In this paper, we report on the improvements in fabrication processes of polymer micro cantilevers for force controlled atomic force microscopy. The improvement of photo mask patterns results in the enhancement of controllability of the pyramidal probe tip geometry and in rapid dissolving of sacrificial layer. It also contributes to process simplicity.

View full abstract

Since mass can be calculated with a resonant frequency of vibration, it is considered that the frequency estimation method using LMS algorithm is applicable to a dynamic mass measurement. In this year, mass measurement is experimented for real time under the Windows environment when a measuring object is sliding on the goods stand.

View full abstract

A method for the detection of friction anomalies in a pneumatic linear bearing is described. In this method an optical interferometer is used to accurately measure the force acting on the moving part of the bearing.

View full abstract

This paper proposes a local phase velocimetry based on the spatio-temporal gradient analysis. The method has an ability to measure the local phase velocity of the zeroth order anti-symmetric Lamb-wave field through the linearity among the four-dimensional vectors which is defined by following components: 1:a vertical displacement, 2: vertical particle velocity, and 3,4:a pair of shear strains of the surface.

View full abstract

In this application, it is used ultrasonic sound waves which are not affected from bad operating conditions to be able to give the quantity of fuel level as litre in the tank to the driver and make measurement system more sensitive. The model according to dimensions which the tank is produced is also stored to training phase of neural network.

View full abstract

A filter for enhancement of intravascular ultrasound images is proposed. This filter performs this task in two phases: first a pre-processing based on morphological smoothing and then a fuzzy enhancement is done on the basis of the intensity and variance of input images. Results have shown that this technique is useful to facilitate the segmentation of ROIs in IVUS images.

View full abstract

This paper proposes non-destructive inspection of concrete structures using ultrasonic sensors. We first try to filter the ultrasonic sensor output with an adequate band pass filter to remove the affect from the disturbance and next to apply a pattern matching method based on a multi-reflected wave model in order to realize a reliable and accurate defect detection and depth measurement.

View full abstract

Records of the silver- and aluminum-point blackbodies at the NMIJ/AIST since 1992 were reviewed in the features such as the plateau profile, the super cooling, the lifetime and the plateau temperature. The temperature agreement between two silver-point blackbodies was within 0.03 K in recent three years. The temperature of two aluminum-point blackbodies agreed within 0.03 K in recent seven years.

View full abstract

A novel metal-carbon eutectic fixed-point cell taking advantage of a hyper-eutectic porous structure to improve its robustness is proposed. Temperature repeatability comparable to conventional metal-carbon eutectic fixed point cells has been demonstrated. The applications of this technique to realizing long cells for thermocouple calibration and large aperture cells for irradiance standards in radiometry / photometry are of particular interest.

View full abstract

High temperature fixed-points based on eutectic alloys of metals and carbon have been proposed as secondary reference points for the International Temperature Scale of 1990. These allow checks of thermometry with a precision rivalling the uncertainties of many National Measurement Institutes. Aspects of the construction of such fixed-points are considered with guidelines for making fixed-points that have best performance.

View full abstract

A facility for infrared radiation thermometry in the middle temperature range from 100 oC to 500 oC is being established by the NMIJ. This calibration facility consists of an infrared radiation thermometer and the fixed point blackbodies of In, Sn, and Zn. An InSb infrared radiation thermometer and a zinc-point blackbody with a large opening of 15mm in diameter are constructed and characterized.

View full abstract

The fixed-point blackbodies had been developed for establishing international temperature scale at NIM since 1965. This paper reports the development of blackbodies, including configuration, making fabrication and realization, of the blackbodies of gold, copper, silver, aluminum and zinc freezing points at NIM and their applications in realizing and disseminating international temperature scale.

View full abstract

This paper proposes a self-organizing map for the visual perception of a partner robot. The robot memorizes the background by using temporal difference extraction of images. Next, the color patterns of a human are learned by using k-means algorithm and self-organizing map. The experimental results show that the robot can recognize the human efficiently.

View full abstract

An on-line unsupervised learning mechanism is proposed for unlabeled data which is polluted by noises. By using a similarity threshold and local error based insertion criterion, the system is able to grow incrementally and to accommodate input patterns of on-line non-stationary data distribution.

View full abstract

It is well known that a brain consists of many sub-modules, each of which performs a specific role. The experimental results of many studies have indicated that prefrontal cortex appropriately switches each module. In this paper, we consider the brain as a hybrid dynamical system which is composed of a higher module having discrete dynamics and a lower module having continuous dynamics.

View full abstract

Various flexible neural probes with sufficient flexibility to achieve minimally invasive interfacing have been developed. Some have micro-fluidic channels useful for injecting chemicals into neural tissues and other purposes. A rat car system is introduced as an application of a neural interface system in which signals recorded from a rat motor cortex are used to control a car.

View full abstract

We developed a method to estimate functional connectivity between cultured neurons. Spontaneous action potentials were recorded by multielectrode arrays and “connection map analysis” was performed. Functional connectivity map was generated from spatio-temporal pattern of spontaneous action potentials in dissociated neurons. “Connectivity map analysis” revealed that the functional connectivity between neurons changed drastically by the induction of synaptic potentiation in cultured living neuronal networks.

View full abstract

In this paper, for the safety of human operators in the real process, we apply feedback error learning scheme (we abbreviate to FEL in the following ), which is a novel architecture that models human learning, to reliable control systems in order to prevent damages from accidental faults.

View full abstract

We propose a face recognition system which distinguishes between a person and a face photo from infrared images. The system can detect a facial region robust for background variations because of the infrared image characteristic. Person-photo distinction is performed by a method based on the photometric stereo. The method can detect a curved photo as well as a flat photo.

View full abstract

We propose an implementation of automated monitoring system using multiple cameras embedded in the environment. At first, the object intruding to a sensitive area is detected using planar view-volume intersection method, and the shape of the object is measured. Then the motion of the object is tracked using generated CG images. This system requires no prior knowledge of the object.

View full abstract

This paper discusses wheelchair support system using humanoid robots as a example of human support using humanoid robots. In this proposal system, humanoid robots provide both of mobility and manipulation supports for the wheelchair users. Then we show the successful experimental results on pushing a wheelchair using a real humanoid robot HRP-2.

View full abstract

This paper describes a method of building a three-dimensional map by a moving camera using measurement of artificial landmarks and recognition of human actions. The proposed method distinguishes static and dynamic objects by recognition of human actions. The experimental result shows the feasibility of monitoring system using the cooperation of fixed and moving cameras.

View full abstract

An ECT (Eddy Current Testing) camera system is proposed to quantitatively evaluate/visualize cracks of the metal structures. It is composed of the multi-coil ECT probe, ECT instrument, and computer-aided analysis/visualization that correspond to the lens, film and development/picture in the elements of photography, respectively. It is applied to inspection of an austenite stainless steel plates with fatigue cracks so as to discuss the accuracy of the system.

View full abstract

This paper presents a novel localization method for a magnetic dipole source. Using a certain combination of gradiometers which consist of oppositely wound circular loops by turns, complex gradients of the magnetic flux can be measured. An algorithm to express the dipole location directly by the complex gradients is shown.

View full abstract

A method that reduces sensor noise and artifacts from MEG data is proposed. Factor analysis(FA) and Kalman filter are employed for sensor noise reduction. FA estimates noise covariance for Kalman filter. After the sensor noise reduction, ICA is used to eliminate artifacts. Simulation studies confirmed that the SNR of estimated independent component increases.

View full abstract

A spatial filter for MEG analysis without any temporal and prior information is proposed. The spatial filter outputs the largest value at its target component. The current distribution estimated with spatial filter is localized by using Mallows Cp statistic. Simulations verify that the method estimates current distribution well. The new method gives more reliable estimation than our conventional method.

View full abstract

This paper is concerned with three dimensional defect reconstruction from magnetic images. Magnetic measurements arising in eddy current testing are used for detecting and characterizing defect shape. A computational method using virus-evolutionary genetic algorithm (VEGA) is proposed for sizing such segments. Successful computational experiments will be summarized in the presentation.

View full abstract

We have presented a new MATLAB toolbox for robust control synthesis by symbolic computation. It is based on a parameter space approach accomplished by quantifier elimination and also equipped with a machinery for numerical simulation of the properties in a frequency domain. The toolbox is explained with several examples.

View full abstract

Simple design methods of attachable robust compensators for any existing control systems are presented. Principle of the new methods is to settle on real plants as their nominal models with local compensators. Since the additional local compensators are designed independent of previously designed controllers, they are applicable for any existing control systems including nonlinear and/or non-closed form control scheme.

View full abstract

A systematic framework of robust state-feedback design is proposed for LPV plants without assuming measurability of dependent parameters. Conditions for the existence of the state-feedback controllers which achieve robust quadratic L2 gain performance for not only nominal parameters but also real parameters in a combined parameter space are given in terms of Linear Matrix Inequalities (LMI’s) expression.

View full abstract

This paper presents new results on designing a robust adaptive direct controller for a class of non-linear first order systems, based on the use of dead zone in parameters update law. It is shown that size of the dead zone does not depend on upper bounds of disturbances. Computer simulation results show the effectiveness of the controller.

View full abstract

An analog current regulator for stepping motors is converted into a digital driver using the so-called PIM method, taking into account the integral action. Experimental plant models are obtained and the robustness of the resulting PIM controller is assessed. It gives performances very close to those of the analog driver even at a very low sample rate of 4kHz.

View full abstract

This paper is concerned with the Black Box modelling of induction motor from test data. The aim of the work is to obtain model of the induction motor directly from test data. The process of modelling and validation is then carried out using MATLB nonlinear system identification toolbox. In this application a model structure (NNARX) was assumed.

View full abstract

In this paper, a recursive algorithm is presented, based on the CVA subspace identification algorithm. The main idea was to explore the relations between the orthogonal and oblique projections involved and to provide simpler expressions that allowed a recursive version of the algorithm - guaranteeing most of the advantages of this kind of methods, and still improving the numerical efficiency.

View full abstract

Today's automobiles with petrol engines are equipped with 3-way catalytic converters that purify automobile exhaust emissions. In this paper, we construct a simplified model for the dynamics of 3-way catalyst based on the assumption that catalyst behavior is dominated by the dynamics of O₂ storage. Then we propose a method for parameter identification of the constructed model.

View full abstract