Implementing Control-oriented Meta-learning on Hardware
Unpredictable weather conditions pose a daunting challenge for the robust control of unmanned aerial vehicles, also known as drones. The control-oriented meta-learning algorithm aims to solve this problem by learning a controller that can adapt to dynamic environments. This algorithm has already bee...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Published: |
Massachusetts Institute of Technology
2024
|
| Online Access: | https://hdl.handle.net/1721.1/156775 |
| _version_ | 1826204344390778880 |
|---|---|
| author | Sohn, Joshua C. |
| author2 | Azizan, Navid |
| author_facet | Azizan, Navid Sohn, Joshua C. |
| author_sort | Sohn, Joshua C. |
| collection | MIT |
| description | Unpredictable weather conditions pose a daunting challenge for the robust control of unmanned aerial vehicles, also known as drones. The control-oriented meta-learning algorithm aims to solve this problem by learning a controller that can adapt to dynamic environments. This algorithm has already been derived and simulated for a two-dimensional model. This project explores the implementation of the control-oriented meta-learning algorithm on a hardware platform. After extending the algorithm to a three-dimensional model, it was tested in a physics-based simulator and deployed on a hexarotor in the real world. Both in simulation and in real life, the learned controller outperformed a traditional controller in the presence of wind. |
| first_indexed | 2024-09-23T12:52:47Z |
| format | Thesis |
| id | mit-1721.1/156775 |
| institution | Massachusetts Institute of Technology |
| last_indexed | 2024-09-23T12:52:47Z |
| publishDate | 2024 |
| publisher | Massachusetts Institute of Technology |
| record_format | dspace |
| spelling | mit-1721.1/1567752024-09-17T03:33:02Z Implementing Control-oriented Meta-learning on Hardware Sohn, Joshua C. Azizan, Navid Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Unpredictable weather conditions pose a daunting challenge for the robust control of unmanned aerial vehicles, also known as drones. The control-oriented meta-learning algorithm aims to solve this problem by learning a controller that can adapt to dynamic environments. This algorithm has already been derived and simulated for a two-dimensional model. This project explores the implementation of the control-oriented meta-learning algorithm on a hardware platform. After extending the algorithm to a three-dimensional model, it was tested in a physics-based simulator and deployed on a hexarotor in the real world. Both in simulation and in real life, the learned controller outperformed a traditional controller in the presence of wind. M.Eng. 2024-09-16T13:48:21Z 2024-09-16T13:48:21Z 2024-05 2024-07-11T14:37:17.645Z Thesis https://hdl.handle.net/1721.1/156775 Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) Copyright retained by author(s) https://creativecommons.org/licenses/by-nc-nd/4.0/ application/pdf Massachusetts Institute of Technology |
| spellingShingle | Sohn, Joshua C. Implementing Control-oriented Meta-learning on Hardware |
| title | Implementing Control-oriented Meta-learning on Hardware |
| title_full | Implementing Control-oriented Meta-learning on Hardware |
| title_fullStr | Implementing Control-oriented Meta-learning on Hardware |
| title_full_unstemmed | Implementing Control-oriented Meta-learning on Hardware |
| title_short | Implementing Control-oriented Meta-learning on Hardware |
| title_sort | implementing control oriented meta learning on hardware |
| url | https://hdl.handle.net/1721.1/156775 |
| work_keys_str_mv | AT sohnjoshuac implementingcontrolorientedmetalearningonhardware |