Enabling Topological Planning with Monocular Vision
© 2020 IEEE. Topological strategies for navigation meaningfully reduce the space of possible actions available to a robot, allowing use of heuristic priors or learning to enable computationally efficient, intelligent planning. The challenges in estimating structure with monocular SLAM in low texture...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021
|
| Online Access: | https://hdl.handle.net/1721.1/137314 |
| _version_ | 1826207888200171520 |
|---|---|
| author | Stein, Gregory Joseph Bradley, Christopher Preston, Victoria Roy, Nicholas |
| author2 | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
| author_facet | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Stein, Gregory Joseph Bradley, Christopher Preston, Victoria Roy, Nicholas |
| author_sort | Stein, Gregory Joseph |
| collection | MIT |
| description | © 2020 IEEE. Topological strategies for navigation meaningfully reduce the space of possible actions available to a robot, allowing use of heuristic priors or learning to enable computationally efficient, intelligent planning. The challenges in estimating structure with monocular SLAM in low texture or highly cluttered environments have precluded its use for topological planning in the past. We propose a robust sparse map representation that can be built with monocular vision and overcomes these shortcomings. Using a learned sensor, we estimate high-level structure of an environment from streaming images by detecting sparse vertices (e.g., boundaries of walls) and reasoning about the structure between them. We also estimate the known free space in our map, a necessary feature for planning through previously unknown environments. We show that our mapping technique can be used on real data and is sufficient for planning and exploration in simulated multi-agent search and learned subgoal planning applications. |
| first_indexed | 2024-09-23T13:56:32Z |
| format | Article |
| id | mit-1721.1/137314 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T13:56:32Z |
| publishDate | 2021 |
| publisher | IEEE |
| record_format | dspace |
| spelling | mit-1721.1/1373142022-10-01T18:09:19Z Enabling Topological Planning with Monocular Vision Stein, Gregory Joseph Bradley, Christopher Preston, Victoria Roy, Nicholas Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory © 2020 IEEE. Topological strategies for navigation meaningfully reduce the space of possible actions available to a robot, allowing use of heuristic priors or learning to enable computationally efficient, intelligent planning. The challenges in estimating structure with monocular SLAM in low texture or highly cluttered environments have precluded its use for topological planning in the past. We propose a robust sparse map representation that can be built with monocular vision and overcomes these shortcomings. Using a learned sensor, we estimate high-level structure of an environment from streaming images by detecting sparse vertices (e.g., boundaries of walls) and reasoning about the structure between them. We also estimate the known free space in our map, a necessary feature for planning through previously unknown environments. We show that our mapping technique can be used on real data and is sufficient for planning and exploration in simulated multi-agent search and learned subgoal planning applications. Office of Naval Research (Contract N00014-17-1-2699) 2021-11-03T20:16:02Z 2021-11-03T20:16:02Z 2020-09 2021-05-03T18:42:32Z Article http://purl.org/eprint/type/ConferencePaper https://hdl.handle.net/1721.1/137314 Stein, Gregory Joseph, Bradley, Christopher, Preston, Victoria and Roy, Nicholas. 2020. "Enabling Topological Planning with Monocular Vision." Proceedings - IEEE International Conference on Robotics and Automation. en http://dx.doi.org/10.1109/ICRA40945.2020.9197484 Proceedings - IEEE International Conference on Robotics and Automation Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf IEEE MIT web domain |
| spellingShingle | Stein, Gregory Joseph Bradley, Christopher Preston, Victoria Roy, Nicholas Enabling Topological Planning with Monocular Vision |
| title | Enabling Topological Planning with Monocular Vision |
| title_full | Enabling Topological Planning with Monocular Vision |
| title_fullStr | Enabling Topological Planning with Monocular Vision |
| title_full_unstemmed | Enabling Topological Planning with Monocular Vision |
| title_short | Enabling Topological Planning with Monocular Vision |
| title_sort | enabling topological planning with monocular vision |
| url | https://hdl.handle.net/1721.1/137314 |
| work_keys_str_mv | AT steingregoryjoseph enablingtopologicalplanningwithmonocularvision AT bradleychristopher enablingtopologicalplanningwithmonocularvision AT prestonvictoria enablingtopologicalplanningwithmonocularvision AT roynicholas enablingtopologicalplanningwithmonocularvision |