Robot motion planning with contact from global pseudo-inverse map
Thesis: Ph. D., Massachusetts Institute of Technology, Department of Mechanical Engineering, 2018.
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | eng |
| Published: |
Massachusetts Institute of Technology
2019
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/1721.1/120251 |
| _version_ | 1811084378333249536 |
|---|---|
| author | Choi, Changrak |
| author2 | Emilio Frazzoli. |
| author_facet | Emilio Frazzoli. Choi, Changrak |
| author_sort | Choi, Changrak |
| collection | MIT |
| description | Thesis: Ph. D., Massachusetts Institute of Technology, Department of Mechanical Engineering, 2018. |
| first_indexed | 2024-09-23T12:49:46Z |
| format | Thesis |
| id | mit-1721.1/120251 |
| institution | Massachusetts Institute of Technology |
| language | eng |
| last_indexed | 2024-09-23T12:49:46Z |
| publishDate | 2019 |
| publisher | Massachusetts Institute of Technology |
| record_format | dspace |
| spelling | mit-1721.1/1202512019-04-12T22:37:59Z Robot motion planning with contact from global pseudo-inverse map Choi, Changrak Emilio Frazzoli. Massachusetts Institute of Technology. Department of Mechanical Engineering. Massachusetts Institute of Technology. Department of Mechanical Engineering. Mechanical Engineering. Thesis: Ph. D., Massachusetts Institute of Technology, Department of Mechanical Engineering, 2018. Cataloged from PDF version of thesis. Includes bibliographical references (pages 101-108). In the robot motion planning problems, environment and its objects are often treated as obstacles to be avoided. However, there are situations where contacting with the environment is not costly. Moreover, in many cases, making contact can actually help a robot to maneuver around to reach a goal state which would not have been possible otherwise. This thesis presents a framework for motion planner that utilizes multiple contacts with the environment and its objects. The planner is targeted to autonomously generate motion, where robot has to make multiple contact with different part of its body in order to achieve a task objective. It is motivated by and has significance in developing a robust humanoid planner that is capable of recovering from a fall down. The recent DRC has been marked with compilation of humanoid robots falling down, but only one robot managed to recover to a standing up position. In a real disaster scenario, the inability to stand up would mean end of the rescue mission for what is extremely expensive machinery. A robust planner capable of recovery is must and this work contributes towards it. The developed planner autonomously generates standing up motion from fall down in the presence of torque limits. The proposed multi-contact motion planner leverages upon following two key components. Existing multi-contact planners require good initial seeds to successfully generate a motion. These are hard to find and often manually encoded. Here, we utilize pre-computed global pseudo-inverse map (inverse kinematic map for each contact-state that has property of global resolution, connected by connectivity functions) to generate multi-contact motion from current configuration to the goal without need for an initial seed. Nevertheless, constructing the global pseudo-inverse map is computationally expensive. In an effort to facilitate the construction, we utilize singular configurations as a heuristic to reduce the search space and justify its use based on the physical analysis. Although computationally expensive, once pre-computed, the global map can be used to generate plans fast online in a multi-query manner. by Changrak Choi. Ph. D. 2019-02-05T16:01:12Z 2019-02-05T16:01:12Z 2018 2018 Thesis http://hdl.handle.net/1721.1/120251 1083141520 eng MIT theses are protected by copyright. They may be viewed, downloaded, or printed from this source but further reproduction or distribution in any format is prohibited without written permission. http://dspace.mit.edu/handle/1721.1/7582 108 pages application/pdf Massachusetts Institute of Technology |
| spellingShingle | Mechanical Engineering. Choi, Changrak Robot motion planning with contact from global pseudo-inverse map |
| title | Robot motion planning with contact from global pseudo-inverse map |
| title_full | Robot motion planning with contact from global pseudo-inverse map |
| title_fullStr | Robot motion planning with contact from global pseudo-inverse map |
| title_full_unstemmed | Robot motion planning with contact from global pseudo-inverse map |
| title_short | Robot motion planning with contact from global pseudo-inverse map |
| title_sort | robot motion planning with contact from global pseudo inverse map |
| topic | Mechanical Engineering. |
| url | http://hdl.handle.net/1721.1/120251 |
| work_keys_str_mv | AT choichangrak robotmotionplanningwithcontactfromglobalpseudoinversemap |