Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

A Velocity Observer Based on Friction Adaptation

Lenain, Roland ; Robertsson, Anders LU ; Johansson, Rolf LU orcid ; Shiriaev, Anton and Berducat, Michel (2008) IEEE International Conference on Robotics and Automation, 2008 p.3365-3370
Abstract
Control of robotic systems subject to friction phenomena is an important issue since growing demands on accuracy require elimination of friction disturbances. If several modelse.g., friction model, rigid-body dynamicsare required to describe the behavior with high precision, each model requires the knowledge of numerous parameters (perhaps time-varying) as well as an increased number of signals and sensors. In order to tackle this double limitation, an observer is proposed, addressing both the problem of velocity reconstruction and friction estimation in the joint of an inverted pendulum. Firstly, an adaptive two-level velocity observer is defined to reconstruct relevant unmeasured velocities, using estimation of the friction model error.... (More)
Control of robotic systems subject to friction phenomena is an important issue since growing demands on accuracy require elimination of friction disturbances. If several modelse.g., friction model, rigid-body dynamicsare required to describe the behavior with high precision, each model requires the knowledge of numerous parameters (perhaps time-varying) as well as an increased number of signals and sensors. In order to tackle this double limitation, an observer is proposed, addressing both the problem of velocity reconstruction and friction estimation in the joint of an inverted pendulum. Firstly, an adaptive two-level velocity observer is defined to reconstruct relevant unmeasured velocities, using estimation of the friction model error. Secondly, the observer is exploited for model-based friction compensation. Capabilities of the algorithm proposed are demonstrated by means of experiments on the Furuta pendulum. (Less)
Please use this url to cite or link to this publication:
author
; ; ; and
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
host publication
Proc. 2008 IEEE International Conference on Robotics and Automation (ICRA2008)
pages
3365 - 3370
conference name
IEEE International Conference on Robotics and Automation, 2008
conference location
Pasadena, CA, United States
conference dates
2008-05-19 - 2008-05-23
external identifiers
  • scopus:51649085889
project
RobotLab LTH
language
English
LU publication?
yes
id
643ed5f1-b9a1-4ac3-91ac-d7dd16bda7d2 (old id 1444529)
date added to LUP
2016-04-04 14:01:05
date last changed
2023-04-09 09:02:51
@inproceedings{643ed5f1-b9a1-4ac3-91ac-d7dd16bda7d2,
  abstract     = {{Control of robotic systems subject to friction phenomena is an important issue since growing demands on accuracy require elimination of friction disturbances. If several modelse.g., friction model, rigid-body dynamicsare required to describe the behavior with high precision, each model requires the knowledge of numerous parameters (perhaps time-varying) as well as an increased number of signals and sensors. In order to tackle this double limitation, an observer is proposed, addressing both the problem of velocity reconstruction and friction estimation in the joint of an inverted pendulum. Firstly, an adaptive two-level velocity observer is defined to reconstruct relevant unmeasured velocities, using estimation of the friction model error. Secondly, the observer is exploited for model-based friction compensation. Capabilities of the algorithm proposed are demonstrated by means of experiments on the Furuta pendulum.}},
  author       = {{Lenain, Roland and Robertsson, Anders and Johansson, Rolf and Shiriaev, Anton and Berducat, Michel}},
  booktitle    = {{Proc. 2008 IEEE International Conference on Robotics and Automation (ICRA2008)}},
  language     = {{eng}},
  pages        = {{3365--3370}},
  title        = {{A Velocity Observer Based on Friction Adaptation}},
  year         = {{2008}},
}