Filtern
Dokumenttyp
- Vortrag (3) (entfernen)
Sprache
- Englisch (3)
Referierte Publikation
- nein (3) (entfernen)
Schlagworte
- Gas distribution mapping (3) (entfernen)
Organisationseinheit der BAM
Eingeladener Vortrag
- nein (3)
This seminar series comprises three lectures on drones and ground robots capable of sensing chemical substances and two accompanying hands-on workshops on nano drones equipped with chemical sensors. In the first lecture, Dr. Neumann will talk on a summary of aerial robot olfaction and his research. In the second and third lectures, more details of the nano drones equipped with chemical sensors will be provided, and the participants can get hands on experience of the nano drones in the accompanying workshops. The lectures will be given in English. This seminar series is supported by FY2023 Researcher Exchanges Program in UEC.
This seminar series comprises three lectures on drones and ground robots capable of sensing chemical substances and two accompanying hands-on workshops on nano drones equipped with chemical sensors. In the first lecture, Dr. Neumann will talk on a summary of aerial robot olfaction and his research. In the second and third lectures, more details of the nano drones equipped with chemical sensors will be provided, and the participants can get hands on experience of the nano
drones in the accompanying workshops. The lectures will be given in English. This seminar series is supported by FY2023 Researcher Exchanges Program in UEC.
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.