What is it about?
As humanity branches out into space in the coming decades, such as with the Artemis Gateway space station, robots will need to monitor and maintain these space habitats when humans are not present. To do this, they need to be able to detect changes in their environment. This work demonstrates a way for robots to do this autonomously using 3D point clouds.
Featured Image
Photo by NASA on Unsplash
Why is it important?
This work presents an autonomous method for robots to detect changes in their habitat with results demonstrated in a simulated microgravity environment, NASA Ames' Granite Lab. This work further demonstrates the possibility of detecting multiple changes in the environment at once, as well as objects removed from the environment (rather than just added).
Perspectives
This paper was an extension of my master's thesis at Chalmers University of Technology for my MSc Complex Adaptive Systems degree. It was a great joy to develop and write with all of the other coauthors, and I hope that this paper will inspire others to contribute to open source software to further humanity's reach into the stars.
Jamie Santos
Chalmers tekniska hogskola
Read the Original
This page is a summary of: Unsupervised Change Detection for Space Habitats Using 3D Point Clouds, January 2024, American Institute of Aeronautics and Astronautics (AIAA),
DOI: 10.2514/6.2024-1960.
You can read the full text:
Resources
Unsupervised Change Detection for Space Habitats Using 3D Point Clouds
This presentation was delivered at 2023 AIAA SciTech Conference in Orlando, Florida and supplements our paper, "Unsupervised Change Detection for Space Habitats Using 3D Point Clouds." Abstract: This work presents an algorithm for scene change detection from point clouds to enable autonomous robotic caretaking in future space habitats. Autonomous robotic systems will help maintain future deep-space habitats, such as the Gateway space station, which will be uncrewed for extended periods. Existing scene analysis software used on the International Space Station (ISS) relies on manually-labeled images for detecting changes. In contrast, the algorithm presented in this work uses raw, unlabeled point clouds as inputs. The algorithm first applies modified Expectation-Maximization Gaussian Mixture Model (GMM) clustering to two input point clouds. It then performs change detection by comparing the GMMs using the Earth Mover's Distance. The algorithm is validated quantitatively and qualitatively using a test dataset collected by an Astrobee robot in the NASA Ames Granite Lab comprising single frame depth images taken directly by Astrobee and full-scene reconstructed maps built with RGB-D and pose data from Astrobee. The runtimes of the approach are also analyzed in depth. The source code is publicly released to promote further development. Authors: Jamie Santos, Holly Dinkel, Julia Di, Paulo V.K. Borges, Marina Moreira, Oleg Alexandrov, Brian Coltin, Trey Smith Venue: https://www.aiaa.org/SciTech Paper: https://arxiv.org/abs/2312.02396
Unsupervised Change Detection for Space Habitats Using 3D Point Clouds
This work presents an algorithm for scene change detection from point clouds to enable autonomous robotic caretaking in future space habitats. Autonomous robotic systems will help maintain future deep-space habitats, such as the Gateway space station, which will be uncrewed for extended periods. Existing scene analysis software used on the International Space Station (ISS) relies on manually-labeled images for detecting changes. In contrast, the algorithm presented in this work uses raw, unlabeled point clouds as inputs. The algorithm first applies modified Expectation-Maximization Gaussian Mixture Model (GMM) clustering to two input point clouds. It then performs change detection by comparing the GMMs using the Earth Mover's Distance. The algorithm is validated quantitatively and qualitatively using a test dataset collected by an Astrobee robot in the NASA Ames Granite Lab comprising single frame depth images taken directly by Astrobee and full-scene reconstructed maps built with RGB-D and pose data from Astrobee. The runtimes of the approach are also analyzed in depth. The source code is publicly released to promote further development.
Contributors
The following have contributed to this page
