Fast-Reactive Probabilistic Motion Planning for High-Dimensional Robots
Abstract Many real-world robotic operations that involve high-dimensional humanoid robots require fast reaction to plan disturbances and probabilistic guarantees over collision risks, whereas most probabilistic motion planning approaches developed for car-like robots cannot be directl...
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
Springer Singapore
2021
|
| Online Access: | https://hdl.handle.net/1721.1/136734 |
| _version_ | 1826190631241777152 |
|---|---|
| author | Dai, Siyu Hofmann, Andreas Williams, Brian |
| author2 | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
| author_facet | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Dai, Siyu Hofmann, Andreas Williams, Brian |
| author_sort | Dai, Siyu |
| collection | MIT |
| description | Abstract
Many real-world robotic operations that involve high-dimensional humanoid robots require fast reaction to plan disturbances and probabilistic guarantees over collision risks, whereas most probabilistic motion planning approaches developed for car-like robots cannot be directly applied to high-dimensional robots. In this paper, we present probabilistic Chekov (p-Chekov), a fast-reactive motion planning system that can provide safety guarantees for high-dimensional robots suffering from process noises and observation noises. Leveraging recent advances in machine learning as well as our previous work in deterministic motion planning that integrated trajectory optimization into a sparse roadmap framework, p-Chekov demonstrates its superiority in terms of collision avoidance ability and planning speed in high-dimensional robotic motion planning tasks in complex environments without the convexification of obstacles. Comprehensive theoretical and empirical analysis provided in this paper shows that p-Chekov can effectively satisfy user-specified chance constraints over collision risk in practical robotic manipulation tasks. |
| first_indexed | 2024-09-23T08:43:17Z |
| format | Article |
| id | mit-1721.1/136734 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T08:43:17Z |
| publishDate | 2021 |
| publisher | Springer Singapore |
| record_format | dspace |
| spelling | mit-1721.1/1367342023-12-08T20:05:02Z Fast-Reactive Probabilistic Motion Planning for High-Dimensional Robots Dai, Siyu Hofmann, Andreas Williams, Brian Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Abstract Many real-world robotic operations that involve high-dimensional humanoid robots require fast reaction to plan disturbances and probabilistic guarantees over collision risks, whereas most probabilistic motion planning approaches developed for car-like robots cannot be directly applied to high-dimensional robots. In this paper, we present probabilistic Chekov (p-Chekov), a fast-reactive motion planning system that can provide safety guarantees for high-dimensional robots suffering from process noises and observation noises. Leveraging recent advances in machine learning as well as our previous work in deterministic motion planning that integrated trajectory optimization into a sparse roadmap framework, p-Chekov demonstrates its superiority in terms of collision avoidance ability and planning speed in high-dimensional robotic motion planning tasks in complex environments without the convexification of obstacles. Comprehensive theoretical and empirical analysis provided in this paper shows that p-Chekov can effectively satisfy user-specified chance constraints over collision risk in practical robotic manipulation tasks. 2021-10-29T13:52:37Z 2021-10-29T13:52:37Z 2021-10-15 2021-10-15T03:14:57Z Article http://purl.org/eprint/type/JournalArticle https://hdl.handle.net/1721.1/136734 SN Computer Science. 2021 Oct 15;2(6):484 en https://doi.org/10.1007/s42979-021-00878-0 Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use. The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd application/pdf Springer Singapore Springer Singapore |
| spellingShingle | Dai, Siyu Hofmann, Andreas Williams, Brian Fast-Reactive Probabilistic Motion Planning for High-Dimensional Robots |
| title | Fast-Reactive Probabilistic Motion Planning for High-Dimensional Robots |
| title_full | Fast-Reactive Probabilistic Motion Planning for High-Dimensional Robots |
| title_fullStr | Fast-Reactive Probabilistic Motion Planning for High-Dimensional Robots |
| title_full_unstemmed | Fast-Reactive Probabilistic Motion Planning for High-Dimensional Robots |
| title_short | Fast-Reactive Probabilistic Motion Planning for High-Dimensional Robots |
| title_sort | fast reactive probabilistic motion planning for high dimensional robots |
| url | https://hdl.handle.net/1721.1/136734 |
| work_keys_str_mv | AT daisiyu fastreactiveprobabilisticmotionplanningforhighdimensionalrobots AT hofmannandreas fastreactiveprobabilisticmotionplanningforhighdimensionalrobots AT williamsbrian fastreactiveprobabilisticmotionplanningforhighdimensionalrobots |