Ohno H.,Toyota Central R and D Labs. Inc.
Applied Soft Computing Journal | Year: 2016
Materials informatics is a growing field in materials science. Materials scientists have begun to use soft computing techniques to discover novel materials. In order to apply these techniques, the descriptors (referred to as features in computer science) of a material must be selected, thereby deciding the resulting performance. As a way of describing a material, the properties of each element in the material are used directly as the features of the input variable. Depending on the number of elements in the material, the dimensionality of the input may differ. Hence, it is not possible to apply the same model to materials with different numbers of elements for tasks such as regression or discrimination. In the present paper, we present a novel method of uniforming the dimensionality of the input that allows regression or discriminative tasks to be performed using soft computing techniques. The main contribution of the proposed method is to provide a solution for uniforming the dimensionality among input vectors of different size. The proposed method is a variant of the denoising autoencoder Vincent et al. (2008)  using neural networks and gives a latent representation with uniformed dimensionality of the input. In the experiments of the present study, we consider compounds with ionic conductivity and hydrogen storage materials. The results of the experiments indicate that the regression tasks can be performed using the uniformed latent data learned by the proposed method. Moreover, in the clustering task using these latent data, we observed distance preservation in data space, which is also the case for the denoising autoencoder. This result may enable the proposed method to be used in a broad range of applications. © 2016 Elsevier B.V. All rights reserved.
Nakamura D.,Toyota Central R and D Labs. Inc.
Applied Physics Express | Year: 2016
Newly developed highly reliable low-cost TaC-coated graphite materials prepared by a wet ceramic process were applied to SiC sublimation growth. We demonstrated an increased long-duration growth rate and a resultant increase in crystal size by a factor of ~1.2 (experimental value) after 24 h of growth [and ~1.5 (extrapolated value) after the optimum duration of 53.1 h] by simply and quickly replacing graphite crucibles with TaC-coated graphite crucibles. Growth with the TaC-coated graphite crucibles reduced source gas leakage and increased the material yield for single crystals because the TaC layers were gas-tight and had a low emissivity. © 2016 The Japan Society of Applied Physics.
Kudo K.,Toyota Central R and D Labs. Inc. |
Morimoto Y.,Toyota Central R and D Labs. Inc.
ECS Transactions | Year: 2012
To clarify the rate determining process for the oxygen transport in a catalyst layer of polymer electrolyte fuel cells, the oxygen transport resistance of a Nafion thin film was precisely evaluated using a microelectrode technique for cast Nafion films with 30-150 nm thickness and data analysis on the relationship between the Nafion thickness on the Pt electrode and the diffusion-limited current density on the electrode. While the oxygen transport resistivity inside the Nafion thin film was roughly equal to that of Nafion thick membrane, the interfacial resistance at the Pt/Nafion or Nafion/gas interface was estimated to be equivalent to 60-80 nm thickness of Nafion film and to account for the dominant part of the total oxygen transport loss in a real PEFC electrode, especially with a low platinum loading. © The Electrochemical Society.
Matsushita Y.,Nagoya University |
Hayashida K.,Toyota Central R and D Labs. Inc. |
Takano A.,Nagoya University
Macromolecular Rapid Communications | Year: 2010
Among several star molecules studied so far, the results for two similar but slightly different systems composed of all hydrophobic components, i.e. one low Tg arm and two high Tg arms were compared and discussed mainly. They are polyisoprene-arm-polystyrene-arm-poly(2- vinylpyridine) stars and polystyrene-arm-polybutadiene-arm-poly(2-vinylpyridine) ones, and are abbreviated as ISP and SBV, respectively, in the original literatures. Several periodic Archimedean tiling patterns can be naturally formed when the relative lengths of three chains are similar one another in both series. Core-shell type morphologies and actually three-phase four layer lamellar structures were also commonly observed for two series. A quasicrystalline tiling with dodecagonal symmetry can also be conformed at the targeted composition of ISP/homopolymer blend and a zinc-blende type structure was created at just a little outside of tiling region for the same blend. Furthermore, the several interesting structures from amphiphilic molecules, composed of a hydrophilic component, poly(ethylene oxide), a hydrophobic component, polymethylmethacrylate and a highly waterrepellent polymer poly(perfluoropropyleneoxide) are described and introduced. © 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim.
Yamanaka K.-I.,Toyota Central R and D Labs. Inc. |
Morikawa T.,Toyota Central R and D Labs. Inc.
Journal of Physical Chemistry C | Year: 2012
Nitrogen-doped TiO 2 (N-TiO 2) is one of the most promising and widely investigated visible-light-responsive semiconductor photocatalysts. Femtosecond time-resolved diffuse reflectance (TDR) spectroscopy was employed under weak excitation conditions to clarify the charge-separation and trapping dynamics in N-TiO 2 powder, which exhibits high activity for photocatalytic oxidation under visible light. TDR spectra of N-TiO 2 after 360 nm light excitation (Ti 3d ← O 2p transition) revealed that the surface-trapped electrons and holes were generated immediately after excitation, similar to that for TiO 2, and the population of surface-trapped electrons decreased more than that for TiO 2 due to deep trapping by additionally induced oxygen vacancies. The dependence of trapping dynamics on the N concentration is discussed with respect to the diffusion coefficient of electrons. TDR spectra of N-TiO 2 after 450 nm light excitation (Ti 3d ← N 2p transition) clearly indicated the generation of charge carriers. Compared with the 360 nm excitation, two differences of time evolution were detected: the significant decrease just after excitation and the deep trapping of electrons within 1 ps (time resolution). Possible mechanisms for the dependence of the charge-carrier dynamics in N-TiO 2 on the excitation wavelength are discussed. © 2011 American Chemical Society.
Nakano H.,Toyota Central R and D Labs. Inc.
Nippon Seramikkusu Kyokai Gakujutsu Ronbunshi/Journal of the Ceramic Society of Japan | Year: 2014
Soft chemical synthesis of silicon nanosheets (SiNSs) and functionalized modified SiNSs are reviewed. Free standing SiNSs with sub-nanometer thicknesses have been prepared by exfoliating layered silicon compounds, and they are found to be composed of crystalline single-atom-thick silicon layers. SiNSs are new silicon nanomaterials, as are silicon nanoparticles and nanowires. Organic modified SiNS can be prepared by modifying a layered polysilane, which has an analogous structure to that of graphite, and this allows the properties of the SiNS to be controlled in order to make it suitable for particular applications. The potential applications of these SiNSs and organic modified SiNSs are also reviewed. © 2014 The Ceramic Society of Japan. All rights reserved.
Yamaguchi K.,TTI Chicago |
Yamaguchi K.,Toyota Central R and D Labs. Inc. |
McAllester D.,TTI Chicago |
Urtasun R.,TTI Chicago
Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition | Year: 2013
We consider the problem of computing optical flow in monocular video taken from a moving vehicle. In this setting, the vast majority of image flow is due to the vehicle's ego-motion. We propose to take advantage of this fact and estimate flow along the epipolar lines of the ego-motion. Towards this goal, we derive a slanted-plane MRF model which explicitly reasons about the ordering of planes and their physical validity at junctions. Furthermore, we present a bottom-up grouping algorithm which produces over-segmentations that respect flow boundaries. We demonstrate the effectiveness of our approach in the challenging KITTI flow benchmark achieving half the error of the best competing general flow algorithm and one third of the error of the best epipolar flow algorithm. © 2013 IEEE.
Yoshida H.,Toyota Central R and D Labs. Inc. |
Nagaoka M.,Toyota Central R and D Labs. Inc.
Journal of Computational Physics | Year: 2010
A lattice Boltzmann model with a multiple-relaxation-time (MRT) collision operator for the convection-diffusion equation is presented. The model uses seven discrete velocities in three dimensions (D3Q7 model). The off-diagonal components of the relaxation-time matrix, which originate from the rotation of the principal axes, enable us to take into account full anisotropy of diffusion. An asymptotic analysis of the model equation with boundary rules for the Dirichlet and Neumann-type (specified flux) conditions is carried out to show that the model is first- and second-order accurate in time and space, respectively. The results of the analysis are verified by several numerical examples. It is also shown numerically that the error of the MRT model is less sensitive to the variation of the relaxation-time coefficients than that of the classical BGK model. In addition, an alternative treatment for the Neumann-type boundary condition that improves the accuracy on a curved boundary is presented along with a numerical example of a spherical boundary. © 2010 Elsevier Inc.
Niclass C.,Toyota Central R and D Labs Inc. |
Soga M.,Toyota Central R and D Labs Inc. |
Matsubara H.,Toyota Central R and D Labs Inc. |
Kato S.,Toyota Central R and D Labs Inc. |
Kagami M.,Toyota Central R and D Labs Inc.
IEEE Journal of Solid-State Circuits | Year: 2013
This paper introduces a single-photon detection technique for time-of-flight distance ranging based on the temporal and spatial correlation of photons. A proof-of-concept prototype achieving depth imaging up to 100 meters with a resolution of 340×, 96 pixels at 10 frames/s was implemented. At the core of the system, a sensor chip comprising 32 macro-pixels based on an array of single-photon avalanche diodes featuring an optical fill factor of 70% was fabricated in a 0.18-μm CMOS. The chip also comprises an array of 32 circuits capable of generating precise triggers upon correlation events as well as of sampling the number of photons involved in each correlation event, and an array of 32 12-b time-to-digital converters. Characterization of the TDC array led to-0.52 LSB and 0.73 LSB of differential and integral nonlinearities, respectively. Quantitative evaluation of the TOF sensor under strong solar background light, i.e., 80 klux, revealed a repeatability error better than 10 cm throughout the distance range of 100 m, thus leading to a relative precision of 0.1%. In the same condition, the relative nonlinearity error was 0.37%. In order to show the suitability of our approach in a real-world situation, experimental results in which the depth sensor was operated in a typical traffic scenario are also reported. © 1966-2012 IEEE.
Takeda H.,Toyota Central R and D Labs. Inc. |
Takeda H.,Japan Science and Technology Agency |
Ishitani O.,Tokyo Institute of Technology |
Ishitani O.,Japan Science and Technology Agency
Coordination Chemistry Reviews | Year: 2010
Photocatalytic systems for CO2 reduction using metal complexes, especially rhenium(I) complexes as a main component, are reviewed: mononuclear Re(I) complexes, mixed systems with two different Re(I) complexes, and supramolecule systems with a Re(I) complex connected to a ruthenium(II) complex. We focus on the mechanistic studies and the architecture for constructing the photocatalytic systems based on the mechanism. © 2009 Elsevier B.V. All rights reserved.