When returning on the earth by the space flight, the space deconditioning may be developed. As this countermeasure, the artificial gravity load device using the centrifuge is proposed in the space station. But the gravity load might cause the faint, and safe gravity load is uncertainty. We proposed that discriminate strength of gravity tolerance using heart rate variability time series. Step function was inputted to AR model estimated from heart rate variability time series during rest or under light gravity load, and strength of the gravity tolerance was discriminated by the step response function. On the result, discriminant accuracy was 87.5% by using heart rate variability time series when gravity load of 1.0 G was added to the human lying on the supine. Therefore, possibility of discriminant of gravity tolerance was obtained by using heart rate variability time series when sympathetic hyperactivity. Discriminant of the gravity tolerance is expected before countermeasure of space deconditioning is executed.

View full abstract

In this paper peak count correction methods for a self-coupling distance sensor with a Vertical-Cavity Surface-Emitting Laser (VCSEL) are reported on. The distance to the target in these distance sensors is usually calculated by determining the frequency of power fluctuations that are produced by the self-coupling effect. In this study it is proposed that the peak count correction method using statistical feature with little computational effort than Fast Fourier Transform (FFT) and higher accuracy than traditional peak count methods. The counting error is reduced to less than 1/100 with the peak count correction methods under adverse condition.

View full abstract

In this paper we study a method for designing a PID controller which can assign pole and zero locations of multi-input systems. The basic idea of our synthesis is to draw a PID control law from the state-space method which can assign pole-zero locations. We propose a state feedback control law in a part of input channels to solve the zero assignment problem. This loop is so designed as to preserve the controllability of the closed loop system. With the help of the controllability, we can deal with the pole assignment problem by means of another state feedback control law in the remaining input channels. Numerical example illustrates the effectiveness of the results.

View full abstract

A traffic accident is a disturbing matter caused by a rise of the recent vehicle diffusion rate worldwide. Therefore, estimation of driver's action unrelated to driving may help to reduce traffic accidents. This paper proposes a novel method to estimating driver's action unrelated to driving by measuring signals from air mass flow sensor placed on the seat belt. A previous study tried to estimate driver's action by measuring signals from pressure distribution sensors placed on the driving seat. However, the previous study was concerned about a large difference of accuracy rate when the drivers reseat. We focus on the seat belt since there is very little change of wearing form via three stationary points. As a result, the accuracy rate when drivers reseat with our proposed method came over 90%. Consequently, proposed method is quite effective and useful to estimate driver's action.

View full abstract

In the research field of bipedal locomotion, a central pattern generator (CPG) and passive dynamic walking (PDW) have attracted much attention. In this paper, we describe a motion control system for biped robots based on dynamic joint passivization. Our motion control system is based on a mixture of the CPG and PDW, that is, the multiple dynamic passivization of joint control (MDPJC). Our intention is to make the joint control of the swing leg temporarily passive in the swing leg phase. The important part is the passive phase time and the switch timings of the joint control. We optimize the switch timing parameters using simulated annealing with advanced adaptive neighborhood (SA/AAN). Experiments using the motion control system based on multiple dynamic passivization of joint control successfully generated energy efficient walking and enabled superior gaits.

View full abstract

We propose an inchworm type microactuator which uses levitation caused by vertical vibration. The vertical vibration is generated by a piezoelectric actuator. The inchworm usually consists of thrust elements and adhesive elements. The proposed inchworm type microactuator does not use any adhesive elements. Three masses are connected with two thrust elements which are stacked-type piezoelectric actuators. One mass which is at the center of the inchworm is also connected with a vertical vibrating piezoelectric actuator. The vertical vibration levitates the central mass and eliminates the friction. The other masses without vibrating elements have constant friction force. By controlling the phase of the thrust elements and vibrating element, the inchworm moves in the liniear directions. The position of the inchworm is measured, and some experimental results indicate the feasibility of the proposed inchworm. The mechanism described in this paper is effective in a precise positioning system in a cleanroom.

View full abstract

This paper proposes a new on-line signature verification technique which employs gradient features and a pooled within-covariance matrix of training samples not only of the user but also of the others. Gradient features are extracted from a signature image reflecting the velocity of pen movement as the grayscale so that both on-line and off-line features are exploited. All training samples of different signatures collected in design stage are pooled together with the user's samples and used for learning within-individual variation matrix to reduce required sample size of the user to minimum number. The result of evaluation test shows that the proposed technique improves the verification accuracy by 4.9% when user's sample of size three is pooled with samples with others. This result shows that the samples of different signatures are useful for training within-individual variation of a specific user.

View full abstract

This paper proposes a technique for direction estimation by means of square grid points in order to improve the performance of cadastral map assembling technique based on Merlin-Farber (MF) algorithm. The MF algorithm requires direction normalization of the segments (of cadastral map) preceding the assembling. Proposed direction estimation technique is based on the spatial frequency analysis of autocorrelation by MF algorithm for the square grid points regularly drawn with constant intervals on the segments. Since many square grid points are drawn over entire area of the segments the direction can be estimated more accurately with those points when compared the direction is estimated with single north arrow. To assemble two adjacent segments the longest common boundary is detected by MF algorithm. Evaluation experiments are performed to compare the accuracy and the success rate of map assembling when the direction is estimated and normalized based on the square grid points and when estimated and normalized based on the north arrow. Total of 324 map segments of 47 district provided by Institut Geographique National France are used in the experiments. While the map assembling based on the north arrow tends to form inaccurate cadastral maps the proposed technique assembles the map more accurately. The results of experiments shows that the proposed technique achieves sufficient success rate and accuracy so that it effectively reduces the labor cost and time of the cadastral map assembling.

View full abstract

We propose a novel approach for detection of the facial midline (facial symmetry axis) from a frontal face image. The facial midline has several applications, for instance reducing computational cost required for facial feature extraction (FFE) and postoperative assessment for cosmetic or dental surgery. The proposed method detects the facial midline of a frontal face from an edge image as the symmetry axis using the Merlin-Faber Hough transformation. And a new performance improvement scheme for midline detection by MFHT is present. The main concept of the proposed scheme is suppression of redundant vote on the Hough parameter space by introducing chain code representation for the binary edge image. Experimental results on the image dataset containing 2409 images from FERET database indicate that the proposed algorithm can improve the accuracy of midline detection from 89.9% to 95.1 % for face images with different scales and rotation.

View full abstract

In this paper, we establish effort and error prediction models using an artificial neural networks (ANNs). We propose the normalizing method to reduce the margin of errors for ANN models. In addition, we perform an evaluation experiment to compare the accuracy of the ANN models with that of the regression analysis (RA) model and that of two ANN models using Steel-Dwass's multiple comparison test. The results show that each ANN model is more accurate than the RA model and the proposed method can reduce the errors for some cases, since the mean errors of the ANN models are statistically significantly lower.

View full abstract

There is code shift keying as one of effective modulation and demodulation method of signal on optical code division multiple access. We propose multiplexing code shift keying which is improved data rate on each user, by using two or more codewords in one frame, and evaluate it.

View full abstract

This paper proposes the method to convert digital image containing distinguishing difficulty sets of colors into the image with high visibility. We set up four criteria, automatically processing by a computer, retaining continuity in color space, not making images into lower visible for people with normal color vision, and not making images not originally having distinguishing difficulty sets of colors into lower visible. We conducted the psychological experiment. We obtained the result that the visibility of a converted image had been improved at 60% for 40 images, and we confirmed the main criterion of the continuity in color space was kept.

View full abstract

This paper attempts to activate a large scale shopping streets (shotengai) using Internet technique. Shotengai is a style of Japanese commercial district street. Recently, decline in shotengai is a serious problem. So, we take up “Osu” which is famous shotengai in Nagoya, Japan. We architect Web visualization system, which has useful usability and findability for Osu shotengai information. We also develop methods for competing for a store owner with another one by their prominency and for attractive design to increase motivation of store owners. The number of users of Osu map is increased rapidly. We hope to apply our instance to the activation of shotengai in Japan.

View full abstract

We aim to discover the factors of an unpleasant sound by processing the sound in the frequency domain. The processed sounds have the same amplitude envelope as an unpleasant sound, and consist of only one or two peak frequencies of the original unpleasant sound. Using these sounds, we attempt to specify a frequency band relating to unpleasantness. As a result, the sound with two pair of peak frequencies tends to make subjects feel unpleasant compared with one peak, and subjects feel unpleasant by the sounds having peaks in 2kHz and 9kHz around. Next, we investigate whether the sound becomes unpleasant by adding the amplitude envelope of unpleasant sounds. The result shows that addition of the envelope makes subjects feel unpleasant compared with the carrier alone.

View full abstract

A specific crystalline defect in diamond may work as a NOT gate in a quantum computer. It is called NV-N center that is composed of two substitutional nitrogen atoms that are separated by c.a. 1.5 nm and one of them is adjacent to a vacancy. In order to find an optimal condition to produce it using a low-energy (100 ∼ 500 eV/atom) N₂ beam implantation, we performed an empirical molecular dynamic simulation. By a high-temperatures implantation at 900∼1000 K, the formation probability of such desired NV-N centers was 5 ∼ 10%, which is comparable with a measured data.

View full abstract

We demonstrate triple ion implanted 4H-SiC bipolar junction transistor (BJT). By etching the extrinsic base regions using inductively coupled plasma dry etching, the characteristics of triple ion implanted 4H-SiC BJT were significantly improved. Maximum common emitter current gain was improved from 1.7 to 7.5.

View full abstract

A wireless power transfer system based on two coupled resonators is analyzed by the filter design theory. Many useful relations between the equivalent circuit components are derived to comply with the change of power transfer condition along with the basic design of the system.
Some design examples are given to deepen understanding of the theory and thus to stimulate using it for the system design. The effect of the resonator losses is also addressed to show the robustness of the theory, indicating the circuit loss of almost 10 dB does not deteriorate the matching condition too much.

View full abstract

A simple design method for the resonator-coupled wireless power transfer system is proposed on the basis of the BPF theory. Three parameters of the constituting resonators are obtained in advance to design a BPF, which include the resonant frequency, coupling coefficient between adjacent resonators, and the external Q of the outermost resonators. The design is carried out adjusting those parameters to cope with a 2-stage BPF made of two direct coupled resonators. Since the present design theory can be applied for any resonators as long as their resonant frequency is chosen the same, it has a wide range of practical application. The adopted design example is a magnetically coupled 2-stage resonator system proposed by MIT group, since it is most popular, but any usable design theory has not been presented so far.

View full abstract

Motor evoked potentials (MEPs) elicited by transcranial magnetic stimulation (TMS) over the primary motor cortex (M1) vary in their amplitude from trial to trial. To investigate the functions of motor cortex by TMS, it is necessary to confirm the causal relationship between stimulated sites and variable MEPs. We created artificial neural networks to classify sets of variable MEP signals and finger forces into the corresponding stimulated sites. We conducted TMS at three different positions over M1 and measured MEPs of hand and forearm muscles and forces of the index finger in four subjects. We estimated the sites within motor cortex stimulated by TMS based on cortical columnar structure and nerve excitation properties. Finally, we tried to classify the various MEPs and finger forces into three groups using artificial neural networks. MEPs and finger forces varied from trial to trial, even if the stimulating coil was fixed on the subject's head. Our proposed neural network was able to identify the MEPs and finger forces with the corresponding stimulated sites in M1. We proposed the artificial neural networks to confirm the TMS-stimulated sites using various MEPs and evoked finger forces.

View full abstract

Falling is one of the most serious problems for the elderly. The aim of this study was to develop a screening method for identifying factors that increase the risk of falling among the elderly, particularly with regard to lower limb muscular strength. Subjects were 48 elderly volunteers, including 25 classed as healthy and 23 classed as frail. All subjects underwent measurement of lower limb muscular strength via toe gap force and measurement of muscle strength of the hip joint adductor via knee gap force. In the frail group, toe gap force of the right foot was 20% lower than that in the healthy group; toe gap force of the left foot in the frail group was 23% lower than that in the healthy group, while knee gap force was 20% lower. Furthermore, we found that combining left toe gap force and knee gap force gave the highest odds ratio (6.05) with 82.6% sensitivity and 56.0% specificity when the toe gap force was 24 N and the knee gap force was 100 N. Thus, lower limb muscular strength can be used for simple and efficient screening, and approaches to prevent falls can be based on quantitative data such as lower limb muscular strength.

View full abstract

In practical situations, poor control performance due to parameter uncertainty is frequently unavoidable. So, robustness against uncertainty has become a desirable property in any control system, especially one that requires accurate parameters such as Model Predictive Control (MPC). With just a nominal model, MPC fails to stabilize a plant that incorporates parameter uncertainty. Robust constrained LMI-based MPC offers a way to handle parameter uncertainty. Therefore, it is appropriate to extend this particular method to cover more types of problems. This research analyzes and synthesizes reference trajectory tracking extension and constant disturbance rejection extension to robust constrained LMI-based MPC in order to broaden the practicality of MPC to process control. The control laws for both reference trajectory tracking extension and constant disturbance rejection extension are obtained through system augmentation to include error dynamics and integrator dynamics, respectively.

View full abstract

The conventional electromagnetic shielding technique is all but impossible to fundamental solution of the problems in the information and communication fields, such as virtual image for radar. Namely, it is necessary to receive a required electromagnetic wave as the information signal, and to shield a needless electromagnetic wave as the noise. the present research has developed the carbon, copper, ferrite, and BPSCCO plates, as the typical shielding material, having the orientation characteristics in the RF (radio frequency) magnetic field. To exhibit the orientation characteristics in the plane wave, it has formed the slit on the surface of typical shielding materials; termed the slit plate. For example, the value of RF magnetic shielding degree SD_HP of slit carbon plate for holding the slit perpendicularly to the ground increased with frequency in the region from 1 MHz (7 dB) to 3 GHz (70 dB). And, the value of SD_HH when holding the slit horizontally is indicated an average value of approximately 10 dB in this frequency region. That is, the difference values, SD_HP-SD_HH, indicated the orientation characteristics. Experimental results revealed several characteristics of the slit plates that include the influences of orientation characteristics on the slit length, slit width, and slit number. In the present paper, it was succeeded to improved the difference average value of approximately 35 dB for SD_HP-SD_HH, by the sandwich of slit ferrite plate over a slit carbon plate, in the civilian communication frequency region from 1 MHz to 3 GHz. In addition, important criteria are discussed for the design of an effective RF magnetic shielding plate having orientation characteristics.

View full abstract

It is important to have ‘Inquiry-mind’ which is the disposition to access information in a critical manner. There are several previous approaches that use discussion ‘Inquiry-mind’. However it is not easy to cultivate Inquiry-mind only by discussion. This study proposes a method for cultivating Inquiry-mind that focuses on opening one's eyes to the possibility of belief bias, that is the tendency to gather agreeable information. This study proposes a method for the quantification of belief bias based on three factors, accessing information, deciding the degree of importance and viewpoint. From the evaluation experiment in which 33 college students participated, it was observed that, accessing information and viewpoint are especially important for cultivating Inquiry-mind. This result confirms the efficacy of the model and argue in favor of its combined use with other cultivating methods.

View full abstract

This paper intends to verify effectiveness of real options approach for evaluating investments in Enterprise Resource Planning systems (ERP) and proves how important it is to disclose shadow options potentially embedded in ERP investment. The net present value (NPV) method is principally adopted to evaluate the value of ERP. However, the NPV method assumes no uncertainties exist in the object. It doesn't satisfy the current business circumstances which are filled with dynamic issues. Since the 1990s the effectiveness of option pricing models for Information System (IS) investment to solve issues in the NPV method has been discussed in the IS literature. This paper presents 3 business cases to review the practical advantages of such techniques for IS investments, especially ERP investments. The first case is EDI development. We evaluate the project by a new approach with lighting one of shadow options, EDI implementation. In the second case we reveal an ERP investment has an “expanding option” in a case of eliminating redundancy. The third case describes an option to contract which is deliberately slotted in ERP development to prepare transferring a manufacturing facility.

View full abstract

Most existing query-biased summarization methods generate the summary using extracted sentences based on similarity measure between all sentences in documents and the query. If there are plural sentences having high similarity to the query in the documents, however, these methods cannot decide from which sentence the summary should be made. This paper proposes an algorithm considering difference of paragraphs, adopting new indicator that shows the difference between one paragraph and the others. In a word space composed of all words in the target document, the algorithm determines the axis that maximizes the difference when a paragraph and the others are projected onto it. There are many combinations of a paragraph and a set of other paragraphs. For each combination, the above-mentioned axis that maximizes the difference and gives a conformity degree to the given query is calculated. With these conformities, the algorithm decides one paragraph for generating the summary. To obtain the axes, topic distinctiveness factor analysis is applied. The basic idea for making final summary is concatenating the sentences extracted from the paragraph. The resultant summary is evaluated from the points of readability, understandability and the easiness to judge whether the link works well or not.

View full abstract

HugeTLBfs improves performance of the applications which need huge data area by using Huge Page because of reducing TLB misses that affect an efficiency of address translation. However, Huge Page sizes which can be used concurrently are limited to only one size. If a pretty large size in Huge Pages is adopted to reduce TLB misses dramatically, its size is allocated even to the remaining fragment in the last part of data area. As a result of allocating a large size to small area, large useless data area that any processes can not use is generated. On the other hand, if a small size is adopted, improvement of performance is trivial. Then we added the second Huge Page size to fill fragment that is concurrently available with the large ones. Consequently, our method makes it possible to minimize useless data area without going down in performance.

View full abstract

In recent years, with spread of Web application and performance gain of Web browsers, the demand of the web-based supervisory and control(WSCADA) systems based on RIA(Rich Internet Application) is increased.
To develop CRUD operations(Create, Read, Update, Delete which corresponds to the basic database operations) of RIA-based web applications, various frameworks and libraries are being provided. However, to develop behavior operations, a lot of program must be written manually. The typical operations of WSCADA are behavior operations, so even if RIA frameworks and libraries are used to develop WSCADA, the productivity of development doesn't improve. Although conceptual models and development environment have been proposed for typical web applications consisted mostly of CRUD operations, those for WSCADA is still the unsolved problem.
This paper proposes the user interface model and the development environment for the monitoring user interface program of WSCADA. We focus on the productivity enhancement of the WSCADA development, and propose the Monitoring User Interface Model(MUM) extended Model-View-Controller(MVC) model. We design the Ajax framework and the development environment based on our model. We define the DisplayItem as the advanced View and the MonitoringItem as the advanced Model, and classify the Controller into the Interaction and the Behavior. Our Ajax framework based on web browser's standard technologies, provides the mapping between conceptual model elements. We define the domain specific language for writing the mapping. We design development environment for auto-generating Behavior program from the mapping. In this paper, we evaluate our model and development environment through the experimental development of the typical WSCADA. As a result, the development cost of the WSCADA based on our framework is only one fifth of that based on the typical Ajax library.

View full abstract

Recently, BBS (Bulletin Board System) has been frequently used to exchange users' opinions on the Internet. An efficient retrieval method that uses past other users' evaluation has been proposed to retrieve target information efficiently and correctly. In this method, it is necessary to cluster past queries to calculate accurately thread score. However, this method has defect from practical points of views, because the time for clustering is so long that an user feels stress. Therefore, we proposed an improvement method which utilizes past clustering result for online response of BBS articles retrieval with users evaluation. Through an experiment, the accuracy of retrieval was almost same as compared with the conventional method, and the average time for retrieval reduced from 25.4 seconds to 3.1 on average using 890 threads.

View full abstract