Simulation of ultrasonic-vibration drawing using the finite element method (FEM)

In ultrasonic-vibration drawing, wires are drawn while ultrasonic vibration is applied to a drawing die. Prior studies provide experimental proof that ultrasonic-vibration drawing reduces drawing resistance, improves lubrication and prevents wire breakage. In the future, ultrasonic-vibration drawing is expected to contribute to the drawing of difficult-to-draw materials and operations, such as shaped wires, ultrafine wires, and the wire drawing operation in semidry or dry condition. However, a detailed analysis and understanding of the mechanism of improvement is not possible on the basis of conventional experimental observations because the ultrasonic-vibration processing phenomenon occurs at high speed. Therefore, we attempted to understand the processing mechanism of ultrasonic-vibration drawing using the finite element method (FEM). ABAQUS was used for the FEM. Drawing force and stress–strain distributions in drawn wires were analyzed. From these studies, we quantitatively clarified the mechanism of improved drawing characteristics, such as decreased drawing force.     

What kind of floor am I standing on? Floor surface identification by a small humanoid robot through full-body motions

This study addresses a floor identification method for small humanoid robots that work in such daily environments as homes. The fundamental difficulty lays in a method to understand the physical properties of floors. To achieve floor identification with small humanoid robots, we used inertial sensors that can be easily installed on such robots, and dynamically selected a full-body motion that physically senses floors to achieve accurate floor identification. We collected a training data-set over 10 different kinds of common floors in home environments. We achieved 85.7% precision with our proposed method. We also demonstrate that our robot could appropriately change its locomotion behaviours depending on the floor identification results.     

Dynamics of learning near singularities in layered networks

We explicitly analyze the trajectories of learning near singularities in hierarchical networks, such as multilayer perceptrons and radial basis function networks, which include permutation symmetry of hidden nodes, and show their general properties. Such symmetry induces singularities in their parameter space, where the Fisher information matrix degenerates and odd learning behaviors, especially the existence of plateaus in gradient descent learning, arise due to the geometric structure of singularity. We plot dynamic vector fields to demonstrate the universal trajectories of learning near singularities. The singularity induces two types of plateaus, the on-singularity plateau and the near-singularity plateau, depending on the stability of the singularity and the initial parameters of learning. The results presented in this letter are universally applicable to a wide class of hierarchical models. Detailed stability analysis of the dynamics of learning in radial basis function networks and multilayer perceptrons will be presented in separate work.     

Enzyme-linked immunosorvent assay using vertical microreactor stack with microbeads

The advantages of vertical microreactor stack with three-dimensional (3D) structure for immunoassay are discussed. The vertical microreactor stack uses vertical fluid flow operation with multifunctional fluid filters. The multi function of fluid filter is very effective for micromixing and passive valve operation. The mechanism of micromixing is discussed by using computational fluid dynamics (CFD), and we know that the mixing mechanism based on Coanda effect. To evaluate the micromixing performance of fluid filter, we demonstrated enzyme reaction with unique repeat mixing operation. As the results, we proved that the fluid filter has very effective mixing performance. The detection limit, which demonstrated by competition enzyme-linked immunosorvent assay (ELISA), is comparable with recommended detection limit, which suggested by Japanese ministry for the environment.     

Three-dimensional medical image recognition of cancer of the liver by a revised radial basis function (RBF) neural network algorithm

In this study, we propose a revised radial basis function (RBF) neural network algorithm and apply this algorithm to computer-aided diagnosis (CAD) of the liver. First, the revised RBF neural network algorithm is applied to recognition of the liver regions, and the recognition results are compared with those obtained using the conventional RBF neural network and the conventional multilayered neural network trained using the back-propagation algorithm. It is shown that the revised RBF neural network is accurate, and is a useful method because the parameters are automatically determined. Then, the revised RBF neural network is applied to CAD of the liver cancer called hepatocellular carcinoma (HCC).     

Microstructure and properties of (Sr, Ca)CuO2/(Sr, Ca)RuO3 multilayers

Epitaxial multilayer thin films of infinite-layer (Sr, Ca)CuO₂ and perovskite (Sr, Ca)RuO₃ have been prepared on (100) SrTiO₃ substrates by multitarget rf magnetron sputtering. X-ray diffraction analyses revealed that the multilayer structure of (Sr, Ca)CuO₃/(Sr, Ca)RuO₃ was successfully fabricated with a minimum layer thickness of 20 Å. Transmission electron microscopy measurements of the multilayers indicated that there was no dislocation which normally exists in single-layer films with an infinite-layer structure. Resistivities of multilayer films at room temperature ranged from 1 to 10 m cm and showed semiconductor-like dependence against the temperature.     

Rational Filters for Passive Depth from Defocus

A fundamental problem in depth from defocus is the measurement of relative defocus between images. The performance of previously proposed focus operators are inevitably sensitive to the frequency spectra of local scene textures. As a result, focus operators such as the Laplacian of Gaussian result in poor depth estimates. An alternative is to use large filter banks that densely sample the frequency space. Though this approach can result in better depth accuracy, it sacrifices the computational efficiency that depth from defocus offers over stereo and structure from motion. We propose a class of broadband operators that, when used together, provide invariance to scene texture and produce accurate and dense depth maps. Since the operators are broadband, a small number of them are sufficient for depth estimation of scenes with complex textural properties. In addition, a depth confidence measure is derived that can be computed from the outputs of the operators. This confidence measure permits further refinement of computed depth maps. Experiments are conducted on both synthetic and real scenes to evaluate the performance of the proposed operators. The depth detection gain error is less than 1%, irrespective of texture frequency. Depth accuracy is found to be 0.51.2% of the distance of the object from the imaging optics.     

Effect of poly(acrylic acid) on the preparation of thick silica films by electrophoretic sol–gel deposition of re-dispersed silica particles

Thick silica films were prepared by the electrophoretic sol–gel deposition technique in the presence of poly(acrylic acid) (PAA) using monodispersed silica particles; the particles were prepared by the sol–gel method, pre-heat treated and then re-dispersed in the mixture of H2O and ethanol. The weight of deposited silica films was maximized when 0.2 mass % of PAA against the whole amount of sol was added. The particles constructing the thick silica films were packed densely when the amount of added PAA was less than 0.2 mass%. The weight of the film increased with decrease in the content of H2O in the sol when a fixed amount of PAA was added. After the heat treatment of deposited films at 800 °C, crack-free silica films of about 30 m thickness were prepared. © 1998 Chapman & Hall     

Soft magnetic properties of NiFeTa thin films for biasing film applications in magnetoresistive elements

The effects of Ta addition on the magnetic properties of permalloy thin films have been investigated. The alloy compositions on a weight basis are (Ni₈₁Fe₁₉)_1-x Ta _x with 0 x 0.105, and the films are sputtered onto a glass substrate at between room temperature and 300 C. The saturation magnetization and anisotropic magnetic field decrease with increasing Ta content. The saturation magnetization is 0.75 T at 5 wt % Ta. The coercivity remains constant at 125 Am^-1. The electrical resistivity increases linearly with increasing Ta content, then saturates at approximately 7.5 wt% Ta. The saturation resistivity is approximately 1.00 m. The magnetoresistivity ratio (/) decreases with increasing Ta content, mainly due to increased electrical resistivity (). The magnetostriction changes from negative to positive with increasing Ta content and reaches nearly zero at 2 wt% Ta. The NiFeTa films containing 5–6 wt% Ta have potential for use as the soft-biasing film in magnetoresistive elements.