Learning Global Direct Inverse Kinematics

Abstract

We introduce and demonstrate a bootstrap method for construction of an in(cid:173) verse function for the robot kinematic mapping using only sample configuration(cid:173) space/workspace data. Unsupervised learning (clustering) techniques are used on pre-image neighborhoods in order to learn to partition the configuration space into subsets over which the kinematic mapping is invertible. Supervised leam(cid:173) ing is then used separately on each of the partitions to approximate the inverse function. The ill-posed inverse kinematics function is thereby regularized, and a globa1 inverse kinematics solution for the wristless Puma manipulator is devel(cid:173) oped.

Cite

Text

DeMers and Kreutz-Delgado. "Learning Global Direct Inverse Kinematics." Neural Information Processing Systems, 1991.

Markdown

[DeMers and Kreutz-Delgado. "Learning Global Direct Inverse Kinematics." Neural Information Processing Systems, 1991.](https://mlanthology.org/neurips/1991/demers1991neurips-learning/)

BibTeX

@inproceedings{demers1991neurips-learning,
  title     = {{Learning Global Direct Inverse Kinematics}},
  author    = {DeMers, David and Kreutz-Delgado, Kenneth},
  booktitle = {Neural Information Processing Systems},
  year      = {1991},
  pages     = {589-595},
  url       = {https://mlanthology.org/neurips/1991/demers1991neurips-learning/}
}