Efficient Planning in Non-Gaussian Belief Spaces and Its Application to Robot Grasping
© Springer International Publishing Switzerland 2017. The limited nature of robot sensors make many important robotics problems partially observable. These problems may require the system to perform complex information-gathering operations. One approach to solving these problems is to create plans i...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
Springer International Publishing
2021
|
| Online Access: | https://hdl.handle.net/1721.1/137700 |
| _version_ | 1826193252195237888 |
|---|---|
| author | Platt, Robert Kaelbling, Leslie Lozano-Perez, Tomas Tedrake, Russ |
| author2 | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
| author_facet | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Platt, Robert Kaelbling, Leslie Lozano-Perez, Tomas Tedrake, Russ |
| author_sort | Platt, Robert |
| collection | MIT |
| description | © Springer International Publishing Switzerland 2017. The limited nature of robot sensors make many important robotics problems partially observable. These problems may require the system to perform complex information-gathering operations. One approach to solving these problems is to create plans in belief-space, the space of probability distributions over the under-lying state of the system. The belief-space plan encodes a strategy for performing a task while gaining information as necessary. Most approaches to belief-space planning rely upon representing belief state in a particular way (typically as a Gaussian). Unfortunately, this can lead to large errors between the assumed density representation of belief state and the true belief state. This paper proposes a new sample-based approach to belief-space planning that has fixed computational complexity while allowing arbitrary implementations of Bayes filtering to be used to track belief state. The approach is illustrated in the context of a simple example and compared to a prior approach. Then, we propose an application of the technique to an instance of the grasp synthesis problem where a robot must simultaneously localize and grasp an object given initially uncertain object parameters by planning information-gathering behavior. Experimental results are presented that demonstrate the approach to be capable of actively localizing and grasping boxes that are presented to the robot in uncertain and hard-to-localize configurations. |
| first_indexed | 2024-09-23T09:36:23Z |
| format | Article |
| id | mit-1721.1/137700 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T09:36:23Z |
| publishDate | 2021 |
| publisher | Springer International Publishing |
| record_format | dspace |
| spelling | mit-1721.1/1377002023-02-10T18:54:06Z Efficient Planning in Non-Gaussian Belief Spaces and Its Application to Robot Grasping Platt, Robert Kaelbling, Leslie Lozano-Perez, Tomas Tedrake, Russ Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory © Springer International Publishing Switzerland 2017. The limited nature of robot sensors make many important robotics problems partially observable. These problems may require the system to perform complex information-gathering operations. One approach to solving these problems is to create plans in belief-space, the space of probability distributions over the under-lying state of the system. The belief-space plan encodes a strategy for performing a task while gaining information as necessary. Most approaches to belief-space planning rely upon representing belief state in a particular way (typically as a Gaussian). Unfortunately, this can lead to large errors between the assumed density representation of belief state and the true belief state. This paper proposes a new sample-based approach to belief-space planning that has fixed computational complexity while allowing arbitrary implementations of Bayes filtering to be used to track belief state. The approach is illustrated in the context of a simple example and compared to a prior approach. Then, we propose an application of the technique to an instance of the grasp synthesis problem where a robot must simultaneously localize and grasp an object given initially uncertain object parameters by planning information-gathering behavior. Experimental results are presented that demonstrate the approach to be capable of actively localizing and grasping boxes that are presented to the robot in uncertain and hard-to-localize configurations. 2021-11-08T16:28:05Z 2021-11-08T16:28:05Z 2016-08-26 2019-06-04T14:20:10Z Article http://purl.org/eprint/type/ConferencePaper 1610-7438 1610-742X https://hdl.handle.net/1721.1/137700 Platt, Robert, Kaelbling, Leslie, Lozano-Perez, Tomas and Tedrake, Russ. 2016. "Efficient Planning in Non-Gaussian Belief Spaces and Its Application to Robot Grasping." en 10.1007/978-3-319-29363-9_15 Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf Springer International Publishing MIT web domain |
| spellingShingle | Platt, Robert Kaelbling, Leslie Lozano-Perez, Tomas Tedrake, Russ Efficient Planning in Non-Gaussian Belief Spaces and Its Application to Robot Grasping |
| title | Efficient Planning in Non-Gaussian Belief Spaces and Its Application to Robot Grasping |
| title_full | Efficient Planning in Non-Gaussian Belief Spaces and Its Application to Robot Grasping |
| title_fullStr | Efficient Planning in Non-Gaussian Belief Spaces and Its Application to Robot Grasping |
| title_full_unstemmed | Efficient Planning in Non-Gaussian Belief Spaces and Its Application to Robot Grasping |
| title_short | Efficient Planning in Non-Gaussian Belief Spaces and Its Application to Robot Grasping |
| title_sort | efficient planning in non gaussian belief spaces and its application to robot grasping |
| url | https://hdl.handle.net/1721.1/137700 |
| work_keys_str_mv | AT plattrobert efficientplanninginnongaussianbeliefspacesanditsapplicationtorobotgrasping AT kaelblingleslie efficientplanninginnongaussianbeliefspacesanditsapplicationtorobotgrasping AT lozanopereztomas efficientplanninginnongaussianbeliefspacesanditsapplicationtorobotgrasping AT tedrakeruss efficientplanninginnongaussianbeliefspacesanditsapplicationtorobotgrasping |