Change search
Link to record
Permanent link

Direct link
BETA
Publications (10 of 60) Show all publications
Sandin, F. & Nilsson, M. (2020). Synaptic Delays for Insect-Inspired Temporal Feature Detection in Dynamic Neuromorphic Processors. Frontiers in Neuroscience, 14, Article ID 150.
Open this publication in new window or tab >>Synaptic Delays for Insect-Inspired Temporal Feature Detection in Dynamic Neuromorphic Processors
2020 (English)In: Frontiers in Neuroscience, ISSN 1662-4548, E-ISSN 1662-453X, Vol. 14, article id 150Article in journal (Refereed) Published
Abstract [en]

Spiking neural networks are well suited for spatiotemporal feature detection and learning, and naturally involve dynamic delay mechanisms in the synapses, dendrites and axons. Dedicated delay neurons and axonal delay circuits have been considered when implementing such pattern recognition networks in dynamic neuromorphic processors. Inspired by an auditory feature detection circuit in crickets, featuring a delayed excitation by postinhibitory rebound, we investigate disynaptic delay elements formed by inhibitory--excitatory pairs of dynamic synapses. We configured such disynaptic delay elements in the DYNAP-SE neuromorphic processor and characterized the distribution of delayed excitations resulting from device mismatch. Interestingly, we found that the disynaptic delay elements can be configured such that the timing and magnitude of the delayed excitation depend mainly on the efficacy of the inhibitory and excitatory synapses, respectively, and that a neuron with multiple delay elements can be tuned to respond selectively to a specific pattern. Furthermore, we present a network with one disynaptic delay element that mimics the auditory feature detection circuit of crickets, and we demonstrate how varying synaptic weights, input noise and processor temperature affect the circuit. Dynamic delay elements of this kind open up for synapse level temporal feature tuning with configurable delays of up to 100 ms.

Place, publisher, year, edition, pages
Frontiers Media S.A., 2020
Keywords
pattern recognition, Spiking neural network (SNN), Neuromorphic, delay line, embedded intelligence, DYNAP, Insect-inspired computing
National Category
Other Electrical Engineering, Electronic Engineering, Information Engineering
Research subject
Electronic systems
Identifiers
urn:nbn:se:ltu:diva-77666 (URN)10.3389/fnins.2020.00150 (DOI)2-s2.0-85082714457 ()32180698 (PubMedID)2-s2.0-85082714457 (Scopus ID)
Funder
The Kempe Foundations, JCK-1809The Kempe Foundations, SMK-1429The Swedish Foundation for International Cooperation in Research and Higher Education (STINT), IG2011-2025
Note

Validerad;2020;Nivå 2;2020-04-14 (alebob)

Available from: 2020-02-13 Created: 2020-02-13 Last updated: 2020-04-22Bibliographically approved
Martin del Campo Barraza, S., Schnabel, S., Sandin, F. & Marklund, P. (2019). Detection of particle contaminants in rolling element bearings with unsupervised acoustic emission feature learning. Tribology International, 132, 30-38
Open this publication in new window or tab >>Detection of particle contaminants in rolling element bearings with unsupervised acoustic emission feature learning
2019 (English)In: Tribology International, ISSN 0301-679X, E-ISSN 1879-2464, Vol. 132, p. 30-38Article in journal (Refereed) Published
Abstract [en]

The detection of contaminants in the lubricant of rolling element bearings using acoustic emission signals is a challenging problem, in particular at high rotational speeds. This problem calls for new analysis methods beyond the conventional amplitude- and frequency-based methods. Feature learning is successfully used in the machine learning field to characterize complex signals. Here we use an unsupervised feature learning approach to distinguish acoustic emission signals. We investigate the repetition rates of features identified with shift-invariant dictionary learning and find that the signature of contaminated lubricant is significantly stronger than the effect on conventional condition indicators like the RMS and the enveloped RMS at rotational speeds above 300 rpm and up to 3000 rpm.

Place, publisher, year, edition, pages
Elsevier, 2019
Keywords
Acoustic emission, Contamination, Dictionary learning, Unsupervised feature learning
National Category
Tribology (Interacting Surfaces including Friction, Lubrication and Wear) Other Electrical Engineering, Electronic Engineering, Information Engineering
Research subject
Machine Elements; Industrial Electronics
Identifiers
urn:nbn:se:ltu:diva-63112 (URN)10.1016/j.triboint.2018.12.007 (DOI)000456758700004 ()2-s2.0-85058395512 (Scopus ID)
Note

Validerad;2019;Nivå 2;2019-01-03 (svasva)

Available from: 2017-04-21 Created: 2017-04-21 Last updated: 2019-05-14Bibliographically approved
Dadhich, S., Sandin, F., Bodin, U., Andersson, U. & Martinsson, T. (2019). Field test of neural-network based automatic bucket-filling algorithm for wheel-loaders. Automation in Construction, 97, 1-12
Open this publication in new window or tab >>Field test of neural-network based automatic bucket-filling algorithm for wheel-loaders
Show others...
2019 (English)In: Automation in Construction, ISSN 0926-5805, E-ISSN 1872-7891, Vol. 97, p. 1-12Article in journal (Refereed) Published
Abstract [en]

Automation of earth-moving industries (construction, mining and quarry) require automatic bucket-filling algorithms for efficient operation of front-end loaders. Autonomous bucket-filling is an open problem since three decades due to difficulties in developing useful earth models (soil, gravel and rock) for automatic control. Operators make use of vision, sound and vestibular feedback to perform the bucket-filling operation with high productivity and fuel efficiency. In this paper, field experiments with a small time-delayed neural network (TDNN) implemented in the bucket control-loop of a Volvo L180H front-end loader filling medium coarse gravel are presented. The total delay time parameter of the TDNN is found to be an important hyperparameter due to the variable delay present in the hydraulics of the wheel-loader. The TDNN network successfully performs the bucket-filling operation after an initial period (100 examples) of imitation learning from an expert operator. The demonstrated solution show only 26% longer bucket-filling time, an improvement over manual tele-operation performance.

Place, publisher, year, edition, pages
Elsevier, 2019
Keywords
Neural-network, Bucket-filling, Wheel-loader, Automation, Construction
National Category
Other Electrical Engineering, Electronic Engineering, Information Engineering Media and Communication Technology
Research subject
Industrial Electronics; Pervasive Mobile Computing
Identifiers
urn:nbn:se:ltu:diva-71383 (URN)10.1016/j.autcon.2018.10.013 (DOI)000453623600001 ()2-s2.0-85055696994 (Scopus ID)
Note

Validerad;2018;Nivå 2;2018-11-07 (johcin) 

Available from: 2018-10-30 Created: 2018-10-30 Last updated: 2019-01-30Bibliographically approved
Nilsson, J., Sandin, F. & Delsing, J. (2019). Interoperability and machine-to-machine translation model with mappings to machine learning tasks. In: Proceedings: 2019 IEEE 17th International Conference on Industrial Informatics (INDIN). Paper presented at 2019 IEEE 17th International Conference on Industrial Informatics (INDIN), 22-25 July, 2019, Helsinki, Finland (pp. 284-289). IEEE
Open this publication in new window or tab >>Interoperability and machine-to-machine translation model with mappings to machine learning tasks
2019 (English)In: Proceedings: 2019 IEEE 17th International Conference on Industrial Informatics (INDIN), IEEE, 2019, p. 284-289Conference paper, Published paper (Other academic)
Abstract [en]

Modern large-scale automation systems integrate thousands to hundreds of thousands of physical sensors and actuators. Demands for more flexible reconfiguration of production systems and optimization across different information models, standards and legacy systems challenge current system interoperability concepts. Automatic semantic translation across information models and standards is an increasingly important problem that needs to be addressed to fulfill these demands in a cost-efficient manner under constraints of human capacity and resources in relation to timing requirements and system complexity. Here we define a translator-based operational interoperability model for interacting cyber-physical systems in mathematical terms, which includes system identification and ontology-based translation as special cases. We present alternative mathematical definitions of the translator learning task and mappings to similar machine learning tasks and solutions based on recent developments in machine learning. Possibilities to learn translators between artefacts without a common physical context, for example in simulations of digital twins and across layers of the automation pyramid are briefly discussed.

Place, publisher, year, edition, pages
IEEE, 2019
Series
IEEE International Conference on Industrial Informatics (INDIN), ISSN 1935-4576, E-ISSN 2378-363X
National Category
Other Electrical Engineering, Electronic Engineering, Information Engineering
Research subject
Electronic systems
Identifiers
urn:nbn:se:ltu:diva-73562 (URN)10.1109/INDIN41052.2019.8972085 (DOI)
Conference
2019 IEEE 17th International Conference on Industrial Informatics (INDIN), 22-25 July, 2019, Helsinki, Finland
Funder
EU, Horizon 2020, 737459
Note

ISBN för värdpublikation: 978-1-7281-2928-0, 978-1-7281-2927-3

Available from: 2019-04-10 Created: 2019-04-10 Last updated: 2020-04-24Bibliographically approved
Nilsson, J., Sandin, F. & Delsing, J. (2019). Interoperability automation considered as machine learning tasks. In: : . Paper presented at 2nd Productive4.0 Consortium Conference, Budapest, March 12-14, 2019.
Open this publication in new window or tab >>Interoperability automation considered as machine learning tasks
2019 (English)Conference paper, Poster (with or without abstract) (Other academic)
Keywords
Interoperability, machine learning, optimization, translation, semantics
National Category
Other Computer and Information Science
Research subject
Industrial Electronics
Identifiers
urn:nbn:se:ltu:diva-73578 (URN)
Conference
2nd Productive4.0 Consortium Conference, Budapest, March 12-14, 2019
Funder
EU, Horizon 2020, 737459
Available from: 2019-04-11 Created: 2019-04-11 Last updated: 2019-09-06
Martin del Campo Barraza, S., Sandin, F. & Schnabel, S. (2019). Kinematic Frequencies of Rotating Equipment Identified with Sparse Coding and Dictionary Learning. In: N. Scott Clements (Ed.), Proceedings of the Annual Conference of the Prognostics and Health Management Society 2019: . Paper presented at Annual Conference of the Prognostics and Health Management Society, 23-26 September, 2019, Scottsdale, Arizona, USA. Scottsdale, AZ, USA: Prognostics and Health Management Society
Open this publication in new window or tab >>Kinematic Frequencies of Rotating Equipment Identified with Sparse Coding and Dictionary Learning
2019 (English)In: Proceedings of the Annual Conference of the Prognostics and Health Management Society 2019 / [ed] N. Scott Clements, Scottsdale, AZ, USA: Prognostics and Health Management Society , 2019Conference paper, Published paper (Refereed)
Abstract [en]

The detection of faults and operational abnormalities in rotating machine elements like rolling element bearings and gears requires information about kinematic properties, such as ball-pass and gear mesh frequencies. Typically, condition monitoring experts obtain such information from the manufacturers for diagnostics purposes. However, the reliability of such information can be compromised during installation and maintenance, for example, if components are replaced and do not match the documented specifications. Thus, methods enabling verification and online extraction of such kinematic properties are needed to improve diagnostic reliability. Unsupervised machine learning methods, like sparse coding with dictionary learning, enable automatic modeling and characterization of repeating signal structures in the time domain, which are naturally generated by rotating equipment. Sparse coding with dictionary learning represents a vibration signal as a linear superposition of noise and atomic waveforms. The activation rate of the atomic waveforms typically possesses a cyclic nature in rotating environments, similar to how bearing kinematic frequencies correlate with faults in a rolling element bearing. However, there is no explicit relationship between the activation rates of the atoms and the bearing kinematic frequencies. This motivates this investigation of the possibility to extract bearing kinematic frequencies from sparse representations. Former work describes the use of dictionary learning for the detection of anomalies in rolling element bearings. In this paper, we describe how a similar unsupervised machine learning method can be used to extract kinematic frequencies of bearings and gears, for example for anomaly detection purposes and comparisons with an expected signature. We study the activation rates and changes of atoms learned from vibration signals in two case studies. The first case is based on data from a well-known controlled experiment with faults seeded in the bearings. The second case is based on a public dataset recorded from the high-speed shaft of a wind turbine with a bearing failure. Furthermore, we compare the activation rates and weights of the atoms to the bearing kinematic frequencies and harmonics. Sparse coding with dictionary learning offers a possibility for self-learningof the kinematic frequencies of a bearing, which can be useful for the further improvement of automated anomaly detection methods in condition monitoring.

Place, publisher, year, edition, pages
Scottsdale, AZ, USA: Prognostics and Health Management Society, 2019
Series
Proceedings of the Annual Conference of the Prognostics and Health Management Society, ISSN 2325-0178 ; 11(1)
Keywords
sparse coding, dictionary learning, condition monitoring, wind turbine, bearings
National Category
Other Electrical Engineering, Electronic Engineering, Information Engineering
Research subject
Electronic systems; Industrial Electronics
Identifiers
urn:nbn:se:ltu:diva-76337 (URN)10.36001/phmconf.2019.v11i1.837 (DOI)2-s2.0-85083979615 (Scopus ID)
Conference
Annual Conference of the Prognostics and Health Management Society, 23-26 September, 2019, Scottsdale, Arizona, USA
Note

ISBN för värdpublikation: 978-1-936263-29-5

Available from: 2019-10-09 Created: 2019-10-09 Last updated: 2020-05-07Bibliographically approved
Nilsson, M. & Sandin, F. (2019). Spatiotemporal Pattern Recognition with Neuromorphic Processor for Edge Applications. In: Online Proceedings of SAIS 2019: . Paper presented at The 31st Annual Workshop of the Swedish Artificial Intelligence Society (SAIS 2019).
Open this publication in new window or tab >>Spatiotemporal Pattern Recognition with Neuromorphic Processor for Edge Applications
2019 (English)In: Online Proceedings of SAIS 2019, 2019Conference paper, Oral presentation with published abstract (Refereed)
Keywords
neuromorphic, spiking neural network, pattern recognition, edge computing, delay line, DYNAP
National Category
Other Electrical Engineering, Electronic Engineering, Information Engineering
Research subject
Electronic systems
Identifiers
urn:nbn:se:ltu:diva-77668 (URN)
Conference
The 31st Annual Workshop of the Swedish Artificial Intelligence Society (SAIS 2019)
Funder
The Kempe Foundations, JCK-1809
Available from: 2020-02-07 Created: 2020-02-07 Last updated: 2020-04-28
Martin del Campo Barraza, S., Sandin, F. & Strömbergsson, D. (2018). Dataset concerning the vibration signals from wind turbines in northern Sweden.
Open this publication in new window or tab >>Dataset concerning the vibration signals from wind turbines in northern Sweden
2018 (English)Data set, Primary data
Alternative title[en]
Dataset of A dictionary learning approach to monitoring of wind turbine drivetrain bearings
Abstract [en]

In the manuscript, we investigate condition monitoring methods based on unsupervised dictionary learning.

The dataset includes the raw time-domain vibration signals from six turbines within the same wind farm (near geographical location). All the wind turbines are of the same type and possess a three-stage gearbox. All measurement data corresponds to the axial direction of an accelerometer mounted on the housing of the output shaft bearing of each turbine. The sampling rate is 12.8 kilosamples/second and each signal segment is 1.28 seconds long (16384 samples).

Keywords
dataset, wind turbine, condition monitoring
National Category
Other Electrical Engineering, Electronic Engineering, Information Engineering Tribology (Interacting Surfaces including Friction, Lubrication and Wear)
Research subject
Industrial Electronics; Machine Elements
Identifiers
urn:nbn:se:ltu:diva-70730 (URN)
Available from: 2018-09-03 Created: 2018-09-03 Last updated: 2019-05-14Bibliographically approved
Dadhich, S., Bodin, U., Sandin, F. & Andersson, U. (2018). From Tele-remote Operation to Semi-automated Wheel-loader. International Journal of Electrical and Electronic Engineering and Telecommunications, 7(4), 178-182
Open this publication in new window or tab >>From Tele-remote Operation to Semi-automated Wheel-loader
2018 (English)In: International Journal of Electrical and Electronic Engineering and Telecommunications, ISSN 2319-2518, Vol. 7, no 4, p. 178-182Article in journal (Refereed) Published
Abstract [en]

This paper presents experimental results with tele-remote operation of a wheel-loader and proposes a method to semi-automate the process. The different components of the tele-remote setup are described in the paper. We focus on the short loading cycle, which is commonly used at quarry and construction sites for moving gravel from piles onto trucks. We present results from short-loading-cycle experiments with three operators, comparing productivity between tele-remote operation and manual operation. A productivity loss of 42% with tele-remote operation motivates the case for more automation. We propose a method to automate the bucket-filling process, which is one of the key operations performed by a wheel-loader.

Keywords
automation, bucket-filling, construction, quarry, tele-operation, wheel-loader
National Category
Other Electrical Engineering, Electronic Engineering, Information Engineering Media and Communication Technology
Research subject
Industrial Electronics; Pervasive Mobile Computing
Identifiers
urn:nbn:se:ltu:diva-71381 (URN)10.18178/ijeetc.7.4.178-182 (DOI)
Available from: 2018-10-30 Created: 2018-10-30 Last updated: 2018-11-23Bibliographically approved
Dadhich, S., Sandin, F. & Bodin, U. (2018). Predicting bucket-filling control actions of a wheel-loader operator using aneural network ensemble. In: 2018 International Joint Conference on Neural Networks (IJCNN): . Paper presented at 2018 International Joint Conference on Neural Networks (IJCNN);8-13 July 2018;Rio de Janeiro, Brazil. Piscataway, NJ: IEEE, Article ID 8489388.
Open this publication in new window or tab >>Predicting bucket-filling control actions of a wheel-loader operator using aneural network ensemble
2018 (English)In: 2018 International Joint Conference on Neural Networks (IJCNN), Piscataway, NJ: IEEE, 2018, article id 8489388Conference paper, Published paper (Refereed)
Abstract [en]

Automatic bucket filling is an open problem since three decades. In this paper, we address this problem with supervised machine learning using data collected from manual operation. The range-normalized actuations of lift joystick, tilt joystick and throttle pedal are predicted using information from sensors on the machine and the prediction errors are quantified. We apply linear regression, k-nearest neighbors, neural networks, regression trees and ensemble methods and find that an ensemble of neural networks results in the most accurate predictions. The prediction root-mean-square-error (RMSE) of the lift action exceeds that of the tilt and throttle actions, and we obtain an RMSE below 0.2 for complete bucket fillings after training with as little as 135 bucket filling examples

Place, publisher, year, edition, pages
Piscataway, NJ: IEEE, 2018
Series
Proceedings of the International Joint Conference on Neural Networks, E-ISSN 2161-4407
National Category
Media and Communication Technology Other Electrical Engineering, Electronic Engineering, Information Engineering
Research subject
Industrial Electronics; Pervasive Mobile Computing
Identifiers
urn:nbn:se:ltu:diva-71382 (URN)10.1109/IJCNN.2018.8489388 (DOI)2-s2.0-85055724313 (Scopus ID)978-1-5090-6014-6 (ISBN)
Conference
2018 International Joint Conference on Neural Networks (IJCNN);8-13 July 2018;Rio de Janeiro, Brazil
Available from: 2018-10-30 Created: 2018-10-30 Last updated: 2019-02-11Bibliographically approved
Organisations
Identifiers
ORCID iD: ORCID iD iconorcid.org/0000-0001-5662-825x

Search in DiVA

Show all publications