On-line dynamic model learning for manipulator control

Joseph Sun De La Cruz, Ergun Calisgan, Dana Kulić, William Owen, Elizabeth A. Croft

Research output: Chapter in Book/Report/Conference proceedingConference PaperOther

3 Citations (Scopus)


This paper proposes an approach for online learning of the dynamic model of a robot manipulator. The dynamic model is formulated as a weighted sum of locally linear models, and Locally Weighted Projection Regression (LWPR) is used to learn the models based on training data obtained during operation. The LWPR model can be initialized with partial knowledge of rigid body parameters to improve the initial performance. The resulting dynamic model is used to implement a model-based controller. Both feedforward and feedback configurations are investigated. The proposed approach is tested on an industrial robot, and shown to outperform independent joint and fixed model-based control.

Original languageEnglish
Title of host publicationSYROCO 2012 Preprints - 10th IFAC Symposium on Robot Control
PublisherIFAC Secretariat
Number of pages6
ISBN (Print)9783902823113
Publication statusPublished - 1 Jan 2012
Externally publishedYes
EventIFAC Symposium on Robot Control 2012 - Dubrovnik, Croatia
Duration: 5 Sep 20127 Sep 2012
Conference number: 10th


ConferenceIFAC Symposium on Robot Control 2012
Abbreviated title SYROCO 2012


  • Learning control
  • Robot dynamics
  • Robotic manipulators

Cite this