All issues
Volume 30, Issue 6
Displaying 1-20 of 20 articles from this issue
- |<
- <
- 1
- >
- >|
-
Satoshi HONDA, Yutaka TOMITA, Shingo NAGASHIMA1994 Volume 30 Issue 6 Pages 603-608
Published: June 30, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSAlternating magnetic field location technology has been used to establish the location and the depth of a buried pipe without digging; AC voltage applied to the conductive pipe makes current flow out both ways and leaks away to the surrounding soil. These currents along the pipe create a magnetic field of a cylindrical shape, which can be detected above the ground. This paper analyzes the performance of the previously developed locator which drives currents indirectly through a pair of driving coils above the ground. These currents and magnetic field are analyzed under the assumptions of homogeneous soil of pure resistance and quasi-steady field. Evaluated currents agree well with experimental data.View full abstractDownload PDF (1033K) -
Masahiro YOKOMICHI, Masasuke SHIMA1994 Volume 30 Issue 6 Pages 609-616
Published: June 30, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, a normal form for nonlinear affine Hamiltonian control system is proposed. Its Hamiltonian is expressed as a polynomial of canonical momenta. Respondek proposed a similar normal form for usual affine nonlinear control system and gave a condition for the classification of nonlinear control systems by means of this normal form.
However, Hamiltonian control systems have several special structures and some attentions must be paid in appling above results to Hamiltonian control systems. We propose some normal forms corresponding to the dimension of the state space and the number of inputs. For each case, we give the conditions about state equivalence and weak feedback equivalence to the normal forms.View full abstractDownload PDF (1182K) -
Liu QIAO, Mitsuo SATO, Hiroshi TAKEDA1994 Volume 30 Issue 6 Pages 617-624
Published: June 30, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper we investigate methods and applications of pattern recognition. It focuses the problem of recognizing driving environment of a vehicle by using information obtained from some sensors of the vehicle, and we present an environmental recognition algorithm in which the recognition is dealt with fuzzy logic.
Previously, we presented a recognition algorithm based on fuzzy reasoning. The algorithm can not be applied to meet the demands of nonstandard drivers and changes of vehicle properties, because the membership functions are fixed once they are constructed. To cover such weakness we present an adaptive recognition algorithm with adaptive change of membership function.
A result of computer simulation of the recognition model by using actual driving samples taken from a number of drivers under five driving environments is presented. Seven feature extraction variables generated from thirteen sensor measurements are considered to be the recognition feature. The recognition algorithm is constituted with self-adaptive as well as non-supervised method as there is no extra source of knowledge for correcting the decision taken by the classifier. Here the supervisor is made by using α-cut of membership function of the representative class. We show the efficiency of the self-supervised learning system for recognition of driving environment.View full abstractDownload PDF (1361K) -
Keishi HASHIMOTO, Fumio MATSUMURA, Masayuki FUJITA1994 Volume 30 Issue 6 Pages 625-632
Published: June 30, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper describes the experimental evaluation of LQ differential game theoretic interpretation of H∞ control for a magnetic suspension system with double flexible beams. It simplifies both an elastic rotor and body in a magnetic bearing system and emphasizes its vibrational problems. At first, the magnetic suspension system is introduced. Secondly, its linearized model with disturbance terms as unmodeled uncertainties is formulated in a state-space form. Then the generalized plant is constructed from LQ differential game theoretic point of view, which involves weighting matrices for state variables, control inputs and frequency weighting functions for uncertainty models. And we design and analyze the control systems by changing design parameters which correspond to weighting factors in the time domain. Finally, several experiments are carried out for comparison of transient responses with regard to control performance. We verify that transient responses can be improved by adding weight on the parameters in the generalized plant constructed in this paper.View full abstractDownload PDF (1586K) -
Helio TINONE, Nobuharu AOSHIMA1994 Volume 30 Issue 6 Pages 633-640
Published: June 30, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, the implementation of a distributed control system for robotic manipulators with transputers is discussed. Real-time control applications employing the parallel processing model supported by transputers are viewed with suspicion for fear of deadlock and introduction of non-determinism. We review some of the techniques used in real-time concurrent processing applications and propose a system particularly suited for robot control, which is experimentally tested in the control of a Direct Drive manipulator.View full abstractDownload PDF (1242K) -
Satoru GOTO, Masatoshi NAKAMURA, Nobuhiro KYURA1994 Volume 30 Issue 6 Pages 641-643
Published: June 30, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSTo achieve accurate position control of mechanical servo systems, a method for modifying the objective speed data was proposed. The mechanical servo systems, with the aid of the developed method, is capable in drawing the path with an assigned accuracy without any complex operations.View full abstractDownload PDF (466K) -
ShinIchi AIHARA1994 Volume 30 Issue 6 Pages 644-650
Published: June 30, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThe purpose of this paper is to study the identification problem of a spatially varying discontinuous parameter in stochastic diffusion equation.
In order to show the consistency property of M. L. E. for a discontinuous diffusion coefficient, we use the method of sieves, i. e., first the admissible class of unknown parameters is projected into a finite-dimensional space and next the convergence of the derived finite-dimensional M. L. E. to the infinite-dimensional M. L. E. is justified under some conditions.View full abstractDownload PDF (958K) -
Improvement of Interference in Pole Control and Identification of System with Flat DelayIchi TAKUMI, Masayasu HATA1994 Volume 30 Issue 6 Pages 651-659
Published: June 30, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper proposes a new method of a pole controlled IIR adaptive digital filter (ADF) with a preceded FIR section, which consists of parallel connected second order IIR filters controlled by each inherent identification error. The FIR part works to identify a flat delay and responses resulting from poles near the origin of z-plane. On the other hand, the IIR part identifies poles near the unit circle on z-plane by the proposed method. The FIR part and the IIR part cooperate mutually independently to identify the unknown system effectively. Simulation results show that the proposed IIR ADF (of 28th order) has an excellent controllability of poles and zeros.View full abstractDownload PDF (1644K) -
Ryo WATANABE, Kenko UCHIDA, Etsujiro SHIMEMURA, Masayuki FUJITA1994 Volume 30 Issue 6 Pages 660-668
Published: June 30, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn the control of the actual plant, there exits a difference between output of compensator and control input momentarily. This is caused by mechanical constraint on the actuator like saturation and rate limit, and constraint on the control input to protect the plant in the aircraft engine controls etc. Difference between output of compensator with integral element and control input causes windup phenomenon and gives adverse effects to the control performance. One of the techniques to prevent the above phenomenon is anti-windup and bumpless transfer technique.
In this paper, we derive sufficient condition of robust stability for the system consisting of the plant with linear time invariant multiplicative uncertainty, compensator, constraint on the control input, and anti-windup and bumpless transfer block. Then, we propose a new algorism of anti-windup and bumpless transfer block synthesis which directly uses this stability condition. In numerical example, we design the anti-windup and bumpless transfer block to a given system, and evaluate its performance by computer simulation.View full abstractDownload PDF (1575K) -
Satoru GOTO, Masatoshi NAKAMURA, Seiji OKA, Nobuhiro KYURA1994 Volume 30 Issue 6 Pages 669-676
Published: June 30, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA method of synchronous position control for multi servo systems was proposed. The synchronous position control is required for various works such as tap process of screwing and positioning of multi drive crane. To achive the synchronous position control, the position output of the main servo system was fed back into the position input of the slave servo system. If the position output of the main servo system was directly used to the position input of the slave servo system, the position output of the slave servo system delayed from the position output of the main system because of the dynamics of the slave servo system. Hence, we modify the input of the slave servo system from the position output of the main servo system by using the inverse dynamics of the slave servo system.
The effectiveness of the proposed method was assured from the experimental results of the synchronous position control of an X-Y table which was one of the mechanical multi servo systems of 2 degrees of freedom.View full abstractDownload PDF (1629K) -
Toshiro NORITSUGU, Masahiro TAKAIWA1994 Volume 30 Issue 6 Pages 677-684
Published: June 30, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSPneumatic servo system is expected to be applied for the contact tasks because of its function of impact force absorption and the possibility of minute force control due to air compressibility. An impedance control, unified approach to control the dynamic relation between the external force and displacement, has been confirmed to be effective for the contact tasks. However, it requires a force sensor, which makes the structure of control system intricate.
We propose an impedance control method without force sensor and applied it to a pneumatic servo system comprising a bellophram type cylinder and PCM control valves. A control system using a disturbance observer which estimate the external force is constructed in an internal position control loop.
There exists a dead time element in most of the pneumatic servo systems, so a Smith's method is introduced to increase the stability margin. Then an estimation function, which is a realization indicator of the setting impedance, is proposed. By using this estimation function, the stability of the control system also can be inspected so the control parameters can be chosen rationally with considering the stability of the control system simultaneously. Through some experiments, the proposed control method is confirmed to be available for the pneumatic servo system.View full abstractDownload PDF (1595K) -
Haruo HASEGAWA1994 Volume 30 Issue 6 Pages 685-693
Published: June 30, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper describes a method of verifying an SDL specification of a switching system by utilizing the algebraic properties of Petri Nets (PN). In order to verify the SDL specification, it is converted into PN, which have been an excellent tool for modeling an asynchronous and parallel system. SDL is based on a finite state machine (FSM) model and FSM can be easily converted into PN. However, SDL has other functions than FSM such as creation and deletion of a process instance, or realization of a synchronous communication between processes by adding a ‘wait’ state. The SDL specification is converted into PN considering these functions. In this paper, Colored Petri Nets (CPN) are used because they have more powerful modeling ability than conventional Petri Nets. Two kinds of invariants, S-invariant and T-invariant, can be calculated from the incidence matrix of the created Petri Nets which means connections between transitions and places. Then, boundness of the specification is verified by S-invariants, and liveness and deadlock-freeness by T-invariants. It is very difficult to get the invariants from the incidence matrix in CPN, because the elements of the matrix are not integers. However, the proposed method can utilize the invariants got from the incidence matrix of conventional PN, which are created from CPN. Besides, in a case that engineers design the SDL specification incrementally, they can ensure the correctness which the specification had before they add new services to it, by checking the invariants. Finally, the method is successfully applied into design of a PBX prototyping system.View full abstractDownload PDF (1649K) -
Atsushi OHTA, Tomiji HISAMURA1994 Volume 30 Issue 6 Pages 694-701
Published: June 30, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSLiveness is one of the most important properties in Petri net, which is a powerful tool for modeling of discrete event systems.
As an extended class of Petri nets, timed Petri nets with finite firing duration of transitions have a wide range of applications, such as scheduling problems in FMS, parallel computing systems and so on. The earliest firing rule for timed Petri net model offers an easy and effective way for solving sub-optimization problems. On the other hand, it is often the case in practical problems that some tasks share processors with other tasks. Such processors can be modeled by ‘resource places’ in Petri net.
In this paper, we treat liveness problems of the timed Petri nets which have shared resource places and obey the earliest firing rule. In general, liveness of a usual Petri net is neither necessary nor sufficient condition for liveness of the Petri net with shared resource places under the earliest firing rule. The relations between above two cases are clarified for some subclasses.
First, considering timed Petri nets with uniform firing durations, we show that if the net has a POC (partially ordered condition) structure and if the net is live, the net with arbitrary resource places is live under the earliest firing rule. The converse is not always true, but if the underlying net is SMA net, the two notions of liveness are equivalent.
For timed Petri nets with arbitrary firing durations, providing some additional conditions on resource places and the underlying net, we show that liveness is preserved under the earliest firing rule (1) if addition of resource places preserves POC structure, or (2) if resource places are connected to “fair” transitions. Some typical models in scheduling problems satisfy these conditions.View full abstractDownload PDF (1335K) -
Katsuaki ONOGI, Manabu TAKEMURA, Yoshiyuki NISHIMURA1994 Volume 30 Issue 6 Pages 702-711
Published: June 30, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIt becomes difficult to obtain an optimal combination or permutation of objects as the number of objects increases. In such cases good suboptimal solutions are valuable in practice. This paper proposes a heuristic method of obtaining suboptimal solutions of the above combinatorial optimization problems. Its basic idea is to reduce an original problem to a small one by regarding similar objects as the same. The inherent knowledge about the original problem is used for this partially identifying approximation. Suboptimal solutions of the original problem are constructed from the exact optimal solution of the reduced problem. The paper then applies the method to flow-shop scheduling problems, and discusses the effects of problem structures on the precision of obtained suboptimal solutions and the computation amount. The method seems to be easily applied to the actual large problems, because its principle is simple to understand.View full abstractDownload PDF (2888K) -
Determination of Optimal Camera and Light-Source PositionsShin-ichiro GOMI, Koichiro DEGUCHI1994 Volume 30 Issue 6 Pages 712-719
Published: June 30, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper proposes a hierarchical design method of camera and light-source positioning for model based vision systems. In this method, first, the geodesic dome is constructed for given objects to place initial candidate positions of camera and light-source. Then, the candidate positions are selected out step by step using criterions of viewing angle, occlusion by other objects, specular reflection, and edge contrast. Then the final optimal positionings are obtained. We show that our hierarchical and step by step strategy achieves great reduction of computational time over the conventional direct configuration methods to obtain good quality of images with high contrasts of object edges.View full abstractDownload PDF (3871K) -
Masahiro INUIGUCHI, Tadashi SHIRAI, Masatoshi SAKAWA1994 Volume 30 Issue 6 Pages 720-728
Published: June 30, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, the concept of stochastic dominance is introduced to Dempster-Shafer theory of evidence and several kinds of evidential dominance are proposed. In Dempster-Shafer theory of evidence, the expected utility is obtained as an interval. The seven kinds of inequalities between intervals are defined. The relationships among these seven inequalities are investigated. Using these inequalities, seven kinds of evidential dominance are defined. A necessary and sufficient condition for each evidential dominance is discussed. The usefulness of each evidential dominance is examined by a numerical simulation.View full abstractDownload PDF (1363K) -
Tatsuya KUNIOKA, Nobuo SHIMAZU1994 Volume 30 Issue 6 Pages 729-731
Published: June 30, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA capacitance method was developed to measure chucking and release time in an electrostatic chuck. This method is more practical than the conventional method, and we can obtain both chucking and release time in one measurement.View full abstractDownload PDF (1289K) -
Akira INOUE, Valeri T. KROUMOV, Shiro MASUDA1994 Volume 30 Issue 6 Pages 732-734
Published: June 30, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA general design scheme for model reference adaptive control system (MRACS) is proposed. The scheme is obtained by using the most general exact model matching (EMM) control law, which is based on two-parameter compensation scheme using factorization approach in the algebraic ring of proper stable rational functions. The scheme includes several exiting MRACS design methods which have been proposed by other authors. Also the general scheme gives a new design method, by which the poles of the closed loop system of MRACS can be placed arbitrarily.View full abstractDownload PDF (414K) -
Zhibin SHU, Keiji WATANABE, Kou YAMADA1994 Volume 30 Issue 6 Pages 735-737
Published: June 30, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper consider the internal model control of time-delay systems. A design method for robust stabilization is presented.View full abstractDownload PDF (292K) -
Yasumasa FUJISAKI, Kazuhiro TOGAWA, Kazumasa HIRAI1994 Volume 30 Issue 6 Pages 738-740
Published: June 30, 1994
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSA new sliding mode controller is proposed for multi-input discrete time systems. The controller consists of a linear state feedback and a switching feedback. A class of stabilizing switching gains is derived. Another class of those without chattering is also derived.View full abstractDownload PDF (342K)
- |<
- <
- 1
- >
- >|