Publication:
Episodic Non-Markov Localization: Reasoning About Short-Term and Long-Term Features

Simple item page
dc.contributor.authorBiswas, Joydeep
dc.contributor.authorVeloso, Manuela M
dc.contributor.departmentUniversity of Massachusetts Amherst
dc.contributor.departmentCarnegie Mellon University
dc.date2023-09-23T16:05:24.000
dc.date.accessioned2024-04-26T09:34:09Z
dc.date.available2024-04-26T09:34:09Z
dc.date.issued2014-01-01
dc.description.abstractMarkov localization and its variants are widely used for localization of mobile robots. These methods assume Markov independence of observations, implying that observations made by a robot correspond to a static map. However, in real human environments, observations include occlusions due to unmapped objects like chairs and tables, and dynamic objects like humans. We introduce an episodic non-Markov localization algorithm that maintains estimates of the belief over the trajectory of the robot while explicitly reasoning about observations and their correlations arising from unmapped static objects, moving objects, as well as objects from the static map. Observations are classified as arising from longterm features, short-term features, or dynamic features, which correspond to mapped objects, unmapped static objects, and unmapped dynamic objects respectively. By detecting time steps along the robot’s trajectory where unmapped observations prior to such time steps are unrelated to those afterwards, nonMarkov localization limits the history of observations and pose estimates to “episodes” over which the belief is computed. We demonstrate non-Markov localization in challenging real world indoor and outdoor environments over multiple datasets, comparing it with alternative state-of-the-art approaches, showing it to be robust as well as accurate.
dc.description.pages3969–3974
dc.identifier.urihttps://hdl.handle.net/20.500.14394/9755
dc.relation.ispartof2014 IEEE International Conference
dc.relation.urlhttps://scholarworks.umass.edu/cgi/viewcontent.cgi?article=2328&context=cs_faculty_pubs&unstamped=1
dc.source.statuspublished
dc.subjectlocalization
dc.subjectrobotics
dc.subjectArtificial Intelligence and Robotics
dc.subjectComputer Sciences
dc.titleEpisodic Non-Markov Localization: Reasoning About Short-Term and Long-Term Features
dc.typearticle
dc.typearticle
digcom.contributor.authorBiswas, Joydeep
digcom.contributor.authorVeloso, Manuela M
digcom.identifiercs_faculty_pubs/1334
digcom.identifier.contextkey9086660
digcom.identifier.submissionpathcs_faculty_pubs/1334
dspace.entity.typePublication
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
icra2014_enml.pdf
Size:
712.46 KB
Format:
Adobe Portable Document Format
Thumbnail Image
Name: icra2014_enml.pdf
Size: 712.46 KB
Kind: Adobe PDF
Collections
Computer Science Department Faculty Publication Series