Sensors used in alarm security services are reviewed. Security will be maintained when the defensive power is superior to the offensive power. In Japan, crimes such as smash-and-grab, lock picking, cash dispenser destruction & theft, and others have gained prevalence over the past few years. The importance of a sensing system which can be promptly modified for the latest modus operandi is addressed.

View full abstract

[in Japanese]

View full abstract

This paper presents the background and potential importance of the interplay between biology and control, specially from the viewpoint of brain science. Motor control is the main field of the interplay, where control approach has been challenged by the so-called self-organization approach. The difference and common features of the two approaches are discussed and the fundamental limitations of the control approach to the creation of human behavior are discussed. Some future research directions of control approach to human behaviors have been suggested.

View full abstract

In ubiquitous computing, user's location is one of the most important information. We have proposed ALTAIR, a new location system for indoors, using image processing in IR(InfraRed) region and wireless network. The system can automatically detect and track the locations of the mobile PC users so that each user can derive the information based on the place where he/she is in a building. In this paper, we introduce location logging system based on ALTAIR. The system automatically makes log of users' location, and helps reviewing work.

View full abstract

In this paper, we propose a new sensor which detects intruders who breaks in the door, such as lock picking by hand. This new sensor uses dual frequency Doppler Ladar, the sensor is installed over the door, which can measure the real distance between the sensor and the human body. We examined that this new sensor only detects the human body: which has a small continuing motion when lock picking, and it would not be affected by metal handle or door itself, which do not have a motion.

View full abstract

In Japan, as society continues to age and stress levels increase, the number of people suffering from sleep disorders is on the rise. It is said that the rate of people for the sound of a snore to be loud is about 20%. This indicates that many people have no solution to their sleep disorders. Against this backdrop, demand is growing for products that can be easily used at home to manage health.
Then, in order that we might detect a snore, we developed the capacity type sensor and examined the method for a high precise detection of a snore using the microphone and the capacity type sensor.

View full abstract

This paper studies a robust performance analysis method of sampled-data systems against parameter uncertainties.We derive an analysis method to compute the admissible range of the uncertainties for scalar parameters in the continuous-time plant,under which the given sampled-data system remains internally stable and retains robust H_∞-performance.

View full abstract

In this paper, we first extend the Nyquist stability criterion for sampled-data systems based on the 2-regularized determinant pertinent to Hilbert-Schmidt operators in such a way that the continuous-time generalized plant is allowed to have a direct feedthrough term from the disturbance input to the controlled output. By means of this Nyquist stability criterion, we derive robust internal stability theorems of sampled-data systems such as the small gain theorem and the passivity theorem.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

In recent years, metaheuristic algorithms for resource constrained project scheduling problems(RCPSPs) which can obtain satisficing solutions with small computational cost have been studied. In this paper, we propose CATS(Critical Activity based Tabu Search) and SFTS(Space Filling Tabu Search) as new tabu search algorithms for the RCPSP. CATS focuses on critical or nearly critical activities, and SFTS makes use of blank spaces of schedules.We confirm efficiency of our algorithms through computational experiments.

View full abstract

[in Japanese]

View full abstract

In this paper, the optimization algorithm which utilezes the learning performance of the variable hierarchical structure learning automata is constructed for finding the global minimum of unknown objective functions. And the usefuluness of our algorithm is demonstrated by numerical simulations.

View full abstract

Many problems which are useful for engineering are usually formulated as function optimization problems. In this paper, we consider function optimization problem as combinatorial optimization problem. Because we sometimes deal with integer number in the design of industrial products. For solving such an optimization problem, local search methods have been proposed. But most of them search all parameters simultaneously. As the result, the search space size increases exponentially with the number of parameters. Therefore we propose a divisional local search method which is named ``Free&Freeze method'' for such a large scale optimization problem.

View full abstract

[in Japanese]

View full abstract

In this work, an improvement is made for incremental principal component analysis (IPCA) in which the addition of a new eigen-axis is determined based on the accumulation ratio for a currently constructed feature space. In incremental learning situations, the accumulation ratio must be modified every time a new sample is given. Therefore, in order to use this ratio as a criterion for the addition of a new feature in IPCA, we need to develop a one-pass update algorithm for the accumulation ratio. To obtain its approximation without using all the past samples, we present an improved algorithm of IPCA. To see if correct feature construction is carried out by the new IPCA algorithm, the recognition performance is evaluated for some standard datasets when k-NN classifier is adopted as a classifier.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

This paper proposes the method over practical problem of Astolfi's controller which is one of discontinuous control law for nonholonomic system. By reinterpreting the Astolfi's controller as time-state control form proposed by Sampei, a problem of input constraints is solved.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

In this research, we discuss signature of individual authentication. While, the individual authentication with signature is comfort and effective for ordinary authentication system.
Especially, discuss pen pressure. From previous research, the experimental results with various data from several persons show excellent.
However, in practical use, there is some miss-recognition for counterfeit writing, which another person imitatesone's signature.
So, we apply neural network with radial bases function(RBF).
In this paper, RBF is proposed to improve rejection capabilities of the system on the premise of ensuring the effectively on known patterns recognition.

View full abstract

[in Japanese]

View full abstract

Perceptual grouping is a process involved in chunking of visual information and in image segmentation, constituting an important aspect of visual processing. Functional magnetic resonance imaging (fMRI) was used to investigate brain activities associated with visual grouping. Twelve healthy subjects (21-31 years) participated in the experiments. Subjects were asked to press a button when he could identify the location of a target quadrant in a grouping task. Contrasting the grouping task against a control task with no target quadrant revealed increased hemodynamic activities in dorsal visual pathway on primary and higher visual areas, parietal association cortex and superior frontal sulcus.

View full abstract

In this paper, we present identification algorithms for parametric canonical-formed state space models and also for determining the order of models, based on maximum-likelihood and Akaike's Information Criteria (AIC). For obtaining maximum-likelihood estimates of model parameters, we apply the Expectation-Maximization (EM) algorithm which is an iterative method such that choice of the initial estimate is most important. The initial estimate is obtained by N4SID or MOESP methods and the similar transforms to canonical-formed state space models.

View full abstract

[in Japanese]

View full abstract

This paper deals with parameter identification of LPV descriptor systems. We present a natural prediction error identification method for LPV descriptor models.We use gradient and Hessian based nonlinear optimization algorithms to minimize the cost function. The gradients and Hessians are computed using a simulation of a LPV descriptor system. The steepest-descent algorithm and the Gauss-Newton algorithm are employed together to improve the convergence and the computation speed in solving the identification problem. Numerical simulations of a magnetic levitation system, demonstrate the effectiveness of the proposed algorithm.

View full abstract

A recursive subspace identification algorithm is derived in which the size of input and output data matrices is fixed. The proposed method estimates both deterministic and stochastic part of the systems. Efficacy of the algorithm is demonstrated by simulations, comparing the numerical results with that obtained by other methods based on the subspace tracking.

View full abstract

The purpose of this paper is to propose a new identification method which gets around some problems of the MD identification method. To this end, this paper focuses its attention on the relation between the frequency responses of discretized systems that result from different sampling rates. Such a relation can be derived from the impulse modulation formula. The problems of the MD identification method can be circumvented by solving a least squares problem under the constraint posed by the relation derived. Then, a pneumatic servo system is identified with the proposed method and the result is compared with that obtained by a usual MD identification method.

View full abstract

The authors have recently proposed a method for estimating unknown parameters attributed to the received signal which is contaminated by random noise and, its efficacy has been tested to the real ultra sonic radar system. In this paper the method is applied to the parameter estimation for FM-CW radar signals. The effectiveness of the method is confirmed by simulation experiments.

View full abstract

In this paper, we consider the control of periodic motions with delayed feedback. Compared to the servo systems where the reference inputs are always required, Delayed Feedback Control has a remarkable feature that a priori information about the location of unstable periodic orbit is not necessary. We demonstrate the merit of DFC by using a ball and beam system.

View full abstract

In this paper, we propose the design method of Delayed Feedback Control using Pade approximation, which yields a larger allowable delay than the difference approximation method. By simulation and experiment, we show that the proposed method is not much effected by additional noises.

View full abstract

Unlike automatic control, it is generally difficult to avoid vibration by manual control.In the system which depends on manual operations, this vibration is the phenomenon which is undesirable to the system and operator. In this paper, to suppress the vibration in manual control, we use additional feedback control based on a delayed feedback mechanism.

View full abstract

This paper concerns the global structure of the phase space in magnetoelastic beam system under the time delayed feedback control. The numerical investigation of an unstable manifold governing the global structure reveals that the chaotic dynamics of the controlled object possibly remains in the controlled system. We show the remanence causes long and irregular transient behavior before the convergence towards the target orbits.

View full abstract

In order to control a plant with time delays, predictive control schemes which use a plant model, such as a Smith controller or a state predictive controller, are very effective. In such a controller, mismatches between the nominal model and the real plant affect closed-loop stability seriously. So robust stability analysis is important to design the predictive controller. In this study, we first construct a state predictive observer for a plant with multiple output time delays, and show a method to obtain stability margins in the delay-times and the gains for a regulator with the constructed observer.

View full abstract

In this paper, we present a stability ctiterion for neutral time-delay systems. It is derived by small gain approach and is given in the linear matrix inequality (LMI) form. The stability criterion can also be applied to retarded time-delay systems in a special case.

View full abstract

In this report, stability conditions and robust stability conditions of linear time-delay systems are derived in the form of necessary and sufficient condition by using the Lambert W function. Moreover, a graphical method of stability check with the obtained conditions is shown in the example.

View full abstract

This paper presents a new design method of decoupling of systems with unstable zeroes.

View full abstract

Although inexpensive and small sensors are often useful for systemcontrol, we have to compromise with them on some characteristics suchas accuracy, bandwidth and so on. Nevertheless, the performance canbe improved, we compose systems of heterogeneous sensors by usingfilters. However, the design of such filters may not bestraightforward, since both the gain and phase should be dealt with atthe same time. This paper proposes a systematic filter design methodbased on H_∞ control. We examine the effectiveness of theproposed method by an application to rate gyro systems.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

Stability of delta-operator diamond polynomials is known to be equivalent to that of some segment polynomials contained in the family. However, extreme point results do not hold for them.We already proposed two stability analysis methods for delta-operator diamond polynomials: one uses stability analysis method for delta-operator segment polynomials, and the other uses directional stability radii. In this paper, comparison is made to examine which method is more efficient.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

We recently focus on a new integrated drive generator (IDG) in aircraft, which has a generator of 90kVA, 400Hz and an economical traction drive-type continuously variable transmission (CVT) as a constant speed-drive unit. The frequency needs to control a hydraulic system by some micrometer strokes. The purpose of this study is to make a contribution to achieve high reliability by using a minimal observer without stroke sensors.

View full abstract

[in Japanese]

View full abstract