Switched Latent Force Models for Movement Segmentation

Mauricio A. Álvarez, Jan Peters, Bernhard SchölkopfNeil D. Lawrence
,  23:55-63, 2010.

Abstract

Latent force models encode the interaction between multiple related dynamical systems in the form of a kernel or covariance function. Each variable to be modeled is represented as the output of a differential equation and each differential equation is driven by a weighted sum of latent functions with uncertainty given by a Gaussian process prior. In this paper we consider employing the latent force model framework for the problem of determining robot motor primitives. To deal with discontinuities in the dynamical systems or the latent driving force we intro- duce an extension of the basic latent force model, that switches between different latent functions and potentially different dynamical systems. This creates a versatile representation for robot movements that can capture discrete changes and non-linearities in the dynamics. We give illustrative examples on both synthetic data and for striking movements recorded using a Barrett WAM robot as haptic in- put device. Our inspiration is robot motor primitives, but we expect our model to have wide application for dynamical systems including models for human motion capture data and systems biology.

Cite this Paper


BibTeX
@InProceedings{pmlr-v-alvarez-switched10, title = {Switched Latent Force Models for Movement Segmentation}, author = {Mauricio A. Álvarez and Jan Peters and Bernhard Schölkopf and Neil D. Lawrence}, pages = {55--63}, year = {}, editor = {}, volume = {23}, address = {Cambridge, MA}, pdf = {http://books.nips.cc/papers/files/nips23/NIPS2010_1222.pdf}, url = {http://inverseprobability.com/publications/alvarez-switched10.html}, abstract = {Latent force models encode the interaction between multiple related dynamical systems in the form of a kernel or covariance function. Each variable to be modeled is represented as the output of a differential equation and each differential equation is driven by a weighted sum of latent functions with uncertainty given by a Gaussian process prior. In this paper we consider employing the latent force model framework for the problem of determining robot motor primitives. To deal with discontinuities in the dynamical systems or the latent driving force we intro- duce an extension of the basic latent force model, that switches between different latent functions and potentially different dynamical systems. This creates a versatile representation for robot movements that can capture discrete changes and non-linearities in the dynamics. We give illustrative examples on both synthetic data and for striking movements recorded using a Barrett WAM robot as haptic in- put device. Our inspiration is robot motor primitives, but we expect our model to have wide application for dynamical systems including models for human motion capture data and systems biology.} }
Endnote
%0 Conference Paper %T Switched Latent Force Models for Movement Segmentation %A Mauricio A. Álvarez %A Jan Peters %A Bernhard Schölkopf %A Neil D. Lawrence %B %C Proceedings of Machine Learning Research %D %E %F pmlr-v-alvarez-switched10 %I PMLR %J Proceedings of Machine Learning Research %P 55--63 %U http://inverseprobability.com %V %W PMLR %X Latent force models encode the interaction between multiple related dynamical systems in the form of a kernel or covariance function. Each variable to be modeled is represented as the output of a differential equation and each differential equation is driven by a weighted sum of latent functions with uncertainty given by a Gaussian process prior. In this paper we consider employing the latent force model framework for the problem of determining robot motor primitives. To deal with discontinuities in the dynamical systems or the latent driving force we intro- duce an extension of the basic latent force model, that switches between different latent functions and potentially different dynamical systems. This creates a versatile representation for robot movements that can capture discrete changes and non-linearities in the dynamics. We give illustrative examples on both synthetic data and for striking movements recorded using a Barrett WAM robot as haptic in- put device. Our inspiration is robot motor primitives, but we expect our model to have wide application for dynamical systems including models for human motion capture data and systems biology.
RIS
TY - CPAPER TI - Switched Latent Force Models for Movement Segmentation AU - Mauricio A. Álvarez AU - Jan Peters AU - Bernhard Schölkopf AU - Neil D. Lawrence BT - PY - DA - ED - ID - pmlr-v-alvarez-switched10 PB - PMLR SP - 55 DP - PMLR EP - 63 L1 - http://books.nips.cc/papers/files/nips23/NIPS2010_1222.pdf UR - http://inverseprobability.com/publications/alvarez-switched10.html AB - Latent force models encode the interaction between multiple related dynamical systems in the form of a kernel or covariance function. Each variable to be modeled is represented as the output of a differential equation and each differential equation is driven by a weighted sum of latent functions with uncertainty given by a Gaussian process prior. In this paper we consider employing the latent force model framework for the problem of determining robot motor primitives. To deal with discontinuities in the dynamical systems or the latent driving force we intro- duce an extension of the basic latent force model, that switches between different latent functions and potentially different dynamical systems. This creates a versatile representation for robot movements that can capture discrete changes and non-linearities in the dynamics. We give illustrative examples on both synthetic data and for striking movements recorded using a Barrett WAM robot as haptic in- put device. Our inspiration is robot motor primitives, but we expect our model to have wide application for dynamical systems including models for human motion capture data and systems biology. ER -
APA
Álvarez, M.A., Peters, J., Schölkopf, B. & Lawrence, N.D.. (). Switched Latent Force Models for Movement Segmentation. , in PMLR :55-63

Related Material