Pētniecība
Publikācijas
729 publikācijas
- 2019
Adaptive guard time for energy-efficient IEEE 802.15.4 TSCH networks(atveras jaunā logā)
Alexandros Mavromatis, Georgios Papadopoulos, Atis Elsts, Nicolas Montavont, Robert Piechocki, Theo Tryfonas, George Oikonomou, Xenofon Fafoutis
Lecture Notes in Computer Science, Sēj. 11618, lpp. 15 - 26, Springer
Kopsavilkums
Several Internet of Things (IoT) applications have strict performance requirements, in terms of reliability and power consumption. IEEE 802.15.4 Time Slotted Channel Hopping (TSCH) is a recently standardised Medium Access Control (MAC) protocol that supports these requirements by keeping the nodes time-synchronised. In order to ensure successful communication between a sender and a receiver, the latter starts listening shortly before the expected frame’s arrival. This time offset is called guard time and it aims to reduce the probability of missed frames due to clock drift. This paper investigates the impact of the guard time on the energy consumption and proposes a scheme for the decentralised adaptation of the guard time in each node depending on its hop-distance from the sink. Simulations and test-bed experiments demonstrate that guard time adaptation can reduce the energy consumption by up to 50%, without compromising the reliability of the network. © IFIP International Federation for Information Processing 2019.
- 2019
Experimental and numerical investigation of laboratory crystal growth furnace for the development of model-based control of CZ process(atveras jaunā logā)
M. Pudzs
Journal of Crystal Growth, Sēj. 522, lpp. 191 - 194, Elsevier B.V.
Kopsavilkums
M. Pudzs Experimental and numerical investigation of laboratory crystal growth furnace for the development of model-based control of CZ process
Q1/Q2 žurnāls · SNIP: 0.925
- 2019
Sampling and Estimating the Waveform of Analog Signals Based on Precision Event Timing(atveras jaunā logā)
A. S. Rybakov
Automatic Control and Computer Sciences, Sēj. 53 (2), lpp. 107–115, Pleiades Publishing
Kopsavilkums
In this paper, we investigate specific features of compressed sensing methods when estimating the waveform of arbitrary analog signals based on zero-crossings of a signal summed with a periodic oscillation. An algorithm for estimating the waveform of an analog signal with the adaptation of the basis matrix to its spectrum is proposed. To solve the problem of off-grid frequencies, we employ an iterative method for finding the maxima of the spectral peaks between the nodes of a DFT frequency grid. © 2019, Allerton Press, Inc.
Indeksēts Scopus / WoS · SNIP: 0.676
- 2019
Efficient Neural Network Implementations on Parallel Embedded Platforms applied to Real-Time Torque-Vectoring Optimization using Predictions for Multi-Motor Electric Vehicles(atveras jaunā logā)
Martin Dendaluce Jahnke, Francesco Cosco, Rihards Novickis, Joshué Pérez Rastelli, Vicente Gomez-Garay
Electronics (Switzerland), Sēj. 8 (2), lpp. 250, MDPI AG
Kopsavilkums
The combination of machine learning and heterogeneous embedded platforms enables new potential for developing sophisticated control concepts which are applicable to the field of vehicle dynamics and ADAS. This interdisciplinary work provides enabler solutions-ultimately implementing fast predictions using neural networks (NNs) on field programmable gate arrays (FPGAs) and graphical processing units (GPUs)-while applying them to a challenging application: Torque Vectoring on a multi-electric-motor vehicle for enhanced vehicle dynamics. The foundation motivating this work is provided by discussing multiple domains of the technological context as well as the constraints related to the automotive field, which contrast with the attractiveness of exploiting the capabilities of new embedded platforms to apply advanced control algorithms for complex control problems. In this particular case we target enhanced vehicle dynamics on a multi-motor electric vehicle benefiting from the greater degrees of freedom and controllability offered by such powertrains. Considering the constraints of the application and the implications of the selected multivariable optimization challenge, we propose a NN to provide batch predictions for real-time optimization. This leads to the major contribution of this work: efficient NN implementations on two intrinsically parallel embedded platforms, a GPU and a FPGA, following an analysis of theoretical and practical implications of their different operating paradigms, in order to efficiently harness their computing potential while gaining insight into their peculiarities. The achieved results exceed the expectations and additionally provide a representative illustration of the strengths and weaknesses of each kind of platform. Consequently, having shown the applicability of the proposed solutions, this work contributes valuable enablers also for further developments following similar fundamental principles. © 2019 by the authors. Licensee MDPI, Basel, Switzerland.
Indeksēts Scopus / WoS · Q1/Q2 žurnāls · SNIP: 0.985
- 2019
Bluetooth Low Energy Throughput in Densely Deployed Radio Environment(atveras jaunā logā)
Armands Ancans, Juris Ormanis, Ricards Cacurs, Modris Greitans, Elise Saoutieff, Adrien Faucon, Sébastien Boisseau
2019 23rd International Conference Electronics, IEEE
Kopsavilkums
Bluetooth Low Energy (BLE) is a promising 2.4 GHz technology for Body Area Networks (BAN) in health care and lifestyle applications. However, the global increase of wireless devices using the crowded spectrum in the 2.4 GHz frequency band can create coexistence issues. This work studies the performance of BLE in environments with multiple BLE devices. An experimental setup consisting of 10 BLE nodes is used to measure BLE application throughput with different connection parameters and under different interference sources, such as other BLE devices and WiFi. The results quantify the decrease of the application throughput and the influence of BLE connection parameters in the experimental settings, as well as suggest parameter values suitable for densely deployed environments.
Indeksēts Scopus / WoS
- 2019
Influence of Nonlinearity of Fast Sawtooth Voltage on Stroboscopic Transformation of Signals in Receivers of Sensors Locators(atveras jaunā logā)
V. Aristov
Automatic Control and Computer Sciences, Sēj. 53 (4), lpp. 291 - 297, Pleiades Publishing
Kopsavilkums
Sawtooth voltage with the possibility of changing its nonlinearity was modeled for a stroboscopic signal converter used in pulsed ultrawideband gigahertz sensor locators. The pattern of automatic shift in signal sample pulses is described mathematically, on the basis of which the formula of the law of shift in the aforementioned pulses at a nonlinear fast sawtooth voltage has been obtained. The effect of the nonlinearity of sawtooth voltage on the output signal of the stroboscopic converter is shown. © 2019, Allerton Press, Inc.
Indeksēts Scopus / WoS · SNIP: 0.676
- 2019
Generation of synthetic training data for object detection in piles(atveras jaunā logā)
E. Buls, R. kadikis, R. Cacurs, J. Arents
Proceedings of SPIE - The International Society for Optical Engineering, Sēj. 11041, lpp. 110411Z, SPIE
Kopsavilkums
Current state-of-the-art object detectors are based on supervised deep learning approaches. These methods require a large amount of annotated training data, which hinders a wider use of these methods in industry. We propose a method for generating synthetic training data for the task of detecting which objects in a pile can be picked up by a robot arm. The method requires few input images, which are used to create annotated images of piles. After training a state-of-theart detector on the synthetic data, we test it on real images. The results show that the model trained in such a way is not a rival to the best object detectors trained on large datasets of real images, but it is good for the specific task of detecting pickable objects in the piles. The main advantage of the proposed training approach is that the existing models can be easily re-trained to work with piles of different objects by personnel who do not specialize in machine learning
Indeksēts Scopus / WoS · SNIP: 0.261
- 2019
Modular Signal Processing Unit for Motion Control Applications Based on System-on-Chip with FPGA(atveras jaunā logā)
Vlastimil Šetka, Ondřej Ježek, Rihards Novickis
IEEE International Conference on Emerging Technologies and Factory Automation, lpp. 857 - 863, IEEE
Kopsavilkums
Motion control systems with distributed architecture where multiple input/output devices are connected to the upper layer controller by fast digital communication (fieldbus) became an industrial standard. This paper presents design of a modular input/output device which can process signals from multiple sensors, drive multiple actuators and act as a Slave or Master node in EtherCAT fieldbus network. User-defined algorithms can be easily implemented to preprocess input signals, combine multiple signals or close local control loops with extremely high sampling rates which makes the difference to standard off-the-shelf solutions. To meet these requirements and simplify hardware design, our device is based on System-on-Chip with both programmable logic (FPGA) and classic processor (CPU) ARM cores. Data processing including user algorithms can be done entirely in FPGA which provides very low latency and no jitter, and also on CPU for more complex computations with advantage of tight integration between FPGA and CPU. In this paper we provide description of hardware design, system architecture and typical applications.
Indeksēts Scopus / WoS
- 2019
MULTI-PARAMETRIC APPROACH FOR ASSESSMENT OF DETERIORATION INCONSTRUCTION MATERIALS USING ULTRASONIC SURFACE WAVES(atveras jaunā logā)
Alexey Tatarinov, Viktors Kurtenoks, Gennady Shahmenko, Evgeny Barkanov, Jelena Mironova
INTERNATIONAL CONGRESS ON ULTRASONICS, Acoustical Society of America
Kopsavilkums
As opposed to single-parametric ultrasonic pulse velocity measurements, a multi-parametric approach is proposed for assessment of the condition of constructional materials and early stages of its degradation. The method is based on acquisition of ultrasonic signals by stepwise profiling of the material surface at several frequencies in the range from 80 to 300 kHz. The analysis of the obtained 2D spatio-temporal waveform profiles allows us to extract case-specific parameters and features informative of deterioration. The parameters include group and phase velocities of head and surface Rayleigh waves, damping and dispersion of signals, frequency slope of attenuation, intensity distribution parameters in the spatio-temporal waveform profiles processed as images. Sensitivity of the method was confirmed in three use cases, including: a) surface quality in frost resistant concrete with rubber admixtures during freeze-thaw cycles; b) local structural deteriorations in refractory concrete subjected to high temperature; c) appearance of porosity in pultruded composite profiles due to inappropriate curing regimes.
- 2018
FCN and LSTM Based Computer Vision System for Recognition of Vehicle Type, License Plate Number, and Registration Country
N. Dorbe, A. Jaundalders, R. Kadikis, K. Nesenbergs.
Kopsavilkums
<p><img class=" wp-image-6932 aligncenter" src="/storage/wp/2019/09/FCN_LSTM_recognition_classification-300x66.png" alt="" width="514" height="113" /></p> <p>We propose an advanced Automatic number-plate recognition (ANPR) system, which not only recognizes the number and the issuing state, but also the type and location of the vehicle in the input image. The system is based on a combination of existing methods, modifications to neural network architectures and improvements in the training process. The proposed system uses machine-learning approach and consists of three main parts: segmentation of input image by Fully Convolutional Network for localization of license plate and determination of vehicle type; recognition of the characters of the localized plate by a Maxout CNN and LSTM; determination of the state that has issued the license plate by a CNN. The training of these neural network models is accomplished using a manually labeled custom dataset, which is expanded with data augmented techniques. The resulting system is capable of localizing and classifying multiple types of vehicles (including motorcycles and emergency vehicles) as well as their license plates. The achieved precision of the localization is 99.5%. The whole number recognition accuracy is 96.7% and character level recognition accuracy is 98.8%. The determination of the issuing state is precise in 92.8% cases.</p> <p><a href="https://link.springer.com/article/10.3103/S0146411618020104">Dorbe, N., Jaundalders, A., Kadikis, R., & Nesenbergs, K. (2018). FCN and LSTM Based Computer Vision System for Recognition of Vehicle Type, License Plate Number, and Registration Country. <i>Automatic Control and Computer Sciences</i>, <i>52</i>(2), 146-154.</a></p> <p><a href="https://makonis.edi.lv/s/tsE9bdpjg9MFCJ6">pdf</a></p> <p> </p>
- 2018
Scalable in-door positioning system for cooperative Micro Intelligent Vehicle algorithm development
I. Drikis, I. Ribners, J. Ormanis, K. Nesenbergs
Proceedings of the Biennial Baltic Electronics Conference, IEEE
Kopsavilkums
Even though cooperative driving shows great promise in improving road safety and travel efficiency, full scale experimental testing of related algorithms after simulation phase is prohibitively expensive. Micro Intelligent Vehicles (MicroIV) are commonly used as a more cost efficient, easier and faster way to develop innovations in this field. As GPS is a crucial component in autonomous driving scenarios but is not practical for use at MicroIV scales, alternative solutions are needed. This paper proposes a scalable, low-cost positioning system based on image processing, that can be used as a replacement for GPS in indoor facilities. © 2018 IEEE.
- 2018
Mathematical Description of the Operation of a Step-Recovery-Diode-Based Pulse Generator Circuit
V. Aristov
Kopsavilkums
<p>V. Aristov. Mathematical Description of the Operation of a Step-Recovery-Diode-Based Pulse Generator Circuit. Automatic Control and Computer Sciences, 2018, No. 6.<br /> The paper describes the operation of a classic step-recovery-diode-based pulse generator circuit using differential equations. The circuit has parasitic inductances and capacitors for both circuit elements and the diode in addition to standard discrete elements necessary for pumping the diode and generating a sharp edge. The results are compared with experimental data.</p>
- 2018
МАТЕМАТИЧЕСКОЕ ОПИСАНИЕ РАБОТЫ СХЕМЫ ФОРМИРОВАТЕЛЯ ИМПУЛЬСОВ НА ДИОДЕ С НАКОПЛЕНИЕМ ЗАРЯДА
В. Аристов
Kopsavilkums
<p>В. Аристов. МАТЕМАТИЧЕСКОЕ ОПИСАНИЕ РАБОТЫ СХЕМЫ ФОРМИРОВАТЕЛЯ ИМПУЛЬСОВ НА ДИОДЕ С НАКОПЛЕНИЕМ ЗАРЯДА. Рига, Автоматика и вычислительная техника. 2018 (6), стр. 98-108.<br /> С помощью дифференциальных уравнений описывается работа классичес-кой схемы формирователя крутого фронта импульсов на одном диоде с на-коплением заряда. Схема включает наряду со стандартными дискретными элементами, необходимыми для накачки диода и формирования крутого фронта, дополнительно паразитные индуктивности и конденсаторы, как элементов схемы, так и диода. Результаты сравниваются с эксперименталь-ными данными.</p>
- 2018
Increasing the Level of the Reflected Signal of a Broadband Pulse Locator by Changing the Pulse Duration of the Shock Excitation of a Broadband Antenna
Vladimir Aristov
Kopsavilkums
<p>Vladimir Aristov. "Increasing the Level of the Reflected Signal of a Broadband Pulse Locator by Changing the Pulse Duration of the Shock Excitation of a Broadband Antenna". European Journal of Advances in Engineering and Technology, 2018, 5(9): 717-721.<br /> Is investigated possibility increasing the level of the reflected signal at the set frequency at a choice of optimum duration of an impulse of excitement of locator with shock excitement of the antenna.</p>
- 2018
Differences in vertical ground reaction forces during first attempt of barefoot running in habitual shod runners
Abolins, V., Bernans, E., & Lanka, J
Kopsavilkums
<p>Abolins, V., Bernans, E., & Lanka, J. (2018). Differences in vertical ground reaction forces during first attempt of barefoot running in habitual shod runners. <i>Journal of Physical Education and Sport</i>, <i>18</i>(4), 2308-2313.</p>
- 2018
Differences in vertical ground reaction forces during first attempt of barefoot running in habitual shod runners
Abolins, V., Bernans, E., & Lanka, J.
Journal of Physical Education and Sport
Kopsavilkums
<p>Abolins, V., Bernans, E., & Lanka, J. (2018). Differences in vertical ground reaction forces during first attempt of barefoot running in habitual shod runners. <i>Journal of Physical Education and Sport</i>, <i>18</i>(4), 2308-2313.</p>
Q1/Q2 žurnāls · SNIP: 4
- 2018
Highly Reconfigurable Beamformer Stimulus Generator
Vaviļina, E., G. Gaigals
Kopsavilkums
<p><span style="color: #0000ff; font-family: Arial; font-size: small;"><span class="m_-8665640760123692966459473308-23082019"><span style="color: #3f3a38;"><span lang="EN-GB"><span style="font-family: Arial;">Vaviļina, E., G. Gaigals. "Highly Reconfigurable Beamformer Stimulus Generator." Latvian Journal of Physics and Technical Sciences 55.1, 53-63, 2018.</span></span></span></span></span></p>
- 2018
Towards Applying the Wireless Sensor Network Testbed for Studying Collaborative Work Environments based on Smart Spaces
Marchenkov S.A., Judvaitis J., Korzun D.G.
Kopsavilkums
<p>Marchenkov S.A., Judvaitis J., Korzun D.G., "Towards Applying the Wireless Sensor Network Testbed for Studying Collaborative Work Environments based on Smart Spaces", Материалы XII Всероссийской научно-практической конференции "Цифровые технологии в образовании, науке, обществе" (Петрозаводск, 4-6 декабря 2018). - Петрозаводск, 2018. - С. 157-161<br /> ISBN: 978-5-8021-3418-4</p> <p>All authors:</p> <p><span class="aCOpRe">Marchenkov S.A., Judvaitis J., Korzun D.G.</span></p> <p><strong>Abstract:</strong></p> <p>The wireless sensor network (WSN) testbed has been developed by the Institute the Institute of Electronics and Computer Science (EDI) (Riga, Latvia). The EDI WSN testbed provides a rich suit of data from sensorslocated inside and outside the institute building. Monitoring of the physical environment state and controlling its parameters are important for operation of collaborative work environments. In this work, the testbed is used for experimental study of such monitoring problems. We propose a software architecture design solution for integration and application of the EDI WSN testbed in studying collaborative work environments based on the smart spaces paradigm.</p> <p> </p>
- 2018
Modernization of Event Timer RTS 2006
V.Bespal’ko, K. Salmins, I. Buraks
Kopsavilkums
<p>Modernization of Event Timer RTS 2006. V.Bespal’ko, K. Salmins, I. Buraks 21st International Workshop on Laser Ranging, Canberra, Australia, November 04-09, 2018</p> <p><a href="https://cddis.nasa.gov/lw21/docs/2018/presentations/Session8_Salmins_presentation.pdf">https://cddis.nasa.gov/lw21/docs/2018/presentations/Session8_Salmins_presentation.pdf</a></p> <p> </p>
- 2018
Selection of informative hyperspectral band subsets based on entropy and correlation
A. Lorencs, I. Mednieks & J. Sinica-Sinavskis.
International Journal of Remote Sensing
Kopsavilkums
<p>A. Lorencs, I. Mednieks & J. Sinica-Sinavskis. Selection of informative hyperspectral band subsets based on entropy and correlation. International Journal of Remote Sensing, 39:20, pp. 6931-6948, 2018.</p> <p>DOI: <a href="https://doi.org/10.1080/01431161.2018.1468107" target="_blank" rel="noopener noreferrer">10.1080/01431161.2018.1468107</a></p> <p>Abstract<br /> The article proposes two novel and relatively simple unsupervised procedures for the selection of informative small subsets of spectral bands in hyperspectral images. To ensure the informativeness of the subsets, bands featuring higher entropy are included. The correlation of band images is restricted to avoid redundancy of the subsets. The entropy multiple correlation ratio procedure employs the entropy-correlation ratio for the selection of spectral bands. The entropy-based correlated band grouping (ECBG) procedure divides the spectrum into groups of bands featuring highly correlated images. The subsets obtained were characterized by the performance of classifiers using only data from included bands. The ECBG procedure provided better results than the alternatives if the number of selected bands was low. Another advantage of this procedure is the possibility of averaging the images obtained for spectral bands within the groups found. It is shown that classification results are significantly improved if such an averaging is used. In the data acquisition practice, it can be used for a purposeful merging of spectral bands in the configuration of hyperspectral imagers, which allows one to reduce the amount of data to be saved in real time and thus helps one to improve the achievable spatial resolution.</p>
Q1/Q2 žurnāls