Parallel Optimization of Motion Controllers via Policy Iteration
Abstract
This paper describes a policy iteration algorithm for optimizing the performance of a harmonic function-based controller with respect to a user-defined index. Value functions are represented as poten(cid:173) tial distributions over the problem domain, being control policies represented as gradient fields over the same domain. All interme(cid:173) diate policies are intrinsically safe, i.e. collisions are not promoted during the adaptation process. The algorithm has efficient imple(cid:173) mentation in parallel SIMD architectures. One potential applica(cid:173) tion - travel distance minimization - illustrates its usefulness.
Cite
Text
Jr. et al. "Parallel Optimization of Motion Controllers via Policy Iteration." Neural Information Processing Systems, 1995.Markdown
[Jr. et al. "Parallel Optimization of Motion Controllers via Policy Iteration." Neural Information Processing Systems, 1995.](https://mlanthology.org/neurips/1995/jr1995neurips-parallel/)BibTeX
@inproceedings{jr1995neurips-parallel,
title = {{Parallel Optimization of Motion Controllers via Policy Iteration}},
author = {Jr., Jefferson A. Coelho and Sitaraman, R. and Grupen, Roderic A.},
booktitle = {Neural Information Processing Systems},
year = {1995},
pages = {996-1002},
url = {https://mlanthology.org/neurips/1995/jr1995neurips-parallel/}
}