We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:


References & Citations

DBLP - CS Bibliography


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Robotics

Title: Soft Tracking Using Contacts for Cluttered Objects to Perform Blind Object Retrieval

Abstract: Retrieving an object from cluttered spaces suchas cupboards, refrigerators, or bins requires tracking objects with limited or no visual sensing. In these scenarios, contact feedback is necessary to estimate the pose of the objects, yet the objects are movable while their shapes and number may be unknown, making the association of contacts with objects extremely difficult. While previous work has focused on multi-target tracking, the assumptions therein prohibit using prior methods given only the contact-sensing modality. Instead, this paper proposes the method Soft Tracking Using Contacts for Cluttered Objects (STUCCO) that tracks the belief over contact point locations and implicit object associations using a particle filter. This method allows ambiguous object associations of past contacts to be revised as new information becomes available. We apply STUCCO to the Blind Object Retrieval problem, where a target object of known shape but unknown pose must be retrieved from clutter. Our results suggest that our method outperforms baselines in four simulation environments, and on a real robot, where contact sensing is noisy. In simulation, we achieve grasp success of at least 65% on all environments while no baselines achieve over 5%.
Comments: 8 pages, 7 figures, 2 tables, and 3 algorithm blocks, accepted to RA-L, preprint
Subjects: Robotics (cs.RO)
Cite as: arXiv:2201.10434 [cs.RO]
  (or arXiv:2201.10434v1 [cs.RO] for this version)

Submission history

From: Sheng Zhong [view email]
[v1] Tue, 25 Jan 2022 16:31:58 GMT (997kb,D)

Link back to: arXiv, form interface, contact.