Fast, Robust Adaptive Control by Learning only Forward Models (1992)
Tags
Active Learning, Efficient Statistical Algorithms, Kd-trees and Ball-trees, Locally Weighted Learning, Memory-based Learning
Abstract
A large class of motor control tasks requires that on each cycle the conctroller is told its current state and must choose an action to achieve a specified, state-dependent, goal behaviour. This paper argues that the optimization of learning rate, the number of experimental control decisions before adequate performance is obtained, and robustness is of prime importance--if necessary at the expense of computation per control cycle and memory requirement. This is motivated by the observation that a robot which requires two thousand learning steps to achieve adequate performance, or a robot which occasionally gets stuck while learning, will always be undesirable, whereas moderate computational expense can be accommodated by increasingly powerful computer hardware. It is not unreasonable to assume the existence of inexpensive 100 Mflop controllers within a few years and so even processes with control cycles in the low tens of milliseconds will have millions of machine instructions in which to make their decisions. This paper outlines a learning control scheme which aims to make effective use of such computational power.
Full text
Download (application/pdf, 3.0 MB)
Approximate BibTeX Entry
@inproceedings{moore-fastrobust,
Month = {April},
Year = {1992},
Publisher = {Morgan Kaufmann},
Address = {340 Pine Street, 6th Fl., San Francisco, CA 94104},
Booktitle = {Advances in Neural Information Processing Systems},
Editor = {J. E. Moody and S. J. Hanson and R. P. L},
Author = {Andrew Moore},
Title = {Fast, Robust Adaptive Control by Learning only Forward Models}
}