Sparse Gaussian Process SLAM, Storage and Filtering for AUV Multibeam Bathymetry
| Autor: | Nils Bore, John Folkesson, Ignacio Torroba |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2018 |
| Předmět: |
Vehicle Engineering
Computer science Digital storage Autonomous vehicles Gaussian distribution Simultaneous localization and mapping Farkostteknik Sparse Gaussian process symbols.namesake Multi-beam sensors Position (vector) Datorseende och robotik (autonoma system) Dead reckoning Computer vision Bathymetry Autonomous underwater vehicles Representation (mathematics) Alignment optimization Gaussian process Computer Vision and Robotics (Autonomous Systems) business.industry Real-world datasets Velocity sensor Navigation Position estimates symbols Artificial intelligence Multibeam bathymetry business Focus (optics) Gaussian noise (electronic) Data compression |
| Popis: | With dead-reckoning from velocity sensors, AUVs may construct short-term, local bathymetry maps of the sea floor using multibeam sensors. However, the position estimate from dead-reckoning will include some drift that grows with time. In this work, we focus on long-term onboard storage of these local bathymetry maps, and the alignment of maps with respect to each other. We propose using Sparse Gaussian Processes for this purpose, and show that the representation has several advantages, including an intuitive alignment optimization, data compression, and sensor noise filtering. We demonstrate these three key capabilities on two real-world datasets. QC 20191024 |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|