Six degree-of-freedom body-fixed hovering over unmapped asteroids via LIDAR altimetry and reinforcement meta-learning
© 2020 IAA We optimize a six degrees of freedom hovering policy using reinforcement meta-learning. The policy maps flash LIDAR measurements directly to on/off spacecraft body-frame thrust commands, allowing hovering at a fixed position and attitude in the asteroid body-fixed reference frame. Importa...
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
Elsevier BV
2022
|
| Online Access: | https://hdl.handle.net/1721.1/135438.2 |
| _version_ | 1811078321929191424 |
|---|---|
| author | Gaudet, Brian Linares, Richard Furfaro, Roberto |
| author2 | Massachusetts Institute of Technology. Department of Aeronautics and Astronautics |
| author_facet | Massachusetts Institute of Technology. Department of Aeronautics and Astronautics Gaudet, Brian Linares, Richard Furfaro, Roberto |
| author_sort | Gaudet, Brian |
| collection | MIT |
| description | © 2020 IAA We optimize a six degrees of freedom hovering policy using reinforcement meta-learning. The policy maps flash LIDAR measurements directly to on/off spacecraft body-frame thrust commands, allowing hovering at a fixed position and attitude in the asteroid body-fixed reference frame. Importantly, the policy does not require position and velocity estimates, and can operate in environments with unknown dynamics, and without an asteroid shape model or navigation aids. Indeed, during optimization the agent is confronted with a new randomly generated asteroid for each episode, insuring that it does not learn an asteroid's shape, texture, or environmental dynamics. This allows the deployed policy to generalize well to novel asteroid characteristics, which we demonstrate in our experiments. Moreover, our experiments show that the optimized policy adapts to actuator failure and sensor noise. Although the policy is optimized using randomly generated synthetic asteroids, it is tested on two shape models from actual asteroids: Bennu and Itokawa. We find that the policy generalizes well to these shape models. The hovering controller has the potential to simplify mission planning by allowing asteroid body-fixed hovering immediately upon the spacecraft's arrival to an asteroid. This in turn simplifies shape model generation and allows resource mapping via remote sensing immediately upon arrival at the target asteroid. |
| first_indexed | 2024-09-23T10:57:45Z |
| format | Article |
| id | mit-1721.1/135438.2 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T10:57:45Z |
| publishDate | 2022 |
| publisher | Elsevier BV |
| record_format | dspace |
| spelling | mit-1721.1/135438.22024-06-14T16:24:48Z Six degree-of-freedom body-fixed hovering over unmapped asteroids via LIDAR altimetry and reinforcement meta-learning Gaudet, Brian Linares, Richard Furfaro, Roberto Massachusetts Institute of Technology. Department of Aeronautics and Astronautics © 2020 IAA We optimize a six degrees of freedom hovering policy using reinforcement meta-learning. The policy maps flash LIDAR measurements directly to on/off spacecraft body-frame thrust commands, allowing hovering at a fixed position and attitude in the asteroid body-fixed reference frame. Importantly, the policy does not require position and velocity estimates, and can operate in environments with unknown dynamics, and without an asteroid shape model or navigation aids. Indeed, during optimization the agent is confronted with a new randomly generated asteroid for each episode, insuring that it does not learn an asteroid's shape, texture, or environmental dynamics. This allows the deployed policy to generalize well to novel asteroid characteristics, which we demonstrate in our experiments. Moreover, our experiments show that the optimized policy adapts to actuator failure and sensor noise. Although the policy is optimized using randomly generated synthetic asteroids, it is tested on two shape models from actual asteroids: Bennu and Itokawa. We find that the policy generalizes well to these shape models. The hovering controller has the potential to simplify mission planning by allowing asteroid body-fixed hovering immediately upon the spacecraft's arrival to an asteroid. This in turn simplifies shape model generation and allows resource mapping via remote sensing immediately upon arrival at the target asteroid. 2022-03-21T14:51:32Z 2021-10-27T20:23:28Z 2022-03-21T14:51:32Z 2020-07 2020-02 2021-05-06T12:51:39Z Article http://purl.org/eprint/type/JournalArticle 0094-5765 https://hdl.handle.net/1721.1/135438.2 en http://dx.doi.org/10.1016/j.actaastro.2020.03.026 Acta Astronautica Creative Commons Attribution-NonCommercial-NoDerivs License http://creativecommons.org/licenses/by-nc-nd/4.0/ application/octet-stream Elsevier BV MIT web domain |
| spellingShingle | Gaudet, Brian Linares, Richard Furfaro, Roberto Six degree-of-freedom body-fixed hovering over unmapped asteroids via LIDAR altimetry and reinforcement meta-learning |
| title | Six degree-of-freedom body-fixed hovering over unmapped asteroids via LIDAR altimetry and reinforcement meta-learning |
| title_full | Six degree-of-freedom body-fixed hovering over unmapped asteroids via LIDAR altimetry and reinforcement meta-learning |
| title_fullStr | Six degree-of-freedom body-fixed hovering over unmapped asteroids via LIDAR altimetry and reinforcement meta-learning |
| title_full_unstemmed | Six degree-of-freedom body-fixed hovering over unmapped asteroids via LIDAR altimetry and reinforcement meta-learning |
| title_short | Six degree-of-freedom body-fixed hovering over unmapped asteroids via LIDAR altimetry and reinforcement meta-learning |
| title_sort | six degree of freedom body fixed hovering over unmapped asteroids via lidar altimetry and reinforcement meta learning |
| url | https://hdl.handle.net/1721.1/135438.2 |
| work_keys_str_mv | AT gaudetbrian sixdegreeoffreedombodyfixedhoveringoverunmappedasteroidsvialidaraltimetryandreinforcementmetalearning AT linaresrichard sixdegreeoffreedombodyfixedhoveringoverunmappedasteroidsvialidaraltimetryandreinforcementmetalearning AT furfaroroberto sixdegreeoffreedombodyfixedhoveringoverunmappedasteroidsvialidaraltimetryandreinforcementmetalearning |