Real-time control of the endeffector of a humanoid robot in external
coordinates requires computationally efficient solutions of the inverse
kinematics problem. In this context, this paper investigates learning
of inverse kinematics for resolved motion rate control (RMRC) employing
an optimization criterion to resolve kinematic redundancies. Our
learning approach is based on the key observations that learning
an inverse of a non uniquely invertible function can be accomplished
by augmenting the input representation to the inverse model and by
using a spatially localized learning approach. We apply this strategy
to inverse kinematics learning and demonstrate how a recently developed
statistical learning algorithm, Locally Weighted Projection Regression,
allows efficient learning of inverse kinematic mappings in an incremental
fashion even when input spaces become rather high dimensional. The
resulting performance of the inverse kinematics is comparable to
Liegeois (1) analytical pseudo inverse with optimization. Our results
are illustrated with a 30 degree-of-freedom humanoid robot.
%0 Journal Article
%1 DSouza:2001
%A D'Souza, A.
%A Vijayakumar, S.
%A Schaal, S.
%D 2001
%J IEEE International Conference on Intelligent Robots and Systems (IROS
2001)
%K control dimensional high humanoid inverse kinematics learning local motion null optimization rate resolved robotics space
%P 298-303
%T Learning inverse kinematics
%V 1
%X Real-time control of the endeffector of a humanoid robot in external
coordinates requires computationally efficient solutions of the inverse
kinematics problem. In this context, this paper investigates learning
of inverse kinematics for resolved motion rate control (RMRC) employing
an optimization criterion to resolve kinematic redundancies. Our
learning approach is based on the key observations that learning
an inverse of a non uniquely invertible function can be accomplished
by augmenting the input representation to the inverse model and by
using a spatially localized learning approach. We apply this strategy
to inverse kinematics learning and demonstrate how a recently developed
statistical learning algorithm, Locally Weighted Projection Regression,
allows efficient learning of inverse kinematic mappings in an incremental
fashion even when input spaces become rather high dimensional. The
resulting performance of the inverse kinematics is comparable to
Liegeois (1) analytical pseudo inverse with optimization. Our results
are illustrated with a 30 degree-of-freedom humanoid robot.
@article{DSouza:2001,
abstract = {Real-time control of the endeffector of a humanoid robot in external
coordinates requires computationally efficient solutions of the inverse
kinematics problem. In this context, this paper investigates learning
of inverse kinematics for resolved motion rate control (RMRC) employing
an optimization criterion to resolve kinematic redundancies. Our
learning approach is based on the key observations that learning
an inverse of a non uniquely invertible function can be accomplished
by augmenting the input representation to the inverse model and by
using a spatially localized learning approach. We apply this strategy
to inverse kinematics learning and demonstrate how a recently developed
statistical learning algorithm, Locally Weighted Projection Regression,
allows efficient learning of inverse kinematic mappings in an incremental
fashion even when input spaces become rather high dimensional. The
resulting performance of the inverse kinematics is comparable to
Liegeois ([1]) analytical pseudo inverse with optimization. Our results
are illustrated with a 30 degree-of-freedom humanoid robot.},
added-at = {2009-06-26T15:25:19.000+0200},
author = {D'Souza, A. and Vijayakumar, S. and Schaal, S.},
biburl = {https://www.bibsonomy.org/bibtex/26f7fce9cc6d720704e9d17311bd50622/butz},
description = {diverse cognitive systems bib},
interhash = {697e45ad5ffbc5ff190ae675769e5fb1},
intrahash = {6f7fce9cc6d720704e9d17311bd50622},
journal = {IEEE International Conference on Intelligent Robots and Systems (IROS
2001)},
keywords = {control dimensional high humanoid inverse kinematics learning local motion null optimization rate resolved robotics space},
owner = {butz},
pages = {298-303},
timestamp = {2009-06-26T15:25:26.000+0200},
title = {Learning inverse kinematics},
volume = 1,
year = 2001
}