Header logo is

Using Model Knowledge for Learning Inverse Dynamics

2010

Conference Paper

ei


In recent years, learning models from data has become an increasingly interesting tool for robotics, as it allows straightforward and accurate model approximation. However, in most robot learning approaches, the model is learned from scratch disregarding all prior knowledge about the system. For many complex robot systems, available prior knowledge from advanced physics-based modeling techniques can entail valuable information for model learning that may result in faster learning speed, higher accuracy and better generalization. In this paper, we investigate how parametric physical models (e.g., obtained from rigid body dynamics) can be used to improve the learning performance, and, especially, how semiparametric regression methods can be applied in this context. We present two possible semiparametric regression approaches, where the knowledge of the physical model can either become part of the mean function or of the kernel in a nonparametric Gaussian process regression. We compare the learning performance o f these methods first on sampled data and, subsequently, apply the obtained inverse dynamics models in tracking control on a real Barrett WAM. The results show that the semiparametric models learned with rigid body dynamics as prior outperform the standard rigid body dynamics models on real data while generalizing better for unknown parts of the state space.

Author(s): Nguyen-Tuong, D. and Peters, J.
Journal: Proceedings of the 2010 IEEE International Conference on Robotics and Automation (ICRA 2010)
Pages: 2677-2682
Year: 2010
Month: May
Day: 0
Publisher: IEEE

Department(s): Empirical Inference
Bibtex Type: Conference Paper (inproceedings)

DOI: 10.1109/ROBOT.2010.5509858
Event Name: 2010 IEEE International Conference on Robotics and Automation (ICRA 2010)
Event Place: Anchorage, AK, USA

Address: Piscataway, NJ, USA
Digital: 0
Institution: Institute of Electrical and Electronics Engineers
ISBN: 978-1-424-45038-1
Language: en
Organization: Max-Planck-Gesellschaft
School: Biologische Kybernetik

Links: PDF
Web

BibTex

@inproceedings{6232,
  title = {Using Model Knowledge for Learning Inverse Dynamics},
  author = {Nguyen-Tuong, D. and Peters, J.},
  journal = {Proceedings of the 2010 IEEE International Conference on Robotics and Automation (ICRA 2010)},
  pages = {2677-2682},
  publisher = {IEEE},
  organization = {Max-Planck-Gesellschaft},
  institution = {Institute of Electrical and Electronics Engineers},
  school = {Biologische Kybernetik},
  address = {Piscataway, NJ, USA},
  month = may,
  year = {2010},
  doi = {10.1109/ROBOT.2010.5509858},
  month_numeric = {5}
}