TY - GEN
T1 - Gas Distribution Mapping With Radius-Based, Bi-directional Graph Neural Networks (RABI-GNN)
AU - Winkler, Nicolas P.
AU - Neumann, Patrick P.
AU - Schaffernicht, Erik
AU - Lilienthal, Achim J.
N1 - Publisher Copyright:
©2024 IEEE.
PY - 2024
Y1 - 2024
N2 - 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 learning-based 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.
AB - 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 learning-based 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.
KW - gas distribution mapping
KW - graph neural networks
KW - mobile robot olfaction
KW - spatial interpolation
UR - http://www.scopus.com/inward/record.url?scp=85197434833&partnerID=8YFLogxK
U2 - 10.1109/ISOEN61239.2024.10556309
DO - 10.1109/ISOEN61239.2024.10556309
M3 - Conference contribution
AN - SCOPUS:85197434833
T3 - ISOEN 2024 - International Symposium on Olfaction and Electronic Nose, Proceedings
BT - ISOEN 2024 - International Symposium on Olfaction and Electronic Nose, Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2024 IEEE International Symposium on Olfaction and Electronic Nose, ISOEN 2024
Y2 - 12 May 2024 through 15 May 2024
ER -