Mini Cheetah Sensor Suite for Visual Perception
The mini cheetah robot is a small lightweight quadruped capable of dynamic movements. Currently, the mini cheetah does not have a method of collecting stabilized perception data. Research has been conducted at the University of Michigan, where a sensor suite for collecting perception data has been c...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Published: |
Massachusetts Institute of Technology
2022
|
| Online Access: | https://hdl.handle.net/1721.1/143286 |
| _version_ | 1811087502281277440 |
|---|---|
| author | Papalia, Lillian C. |
| author2 | Leonard, John J. |
| author_facet | Leonard, John J. Papalia, Lillian C. |
| author_sort | Papalia, Lillian C. |
| collection | MIT |
| description | The mini cheetah robot is a small lightweight quadruped capable of dynamic movements. Currently, the mini cheetah does not have a method of collecting stabilized perception data. Research has been conducted at the University of Michigan, where a sensor suite for collecting perception data has been created. While the sensor suite is able to collect perception data, it is heavy and affects the agility of the mini cheetah robot and does not incorporate any active stabilization into the design. A lightweight sensor suite was designed and manufactured incorporating a commercial gimbal. Three tests were conducted on a rigid mounted camera and a gimbal stabilized camera. AprilTag detection as well as visual qualitative analysis determined the rigid mounted camera performed better than the current gimbal configuration. Further work involving active stabilization will be needed to determine a satisfactory solution, particularly when the mini cheetah is pronking. |
| first_indexed | 2024-09-23T13:47:07Z |
| format | Thesis |
| id | mit-1721.1/143286 |
| institution | Massachusetts Institute of Technology |
| last_indexed | 2024-09-23T13:47:07Z |
| publishDate | 2022 |
| publisher | Massachusetts Institute of Technology |
| record_format | dspace |
| spelling | mit-1721.1/1432862022-06-16T03:23:52Z Mini Cheetah Sensor Suite for Visual Perception Papalia, Lillian C. Leonard, John J. Massachusetts Institute of Technology. Department of Mechanical Engineering The mini cheetah robot is a small lightweight quadruped capable of dynamic movements. Currently, the mini cheetah does not have a method of collecting stabilized perception data. Research has been conducted at the University of Michigan, where a sensor suite for collecting perception data has been created. While the sensor suite is able to collect perception data, it is heavy and affects the agility of the mini cheetah robot and does not incorporate any active stabilization into the design. A lightweight sensor suite was designed and manufactured incorporating a commercial gimbal. Three tests were conducted on a rigid mounted camera and a gimbal stabilized camera. AprilTag detection as well as visual qualitative analysis determined the rigid mounted camera performed better than the current gimbal configuration. Further work involving active stabilization will be needed to determine a satisfactory solution, particularly when the mini cheetah is pronking. S.B. 2022-06-15T13:09:46Z 2022-06-15T13:09:46Z 2022-02 2022-03-08T19:48:49.599Z Thesis https://hdl.handle.net/1721.1/143286 In Copyright - Educational Use Permitted Copyright MIT http://rightsstatements.org/page/InC-EDU/1.0/ application/pdf Massachusetts Institute of Technology |
| spellingShingle | Papalia, Lillian C. Mini Cheetah Sensor Suite for Visual Perception |
| title | Mini Cheetah Sensor Suite for Visual Perception |
| title_full | Mini Cheetah Sensor Suite for Visual Perception |
| title_fullStr | Mini Cheetah Sensor Suite for Visual Perception |
| title_full_unstemmed | Mini Cheetah Sensor Suite for Visual Perception |
| title_short | Mini Cheetah Sensor Suite for Visual Perception |
| title_sort | mini cheetah sensor suite for visual perception |
| url | https://hdl.handle.net/1721.1/143286 |
| work_keys_str_mv | AT papalialillianc minicheetahsensorsuiteforvisualperception |