pneumatic servo systems have already been proposed. To design these control systems, the mathematical model describing system dynamics must be known beforehand. However, the exact modeling and identification of system parameters are difficult due to various inherent nonlinearities in a pneumatic system. Moreover, the system parameters depend on the operating conditions so strongly that the satisfactory response can not always be obtained with the constant gain controller.
In this study, to cope with these problems an adaptive control scheme, which can adjust its own gains by using estimated system parameters to realize the desired response, is employed in the electro-pneumatic servo system comprising a pneumatic cylinder and electro-pneumatic proportional control valves. The control law is designed with the adaptive pole-placement method applicable to also a non-minimum phase system often encountered in a discrete-time control using a digital computer. The air low stiffness makes a pneumatic system fairly sensitive to frictional or external load force, resulting in a decreased position control accuracy. For dealing with these forces, the conventional adaptive poleplacement control scheme is expanded to consider them as an unknown deterministic disturbance.
From various experimental examinations, the following is verified. The adaptive control can make the pneumatic servo system enough robust to achieve the identical desired response independently of the operating conditions such as a mass load, an operating pressure, a desired position and so on. Further, the effect of considering the disturbance remarkably appears at being subject to the external force. This expanded control method can be satisfied with also the positioning accuracy.View full abstractDownload PDF (1042K) -
Toshimitsu USHIO, Ryuichi MATSUMOTO1988 Volume 24 Issue 11 Pages 1195-1200
Published: November 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSPetri nets with external input places are useful tools for analysing control problems of discrete-event systems. If control synthesis problems can be represented by control-invariant predicates on the reachable set, then there exist control-alternatives, which are static state-feedbacks for the control problems. In this paper, the maximal control-alternative is called the maximally permissive feedback, which does not always exist uniquely.
This paper shows that the weak interaction of the predicate is the necessary and sufficient condition for the uniqueness of the maximally permissive feedback. And modular feedback is considered in the case that control synthesis problems are represented by conjunction and/or disjunction of predicates. Specially, it is shown that the maximally permissive feedback for conjunction of predicates is equal to the modular one at markings satisfying the conjunction if each predicate is weakly interactive.View full abstractDownload PDF (818K) -
Tetsuo SHIOTSUKI, Shoichi NARAHASHI, Shigeyasu KAWAJI1988 Volume 24 Issue 11 Pages 1201-1203
Published: November 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper is concerned with the problem of constructing an observer for linear descriptor systems. The usual realizability conditions of the observer are relaxed by using the elimination of the purely static parts of the system. It is shown that the observer can be realized if the system is observable in the sense of Verghese.View full abstractDownload PDF (289K) -
Takao FUJII, Manabu MIYAOKU1988 Volume 24 Issue 11 Pages 1204-1206
Published: November 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA new design method of optimal servo-systems, called “ILQ design method” is based on both a necessary and a sufficient condition for optimality of feedback control laws. In this paper we relax the latter one to obtain a new sufficient condition of more practical usefulness.View full abstractDownload PDF (389K) -
Tsutomu MITA1988 Volume 24 Issue 11 Pages 1207-1209
Published: November 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper shows that the LQ digital optimal control solution implied in H∞ optimal control solution if the L2 norm of the performance index of LQ control is exchanged by H∞ norm.View full abstractDownload PDF (364K) -
Yoshihiro YAMAMOTO1988 Volume 24 Issue 11 Pages 1210-1212
Published: November 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper presents a model matching approach which includes the freedom to improve the disturbance characteristic.
The result can be extended to use it in adaptive control systems synthesis.View full abstractDownload PDF (393K) -
Hideo ISHIHARA, Seiji KOBAYASHI1988 Volume 24 Issue 11 Pages 1213-1215
Published: November 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSAn adaptive control scheme for nonlinear systems with unknown parameters is proposed. Parameters of an approximate system model are estimated by the least square method, and unknown parameters of the original nonlinear system are estimated from these through the algorithm proposed. Basing on the results, the control input is determined.View full abstractDownload PDF (377K) -
Yoshiaki KAWAMURA1988 Volume 24 Issue 11 Pages 1216-1218
Published: November 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA basic iterative algorithm is proposed which directly constructs the optimal regulator from input-output data without requiring system equations. It is clarified that the algorithm is closely related to Newton-Raphson type algorithms which use state equations instead of data.View full abstractDownload PDF (376K) -
Yasuhiko KAKU, Hidehiko NUMASATO, Tsutomu MITA1988 Volume 24 Issue 11 Pages 1219-1221
Published: November 30, 1988
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, to reduct a time varing torque disturbance of the motor control system, we design the augment system of regulator and obsever using stable digital inverse system that we proposed. Experiment results show that our design method is better than conventional one.View full abstractDownload PDF (1597K)
- |<
- <
- 1
- >
- >|