Where To Start? Transferring Simple Skills to Complex Environments

Vitalis Vosylius, Edward Johns
Proceedings of The 6th Conference on Robot Learning, PMLR 205:471-481, 2023.

Abstract

Robot learning provides a number of ways to teach robots simple skills, such as grasping. However, these skills are usually trained in open, clutter-free environments, and therefore would likely cause undesirable collisions in more complex, cluttered environments. In this work, we introduce an affordance model based on a graph representation of an environment, which is optimised during deployment to find suitable robot configurations to start a skill from, such that the skill can be executed without any collisions. We demonstrate that our method can generalise a priori acquired skills to previously unseen cluttered and constrained environments, in simulation and in the real world, for both a grasping and a placing task.

Cite this Paper


BibTeX
@InProceedings{pmlr-v205-vosylius23a, title = {Where To Start? Transferring Simple Skills to Complex Environments}, author = {Vosylius, Vitalis and Johns, Edward}, booktitle = {Proceedings of The 6th Conference on Robot Learning}, pages = {471--481}, year = {2023}, editor = {Liu, Karen and Kulic, Dana and Ichnowski, Jeff}, volume = {205}, series = {Proceedings of Machine Learning Research}, month = {14--18 Dec}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v205/vosylius23a/vosylius23a.pdf}, url = {https://proceedings.mlr.press/v205/vosylius23a.html}, abstract = {Robot learning provides a number of ways to teach robots simple skills, such as grasping. However, these skills are usually trained in open, clutter-free environments, and therefore would likely cause undesirable collisions in more complex, cluttered environments. In this work, we introduce an affordance model based on a graph representation of an environment, which is optimised during deployment to find suitable robot configurations to start a skill from, such that the skill can be executed without any collisions. We demonstrate that our method can generalise a priori acquired skills to previously unseen cluttered and constrained environments, in simulation and in the real world, for both a grasping and a placing task. } }
Endnote
%0 Conference Paper %T Where To Start? Transferring Simple Skills to Complex Environments %A Vitalis Vosylius %A Edward Johns %B Proceedings of The 6th Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2023 %E Karen Liu %E Dana Kulic %E Jeff Ichnowski %F pmlr-v205-vosylius23a %I PMLR %P 471--481 %U https://proceedings.mlr.press/v205/vosylius23a.html %V 205 %X Robot learning provides a number of ways to teach robots simple skills, such as grasping. However, these skills are usually trained in open, clutter-free environments, and therefore would likely cause undesirable collisions in more complex, cluttered environments. In this work, we introduce an affordance model based on a graph representation of an environment, which is optimised during deployment to find suitable robot configurations to start a skill from, such that the skill can be executed without any collisions. We demonstrate that our method can generalise a priori acquired skills to previously unseen cluttered and constrained environments, in simulation and in the real world, for both a grasping and a placing task.
APA
Vosylius, V. & Johns, E.. (2023). Where To Start? Transferring Simple Skills to Complex Environments. Proceedings of The 6th Conference on Robot Learning, in Proceedings of Machine Learning Research 205:471-481 Available from https://proceedings.mlr.press/v205/vosylius23a.html.

Related Material