Direkt zum Inhalt springen

# Technical University of Munich

Making Robot Learning Controllable: A Case Study in Robot Navigation (bibtex)
Making Robot Learning Controllable: A Case Study in Robot Navigation (bibtex)
by A Kirsch, M Schweitzer and M Beetz
Abstract:
In many applications the performance of learned robot controllers drags behind those of the respective hand-coded ones. In our view, this situation is caused not mainly by deficiencies of the learning algorithms but rather by an insufficient embedding of learning in robot control programs. This paper presents a case study in which RoLL, a robot control language that allows for explicit representations of learning problems, is applied to learning robot navigation tasks. The case study shows that RoLL's constructs for specifying learning problems (1) make aspects of autonomous robot learning explicit and controllable; (2) have an enormous impact on the performance of the learned controllers and therefore encourage the engineering of high performance learners; (3) make the learning processes repeatable and allow for writing bootstrapping robot controllers. Taken together the approach constitutes an important step towards engineering controllers of autonomous learning robots.
Reference:
Making Robot Learning Controllable: A Case Study in Robot Navigation (A Kirsch, M Schweitzer and M Beetz), In Proceedings of the ICAPS Workshop on Plan Execution: A Reality Check, 2005.
Bibtex Entry:
@inproceedings{kirsch05making,
author = {A Kirsch and M Schweitzer and M Beetz},
title = {Making Robot Learning Controllable: A Case Study in Robot Navigation},
booktitle = {Proceedings of the ICAPS Workshop on Plan Execution: A Reality Check},
year = {2005},
bib2html_pubtype = {Workshop Paper},
bib2html_rescat = {Learning,Planning,Action},
bib2html_groups = {Cogito, AGILO},
abstract = {In many applications the performance of learned robot controllers
drags behind those of the respective hand-coded ones. In our view,
this situation is caused not mainly by deficiencies of the learning
algorithms but rather by an insufficient embedding of learning in
robot control programs.
This paper presents a case study in which RoLL, a robot control
language that allows for explicit representations of learning
study shows that RoLL's constructs for specifying learning
problems (1) make aspects of autonomous robot learning explicit
and controllable; (2) have an enormous impact on the
performance of the learned controllers and therefore encourage the
engineering of high performance learners; (3) make the learning
processes repeatable and allow for writing bootstrapping robot
controllers.  Taken together the approach constitutes an important
step towards engineering controllers of autonomous learning
robots.},
}