Reference : Position identification in force-guided peg-in-hole assembly tasks
Scientific congresses, symposiums and conference proceedings : Paper published in a journal
Engineering, computing & technology : Mechanical engineering
Position identification in force-guided peg-in-hole assembly tasks
Ibrahim, Jasim []
Plapper, Peter mailto [University of Luxembourg > Faculty of Science, Technology and Communication (FSTC) > Engineering Research Unit >]
Voos, Holger mailto [University of Luxembourg > Faculty of Science, Technology and Communication (FSTC) > Engineering Research Unit >]
Procedia CIRP
Yes (verified by ORBilu)
The Netherlands
5th CATS 2014 - CIRP Conference on Assembly Technologies and Systems
13-11-2014 to 14-11-2014
[en] Force-guided robots ; peg-in-hole assembly ; position uncertainty
[en] Position uncertainty is inevitable in many force-guided robotic assembly tasks. Such uncertainty can cause a significant delay, extra energy expenditure, and may even results in detriments to the mated parts or the robot itself. This article suggests a strategy for identifying the accurate hole position in force-guided robotic peg-in-hole assembly tasks through employing only the captured wrench (the Cartesian forces and torques) signals of the manipulated. In the framework of using the Contact-State (CS) modeling for such robotic tasks, the identification of the hole position is realized through detecting the CS that corresponds for the phase of the peg-on-hole, that is the phase in which the peg is located precisely on the hole. Expectation Maximization-based Gaussian Mixtures Model (EM-GMM) CS modeling scheme is employed in detecting the CS corresponding for the peg-on-hole phase. Only the wrench signals are used in modeling and detecting the phases of the assembly process. The considered peg-in-hole assembly process starts from free space and as soon as the peg touches the environment with missing the hole, a spiral search path is followed that would survey the whole environment surface. When the CS of the peg-on-hole is detected, the hole position is identified. Experiments are conducted on a KUKA Lightweight Robot (LWR) doing typical peg-in-hole assembly tasks. Multiple hole positions are considered and excellent performance of the proposed identification strategy is shown.

File(s) associated to this reference

Fulltext file(s):

Open access
2014-CIRP.pdfPublisher postprint599.33 kBView/Open

Bookmark and Share SFX Query

All documents in ORBilu are protected by a user license.