To Örebro University

oru.seÖrebro University Publications
Change search
Link to record
Permanent link

Direct link
Winkler, Nicolas P.
Publications (9 of 9) Show all publications
Winkler, N. P., Neumann, P. P., Schaffernicht, E. & Lilienthal, A. J. (2024). Gas Distribution Mapping With Radius-Based, Bi-directional Graph Neural Networks (RABI-GNN). In: 2024 IEEE International Symposium on Olfaction and Electronic Nose (ISOEN): . Paper presented at International Symposium on Olfaction and Electronic Nose (ISOEN 2024), Grapevine, TX, USA, May 12-15, 2024. IEEE
Open this publication in new window or tab >>Gas Distribution Mapping With Radius-Based, Bi-directional Graph Neural Networks (RABI-GNN)
2024 (English)In: 2024 IEEE International Symposium on Olfaction and Electronic Nose (ISOEN), IEEE , 2024Conference paper, Published paper (Refereed)
Abstract [en]

Gas Distribution Mapping (GDM) is essential in monitoring hazardous environments, where uneven sampling and spatial sparsity of data present significant challenges. Traditional methods for GDM often fall short in accuracy and expressiveness. Modern learning-based approaches employing Convolutional Neural Networks (CNNs) require regular-sized input data, limiting their adaptability to irregular and sparse datasets typically encountered in GDM. This study addresses these shortcomings by showcasing Graph Neural Networks (GNNs) for learningbased GDM on irregular and spatially sparse sensor data. Our Radius-Based, Bi-Directionally connected GNN (RABI-GNN) was trained on a synthetic gas distribution dataset on which it outperforms our previous CNN-based model while overcoming its constraints. We demonstrate the flexibility of RABI-GNN by applying it to real-world data obtained in an industrial steel factory, highlighting promising opportunities for more accurate GDM models.

Place, publisher, year, edition, pages
IEEE, 2024
National Category
Computer Sciences
Identifiers
urn:nbn:se:oru:diva-115645 (URN)10.1109/ISOEN61239.2024.10556309 (DOI)001259381600051 ()2-s2.0-85197434833 (Scopus ID)9798350348668 (ISBN)9798350348651 (ISBN)
Conference
International Symposium on Olfaction and Electronic Nose (ISOEN 2024), Grapevine, TX, USA, May 12-15, 2024
Available from: 2024-08-27 Created: 2024-08-27 Last updated: 2024-08-27Bibliographically approved
Winkler, N. P., Kotlyar, O., Schaffernicht, E., Matsukura, H., Ishida, H., Neumann, P. P. & Lilienthal, A. J. (2024). Super-resolution for Gas Distribution Mapping. Sensors and actuators. B, Chemical, 419, Article ID 136267.
Open this publication in new window or tab >>Super-resolution for Gas Distribution Mapping
Show others...
2024 (English)In: Sensors and actuators. B, Chemical, ISSN 0925-4005, E-ISSN 1873-3077, Vol. 419, article id 136267Article in journal (Refereed) Published
Abstract [en]

Gas Distribution Mapping (GDM) is a valuable tool for monitoring the distribution of gases in a wide range of applications, including environmental monitoring, emergency response, and industrial safety. While GDM is actively researched in the scope of gas-sensitive mobile robots (Mobile Robot Olfaction), there is a potential for broader applications utilizing sensor networks. This study aims to address the lack of deep learning approaches in GDM and explore their potential for improved mapping of gas distributions. In this paper, we introduce Gas Distribution Decoder (GDD), a learning-based GDM method. GDD is a deep neural network for spatial interpolation between sparsely distributed sensor measurements that was trained on an extensive data set of realistic-shaped synthetic gas plumes based on actual airflow measurements. As access to ground truth representations of gas distributions remains a challenge in GDM research, we make our data sets, along with our models, publicly available. We test and compare GDD with state-of-the-art models on synthetic and real- world data. Our findings demonstrate that GDD significantly outperforms existing models, demonstrating a 35% improvement in accuracy on synthetic data when measured using the Root Mean Squared Error over the entire distribution map. Notably, GDD appears to have superior capabilities in reconstructing the edges and characteristic shapes of gas plumes compared to traditional models. These potentials offer new possibilities for more accurate and efficient environmental monitoring, and we hope to inspire other researchers to explore learning-based GDM.

Place, publisher, year, edition, pages
Elsevier, 2024
Keywords
Gas Distribution Mapping, Spatiotemporal interpolation, Mobile Robot Olfaction, Sensor network, Deep learning
National Category
Computer Sciences
Identifiers
urn:nbn:se:oru:diva-115514 (URN)10.1016/j.snb.2024.136267 (DOI)001292229400001 ()2-s2.0-85200765573 (Scopus ID)
Available from: 2024-08-21 Created: 2024-08-21 Last updated: 2024-08-21Bibliographically approved
Winkler, N. P., Neumann, P. P., Schaffernicht, E., Lilienthal, A., Poikkimäki, M., Kangas, A. & Säämänen, A. (2022). Gather Dust and Get Dusted: Long-Term Drift and Cleaning of Sharp GP2Y1010AU0F Dust Sensor in a Steel Factory. In: : . Paper presented at 38th Danubia-Adria Symposium on Advances in Experimental Mechanics, Poros Island, Greece, September 20-23, 2022.
Open this publication in new window or tab >>Gather Dust and Get Dusted: Long-Term Drift and Cleaning of Sharp GP2Y1010AU0F Dust Sensor in a Steel Factory
Show others...
2022 (English)Conference paper, Published paper (Refereed)
Abstract [en]

The Sharp GP2Y1010AU0F is a widely used low-cost dust sensor, but despite its popularity, the manufacturer provides little information on the sensor. We installed 16 sensing nodes with Sharp dust sensors in a hot rolling mill of a steel factory. Our analysis shows a clear correlation between sensor drift and accumulated production of the steel factory. An eye should be kept on the long-term drift of the sensors to prevent early saturation. Two of 16 sensors experienced full saturation, each after around eight and ten months of operation.

Keywords
Dust sensor, Low-cost, Sensor drift, Sensor network
National Category
Signal Processing
Identifiers
urn:nbn:se:oru:diva-102768 (URN)
Conference
38th Danubia-Adria Symposium on Advances in Experimental Mechanics, Poros Island, Greece, September 20-23, 2022
Available from: 2022-12-19 Created: 2022-12-19 Last updated: 2024-01-03Bibliographically approved
Winkler, N. P., Kotlyar, O., Schaffernicht, E., Fan, H., Matsukura, H., Ishida, H., . . . Lilienthal, A. (2022). Learning From the Past: Sequential Deep Learning for Gas Distribution Mapping. In: Danilo Tardioli; Vicente Matellán; Guillermo Heredia; Manuel F. Silva; Lino Marques (Ed.), ROBOT2022: Fifth Iberian Robotics Conference: Advances in Robotics, Volume 2. Paper presented at ROBOT2022: Fifth Iberian Robotics Conference, Zaragoza, Spain, November 23-25, 2022 (pp. 178-188). Springer, 590
Open this publication in new window or tab >>Learning From the Past: Sequential Deep Learning for Gas Distribution Mapping
Show others...
2022 (English)In: ROBOT2022: Fifth Iberian Robotics Conference: Advances in Robotics, Volume 2 / [ed] Danilo Tardioli; Vicente Matellán; Guillermo Heredia; Manuel F. Silva; Lino Marques, Springer, 2022, Vol. 590, p. 178-188Conference paper, Published paper (Refereed)
Abstract [en]

To better understand the dynamics in hazardous environments, gas distribution mapping aims to map the gas concentration levels of a specified area precisely. Sampling is typically carried out in a spatially sparse manner, either with a mobile robot or a sensor network and concentration values between known data points have to be interpolated. In this paper, we investigate sequential deep learning models that are able to map the gas distribution based on a multiple time step input from a sensor network. We propose a novel hybrid convolutional LSTM - transpose convolutional structure that we train with synthetic gas distribution data. Our results show that learning the spatial and temporal correlation of gas plume patterns outperforms a non-sequential neural network model.

Place, publisher, year, edition, pages
Springer, 2022
Series
Lecture Notes in Networks and Systems, ISSN 2367-3370, E-ISSN 2367-3389 ; 590
Keywords
Convolutional LSTM, Gas Distribution Mapping, Sequential Learning, Spatial Interpolation
National Category
Computer Sciences
Identifiers
urn:nbn:se:oru:diva-102769 (URN)10.1007/978-3-031-21062-4_15 (DOI)000906176800015 ()2-s2.0-85145267880 (Scopus ID)9783031210617 (ISBN)9783031210624 (ISBN)
Conference
ROBOT2022: Fifth Iberian Robotics Conference, Zaragoza, Spain, November 23-25, 2022
Note

Funding agency:

Ministry of Education, Culture, Sports, Science and Technology, Japan (MEXT)

Japan Society for the Promotion of Science 22H04952

Available from: 2022-12-19 Created: 2022-12-19 Last updated: 2024-01-03Bibliographically approved
Winkler, N. P., Matsukura, H., Neumann, P. P., Schaffernicht, E., Ishida, H. & Lilienthal, A. J. (2022). Super-Resolution for Gas Distribution Mapping: Convolutional Encoder-Decoder Network. In: 2022 IEEE International Symposium on Olfaction and Electronic Nose (ISOEN): . Paper presented at IEEE International Symposium on Olfaction and Electronic Nose (ISOEN 2022), Aveiro, Portugal, May 29 - June 1, 2022. IEEE
Open this publication in new window or tab >>Super-Resolution for Gas Distribution Mapping: Convolutional Encoder-Decoder Network
Show others...
2022 (English)In: 2022 IEEE International Symposium on Olfaction and Electronic Nose (ISOEN), IEEE , 2022Conference paper, Published paper (Refereed)
Abstract [en]

Gas distribution mapping is important to have an accurate understanding of gas concentration levels in hazardous environments. A major problem is that in-situ gas sensors are only able to measure concentrations at their specific location. The gas distribution in-between the sampling locations must therefore be modeled. In this research, we interpret the task of spatial interpolation between sparsely distributed sensors as a task of enhancing an image's resolution, namely super-resolution. Because autoencoders are proven to perform well for this super-resolution task, we trained a convolutional encoder-decoder neural network to map the gas distribution over a spatially sparse sensor network. Due to the difficulty to collect real-world gas distribution data and missing ground truth, we used synthetic data generated with a gas distribution simulator for training and evaluation of the model. Our results show that the neural network was able to learn the behavior of gas plumes and outperforms simpler interpolation techniques.

Place, publisher, year, edition, pages
IEEE, 2022
Keywords
gas distribution mapping, spatial interpolation, deep learning, super-resolution, sensor network
National Category
Computer Sciences
Identifiers
urn:nbn:se:oru:diva-101455 (URN)10.1109/ISOEN54820.2022.9789555 (DOI)000852626300003 ()2-s2.0-85133213201 (Scopus ID)9781665458603 (ISBN)9781665458610 (ISBN)
Conference
IEEE International Symposium on Olfaction and Electronic Nose (ISOEN 2022), Aveiro, Portugal, May 29 - June 1, 2022
Note

Funding agency:

Ministry of Education, Culture, Sports, Science and Technology, Japan (MEXT) Japan Society for the Promotion of Science 19H02103

Available from: 2022-09-26 Created: 2022-09-26 Last updated: 2024-01-03Bibliographically approved
Winkler, N. P., Neumann, P. P., Kohlhoff, H., Erdmann, J., Schaffernicht, E. & Lilienthal, A. (2021). Development of a Low-Cost Sensing Node with Active Ventilation Fan for Air Pollution Monitoring. In: SMSI 2021 Proceedings: . Paper presented at SMSI 2021 Conference: Sensor and Measurement Science International, (Digital conference), May 3-6, 2021 (pp. 260-261).
Open this publication in new window or tab >>Development of a Low-Cost Sensing Node with Active Ventilation Fan for Air Pollution Monitoring
Show others...
2021 (English)In: SMSI 2021 Proceedings, 2021, p. 260-261Conference paper, Published paper (Refereed)
Abstract [en]

A fully designed low-cost sensing node for air pollution monitoring and calibration results for several low-cost gas sensors are presented. As the state of the art is lacking information on the importance of an active ventilation system, the effect of an active fan is compared to the passive ventilation of a lamellar structured casing. Measurements obtained in an urban outdoor environment show that readings of the low-cost dust sensor (Sharp GP2Y1010AU0F) are distorted by the active ventilation system. While this behavior requires further research, a correlation with temperature and humidity inside the node shown.

Keywords
wireless sensing node, environmental monitoring, air pollution, sensor network
National Category
Other Electrical Engineering, Electronic Engineering, Information Engineering
Identifiers
urn:nbn:se:oru:diva-96666 (URN)10.5162/SMSI2021/D3.5 (DOI)
Conference
SMSI 2021 Conference: Sensor and Measurement Science International, (Digital conference), May 3-6, 2021
Available from: 2022-01-25 Created: 2022-01-25 Last updated: 2024-01-03Bibliographically approved
Winkler, N. P., Neumann, P. P., Schaffernicht, E. & Lilienthal, A. (2021). Using Redundancy in a Sensor Network to Compensate Sensor Failures. In: 2021 IEEE SENSORS: . Paper presented at IEEE SENSORS 2021, (Virtual conference), October 31 - November 4, 2021.
Open this publication in new window or tab >>Using Redundancy in a Sensor Network to Compensate Sensor Failures
2021 (English)In: 2021 IEEE SENSORS, 2021Conference paper, Published paper (Refereed)
Abstract [en]

Wireless sensor networks provide occupational health experts with valuable information about the distribution of air pollutants in an environment. However, especially low-cost sensors may produce faulty measurements or fail completely. Consequently, not only spatial coverage but also redundancy should be a design criterion for the deployment of a sensor network. For a sensor network deployed in a steel factory, we analyze the correlations between sensors and build machine learning forecasting models, to investigate how well the sensor network can compensate for the outage of sensors. While our results show promising prediction quality of the models, they also indicate the presence of spatially very limited events. We, therefore, conclude that initial measurements with, e.g., mobile units, could help to identify important locations to design redundant sensor networks.

Keywords
Environmental monitoring, wireless sensor network, sensor placement, machine learning
National Category
Other Electrical Engineering, Electronic Engineering, Information Engineering
Identifiers
urn:nbn:se:oru:diva-96667 (URN)
Conference
IEEE SENSORS 2021, (Virtual conference), October 31 - November 4, 2021
Available from: 2022-01-25 Created: 2022-01-25 Last updated: 2024-01-03Bibliographically approved
Winkler, N. P., Neumann, P. P., Schaffernicht, E. & Lilienthal, A. J. (2021). Using Redundancy in a Sensor Network to Compensate Sensor Failures. In: 2021 IEEE SENSORS: . Paper presented at 20th IEEE Sensors Conference, (Virtual conference), October 31 - November 4, 2021. IEEE
Open this publication in new window or tab >>Using Redundancy in a Sensor Network to Compensate Sensor Failures
2021 (English)In: 2021 IEEE SENSORS, IEEE , 2021Conference paper, Published paper (Refereed)
Abstract [en]

Wireless sensor networks provide occupational health experts with valuable information about the distribution of air pollutants in an environment. However, especially low-cost sensors may produce faulty measurements or fail completely. Consequently, not only spatial coverage but also redundancy should be a design criterion for the deployment of a sensor network. For a sensor network deployed in a steel factory, we analyze the correlations between sensors and build machine learning forecasting models, to investigate how well the sensor network can compensate for the outage of sensors. While our results show promising prediction quality of the models, they also indicate the presence of spatially very limited events. We, therefore, conclude that initial measurements with, e.g., mobile units, could help to identify important locations to design redundant sensor networks.

Place, publisher, year, edition, pages
IEEE, 2021
Series
Proceedings of IEEE Sensors, ISSN 1930-0395, E-ISSN 2168-9229
Keywords
environmental monitoring, wireless sensor network, sensor placement, machine learning
National Category
Computer Sciences
Identifiers
urn:nbn:se:oru:diva-97668 (URN)10.1109/SENSORS47087.2021.9639479 (DOI)000755468300033 ()2-s2.0-85123610376 (Scopus ID)9781728195018 (ISBN)9781728195025 (ISBN)
Conference
20th IEEE Sensors Conference, (Virtual conference), October 31 - November 4, 2021
Note

Funding agency:

SAFeRA

Available from: 2022-02-25 Created: 2022-02-25 Last updated: 2024-01-03Bibliographically approved
Winkler, N. P., Neumann, P. P., Säämänen, A., Schaffernicht, E. & Lilienthal, A. J. (2020). High-quality meets low-cost: Approaches for hybrid-mobility sensor networks. In: MATERIALS TODAY-PROCEEDINGS: . Paper presented at 36th Danubia Adria Symposium on Advances in Experimental Mechanics, SEP 24-27, 2019, Pilsen, CZECH REPUBLIC (pp. 250-253). Elsevier, 32
Open this publication in new window or tab >>High-quality meets low-cost: Approaches for hybrid-mobility sensor networks
Show others...
2020 (English)In: MATERIALS TODAY-PROCEEDINGS, Elsevier, 2020, Vol. 32, p. 250-253Conference paper, Published paper (Refereed)
Abstract [en]

Air pollution within industrial scenarios is a major risk for workers, which is why detailed knowledge about the dispersion of dusts and gases is necessary. This paper introduces a system combining stationary low-cost and high-quality sensors, carried by ground robots and unmanned aerial vehicles. Based on these dense sampling capabilities, detailed distribution maps of dusts and gases will be created. This system enables various research opportunities, especially on the fields of distribution mapping and sensor planning. Standard approaches for distribution mapping can be enhanced with knowledge about the environment's characteristics, while the effectiveness of new approaches, utilizing neural networks, can be further investigated. The influence of different sensor network setups on the predictive quality of distribution algorithms will be researched and metrics for the quantification of a sensor network's quality will be investigated.

Place, publisher, year, edition, pages
Elsevier, 2020
Series
Materials Today: Proceedings, E-ISSN 2214-7853
Keywords
Mobile robot olfaction, Air quality monitoring, Wireless sensor network, Gas distribution mapping, Occupational health
National Category
Computer Sciences
Identifiers
urn:nbn:se:oru:diva-87780 (URN)10.1016/j.matpr.2020.05.799 (DOI)000587965400034 ()
Conference
36th Danubia Adria Symposium on Advances in Experimental Mechanics, SEP 24-27, 2019, Pilsen, CZECH REPUBLIC
Note

Funding Agency:

SAF(sic)RA

Available from: 2020-12-03 Created: 2020-12-03 Last updated: 2024-01-03Bibliographically approved
Organisations

Search in DiVA

Show all publications