Provably Safe Robot Navigation with Obstacle Uncertainty

© 2017 MIT Press Journals. All rights reserved. As drones and autonomous cars become more widespread it is becoming increasingly important that robots can operate safely under realistic conditions. The noisy information fed into real systems means that robots must use estimates of the environment to...

Full description

Bibliographic Details
Main Authors: Axelrod, Brian, Kaelbling, Leslie, Lozano-Perez, Tomas
Other Authors: Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory
Format: Article
Language:English
Published: Robotics: Science and Systems Foundation 2021
Online Access:https://hdl.handle.net/1721.1/137695
_version_ 1826214967648452608
author Axelrod, Brian
Kaelbling, Leslie
Lozano-Perez, Tomas
author2 Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory
author_facet Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory
Axelrod, Brian
Kaelbling, Leslie
Lozano-Perez, Tomas
author_sort Axelrod, Brian
collection MIT
description © 2017 MIT Press Journals. All rights reserved. As drones and autonomous cars become more widespread it is becoming increasingly important that robots can operate safely under realistic conditions. The noisy information fed into real systems means that robots must use estimates of the environment to plan navigation. Efficiently guaranteeing that the resulting motion plans are safe under these circumstances has proved difficult. We examine how to guarantee that a trajectory or policy is safe with only imperfect observations of the environment. We examine the implications of various mathematical formalisms of safety and arrive at a mathematical notion of safety of a long-term execution, even when conditioned on observational information. We present efficient algorithms that can prove that trajectories or policies are safe with much tighter bounds than in previous work. Notably, the complexity of the environment does not affect our method's ability to evaluate if a trajectory or policy is safe. We then use these safety checking methods to design a safe variant of the RRT planning algorithm.
first_indexed 2024-09-23T16:14:04Z
format Article
id mit-1721.1/137695
institution Massachusetts Institute of Technology
language English
last_indexed 2024-09-23T16:14:04Z
publishDate 2021
publisher Robotics: Science and Systems Foundation
record_format dspace
spelling mit-1721.1/1376952023-01-10T15:04:42Z Provably Safe Robot Navigation with Obstacle Uncertainty Axelrod, Brian Kaelbling, Leslie Lozano-Perez, Tomas Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science © 2017 MIT Press Journals. All rights reserved. As drones and autonomous cars become more widespread it is becoming increasingly important that robots can operate safely under realistic conditions. The noisy information fed into real systems means that robots must use estimates of the environment to plan navigation. Efficiently guaranteeing that the resulting motion plans are safe under these circumstances has proved difficult. We examine how to guarantee that a trajectory or policy is safe with only imperfect observations of the environment. We examine the implications of various mathematical formalisms of safety and arrive at a mathematical notion of safety of a long-term execution, even when conditioned on observational information. We present efficient algorithms that can prove that trajectories or policies are safe with much tighter bounds than in previous work. Notably, the complexity of the environment does not affect our method's ability to evaluate if a trajectory or policy is safe. We then use these safety checking methods to design a safe variant of the RRT planning algorithm. 2021-11-08T16:07:48Z 2021-11-08T16:07:48Z 2017-07-12 2019-06-04T14:56:17Z Article http://purl.org/eprint/type/ConferencePaper https://hdl.handle.net/1721.1/137695 Axelrod, Brian, Kaelbling, Leslie and Lozano-Perez, Tomas. 2017. "Provably Safe Robot Navigation with Obstacle Uncertainty." en 10.15607/rss.2017.xiii.023 Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf Robotics: Science and Systems Foundation MIT web domain
spellingShingle Axelrod, Brian
Kaelbling, Leslie
Lozano-Perez, Tomas
Provably Safe Robot Navigation with Obstacle Uncertainty
title Provably Safe Robot Navigation with Obstacle Uncertainty
title_full Provably Safe Robot Navigation with Obstacle Uncertainty
title_fullStr Provably Safe Robot Navigation with Obstacle Uncertainty
title_full_unstemmed Provably Safe Robot Navigation with Obstacle Uncertainty
title_short Provably Safe Robot Navigation with Obstacle Uncertainty
title_sort provably safe robot navigation with obstacle uncertainty
url https://hdl.handle.net/1721.1/137695
work_keys_str_mv AT axelrodbrian provablysaferobotnavigationwithobstacleuncertainty
AT kaelblingleslie provablysaferobotnavigationwithobstacleuncertainty
AT lozanopereztomas provablysaferobotnavigationwithobstacleuncertainty