Hartikainen J.,Rocsole Ltd |
Svensson L.,Chalmers University of Technology |
Journal of Advances in Information Fusion | Year: 2016
This article is concerned with Gaussian process quadratures, which are numerical integration methods based on Gaussian process regression methods, and sigma-point methods, which are used in advanced non-linear Kalman filtering and smoothing algorithms. We show that many sigma-point methods can be interpreted as Gaussian process quadrature based methods with suitably selected covariance functions. We show that this interpretation also extends to more general multivariate Gauss-Hermite integration methods and related spherical cubature rules. Additionally, we discuss different criteria for selecting the sigma-point locations: exactness of the integrals of multivariate polynomials up to a given order, minimum average error, and quasi-random point sets. The performance of the different methods is tested in numerical experiments.
Rahomaki J.,University of Eastern Finland |
Hyvarinen H.J.,Rocsole Ltd |
Rehman S.,National University of Singapore |
Rehman S.,Singapore MIT Alliance for Research and Technology |
Turunen J.,University of Eastern Finland
Nanoscale Research Letters | Year: 2013
We propose a scheme based on extraordinary transmission of light through a single nanoaperture, surrounded by periodic corrugations, for direct characterization of focal-region optical fields with subwavelength-scale structure. We describe the design of the device on the basis of rigorous diffraction theory and fabricate a prototype using a process that involves electron beam lithography, dry etching, and template stripping. First experimental results performed with a transmission-type confocal optical microscope demonstrate the potential of the method. © 2013 Rahomäki et al.
Oravisjarvi K.,University of Oulu |
Pietikainen M.,University of Oulu |
Ruuskanen J.,University of Eastern Finland |
Niemi S.,University of Vaasa |
And 6 more authors.
Journal of Aerosol Science | Year: 2014
Regional deposition of diesel particles in the human lungs was analyzed and the chemical composition of inhaled particles was investigated. The off-road diesel engine with a diesel particulate filter (DPF) or a selective catalytic reduction (SCR) unit and without any exhaust after-treatment system was used. Around 85-95% of the measured particles were of ultrafine size and 53-84% of those nanoparticles. Over 70% of the deposited particles under 0.1. μm and about 45-70% of the deposited particles from 0.1 to 1. μm reach also the alveolar-interstitial level. Elements analyzed in particles were C, O, Fe, Si, Ti, Na, K, Ca, Mg, Ba, Mn, Zn, Cu, Cl, P, S and N. The proportion of PAHs in the measured particle mass was 0.05% and carcinogenic ones represented 1.3% of the total PAHs. The DPF system removed particles efficiently and up to 99% of the particles were removed. The total number of particles deposited in the lungs was generally lower when DPF was used compared to other setups. These particles contained though the largest variety of elements, which are commonly considered harmful to humans. Therefore it is difficult to conclude, whether exhaust particles from a diesel engine with a DPF unit would be less harmful to human health. © 2013.
Nissinen A.,University of Eastern Finland |
Lehikoinen A.,Rocsole Ltd |
Mononen M.,Outotec Oyj |
Lahteenmaki S.,Inmet Pyhasalmi Mine Ltd |
Vauhkonen M.,University of Eastern Finland
Minerals Engineering | Year: 2014
Flotation process is widely used in mineral industry for the separation of valuable minerals from low-grade ore slurry. There are several parameters such as the bubble size and bubble loading that predict the efficiency of the flotation process. These parameters can be used for the control of the flotation process. There are already some techniques that can be used for online monitoring of these parameters, for example, the high-speed video imaging and a probe sensor based on electrical resistance tomography (ERT). These methods, however, suffer for some limitations. The high speed video imaging gives information only on the surface of the froth and in the previously proposed ERT based techniques the conductivity of the froth is typically modeled to be smoothly varying. However, in reality the froth is composed of different size of bubbles having highly conductive surface and very low conductive interior which configuration cannot be modeled with smoothly varying conductivity distribution. In this paper, we propose a computational approach in which the structure of the froth is modeled and both the bubble size and the conductivity of the boundary of the bubbles are estimated. The proposed approach utilizes data measured with the standard ERT probe. The estimated bubble size and conductivity of the boundary of the bubbles are compared to online measured camera based estimates of the bubble size and bubble loading. The proposed approach is evaluated with simulated measurements and real data from Pyhäsalmi Mine. The results show that there is a high correlation between the camera based and the ERT based estimates of the bubble size. Furthermore, some of the parameters obtained from the ERT based method correlate well with the camera based estimate of the bubble loading. © 2014 Elsevier Ltd. All rights reserved.
Sarkka S.,Aalto University |
Hartikainen J.,Rocsole Ltd.
IEEE International Workshop on Machine Learning for Signal Processing, MLSP | Year: 2013
We consider joint estimation of state and time-varying noise covariance matrices in non-linear stochastic state space models. We propose a variational Bayes and Gaussian non-linear filtering based algorithm for efficient computation of the approximate filtering posterior distributions. The formulation allows the use of efficient Gaussian integration methods such as unscented transform, cubature integration and Gauss-Hermite integration along with the classical Taylor series approximations. The performance of the algorithm is illustrated in a simulated application. © 2013 IEEE.
Sarkka S.,Aalto University |
Solin A.,Aalto University |
Hartikainen J.,Rocsole Ltd.
IEEE Signal Processing Magazine | Year: 2013
Gaussian process-based machine learning is a powerful Bayesian paradigm for nonparametric nonlinear regression and classification. In this article, we discuss connections of Gaussian process regression with Kalman filtering and present methods for converting spatiotemporal Gaussian process regression problems into infinite-dimensional state-space models. This formulation allows for use of computationally efficient infinite-dimensional Kalman filtering and smoothing methods, or more general Bayesian filtering and smoothing methods, which reduces the problematic cubic complexity of Gaussian process regression in the number of time steps into linear time complexity. The implication of this is that the use of machine-learning models in signal processing becomes computationally feasible, and it opens the possibility to combine machine-learning techniques with signal processing methods. © 1991-2012 IEEE.
Rocsole Ltd | Date: 2013-03-07
A method for investigating permittivity within a target domain comprises a step of measuring, for a plurality of pairs of contact element groups, the contact elements of the contact element groups being arranged in capacitance measurement connection with the target domain, an electrical quantity of interest proportional to the capacitance of a capacitor formed by a pair of contact element groups; the contact elements comprising a plurality of electrodes supported between the target domain and an electrically conductive background body limiting the capacitance measurement zone of the electrodes. The contact elements may further comprise the electrically conductive background body.
Rocsole Ltd | Date: 2013-02-01
A method for determining the location of an interface of interest in a target domain, between a free volume of a flowable material and a solid material limiting said free volume, the method involves the steps of providing a mathematical model of the target domain determining, for a plurality of pairs of electrode groups, a characteristic electrical quantity proportional to the capacitance of a capacitor formed by a pair of electrode groups; receiving measurements of the characteristic electrical quantity for a plurality of pairs of electrode groups; adjusting the mathematical model by varying the location of the boundary surface in order to take into account possible wear of the boundary surface so as to reduce the differences between the measured characteristic electrical quantities and those defined by the mathematical model; and determining the location of the interface of interest on the basis of the adjusted mathematical model.
Rocsole Ltd | Date: 2013-05-17
An arrangement (100) for monitoring scaling in a primary heat exchanger (1) comprises a secondary heat exchanger (101) and a scaling detecting apparatus (116, 117) installed to detect scaling in the secondary heat exchanger (101) as an indication of scaling in the primary heat exchanger (1).
News Article | December 14, 2016
KUOPIO, Finland, December 14, 2016 /PRNewswire/ -- Rocsole Ltd, a Finnish Company which specialises in real-time tomography, announces it has successfully completed an additional funding round which was led by Shell Technology Ventures (STV). The Finnish company, with offices in...