Adaptive guidance and integrated navigation with reinforcement meta-learning
© 2020 IAA This paper proposes a novel adaptive guidance system developed using reinforcement meta-learning with a recurrent policy and value function approximator. The use of recurrent network layers allows the deployed policy to adapt in real time to environmental forces acting on the agent. We co...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier BV
2021
|
| Online Access: | https://hdl.handle.net/1721.1/135440 |
| _version_ | 1811096300314165248 |
|---|---|
| author | Gaudet, B Linares, R Furfaro, R |
| author_facet | Gaudet, B Linares, R Furfaro, R |
| author_sort | Gaudet, B |
| collection | MIT |
| description | © 2020 IAA This paper proposes a novel adaptive guidance system developed using reinforcement meta-learning with a recurrent policy and value function approximator. The use of recurrent network layers allows the deployed policy to adapt in real time to environmental forces acting on the agent. We compare the performance of the DR/DV guidance law, an RL agent with a non-recurrent policy, and an RL agent with a recurrent policy in four challenging environments with unknown but highly variable dynamics. These tasks include a safe Mars landing with random engine failure and a landing on an asteroid with unknown environmental dynamics. We also demonstrate the ability of a RL meta-learning optimized policy to implement a guidance law using observations consisting of only Doppler radar altimeter readings in a Mars landing environment, and LIDAR altimeter readings in an asteroid landing environment thus integrating guidance and navigation. |
| first_indexed | 2024-09-23T16:41:42Z |
| format | Article |
| id | mit-1721.1/135440 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T16:41:42Z |
| publishDate | 2021 |
| publisher | Elsevier BV |
| record_format | dspace |
| spelling | mit-1721.1/1354402021-10-28T03:07:47Z Adaptive guidance and integrated navigation with reinforcement meta-learning Gaudet, B Linares, R Furfaro, R © 2020 IAA This paper proposes a novel adaptive guidance system developed using reinforcement meta-learning with a recurrent policy and value function approximator. The use of recurrent network layers allows the deployed policy to adapt in real time to environmental forces acting on the agent. We compare the performance of the DR/DV guidance law, an RL agent with a non-recurrent policy, and an RL agent with a recurrent policy in four challenging environments with unknown but highly variable dynamics. These tasks include a safe Mars landing with random engine failure and a landing on an asteroid with unknown environmental dynamics. We also demonstrate the ability of a RL meta-learning optimized policy to implement a guidance law using observations consisting of only Doppler radar altimeter readings in a Mars landing environment, and LIDAR altimeter readings in an asteroid landing environment thus integrating guidance and navigation. 2021-10-27T20:23:28Z 2021-10-27T20:23:28Z 2020-04-01 2021-05-06T13:40:28Z Article http://purl.org/eprint/type/JournalArticle https://hdl.handle.net/1721.1/135440 en 10.1016/j.actaastro.2020.01.007 Acta Astronautica Creative Commons Attribution-NonCommercial-NoDerivs License http://creativecommons.org/licenses/by-nc-nd/4.0/ application/pdf Elsevier BV arXiv |
| spellingShingle | Gaudet, B Linares, R Furfaro, R Adaptive guidance and integrated navigation with reinforcement meta-learning |
| title | Adaptive guidance and integrated navigation with reinforcement meta-learning |
| title_full | Adaptive guidance and integrated navigation with reinforcement meta-learning |
| title_fullStr | Adaptive guidance and integrated navigation with reinforcement meta-learning |
| title_full_unstemmed | Adaptive guidance and integrated navigation with reinforcement meta-learning |
| title_short | Adaptive guidance and integrated navigation with reinforcement meta-learning |
| title_sort | adaptive guidance and integrated navigation with reinforcement meta learning |
| url | https://hdl.handle.net/1721.1/135440 |
| work_keys_str_mv | AT gaudetb adaptiveguidanceandintegratednavigationwithreinforcementmetalearning AT linaresr adaptiveguidanceandintegratednavigationwithreinforcementmetalearning AT furfaror adaptiveguidanceandintegratednavigationwithreinforcementmetalearning |