• Classified by Topic • Classified by Publication Type • Sorted by Date • Sorted by First Author Last Name • Classified by Funding Source •
Robot Air Hockey: A Manipulation Testbed for Robot Learning with Reinforcement Learning.
Caleb Chuck, Carl Qi, Michael
J. Munje, Shuozhe Li, Max Rudolph, Chang Shi, Siddhant Agarwal, Harshit Sikchi, Abhinav Peri, Sarthak Dayal, Evan Kuo, Kavan
Mehta, Anthony Wang, Peter Stone, Amy Zhang, and Scott
Niekum.
In ICRA Workshop on Manipulation Skills, May 2024.
Reinforcement Learning is a promising tool for learning complex policies even infast-moving and object-interactive domains where human teleoperation orhard-coded policies might fail. To effectively reflect this challenging categoryof tasks, we introduce a dynamic, interactive RL testbed based on robot airhockey. By augmenting air hockey with a large family of tasks ranging from easytasks like reaching, to challenging ones like pushing a block by hitting it witha puck, as well as goal-based and human-interactive tasks, our testbed allows avaried assessment of RL capabilities. The robot air hockey testbed also supportssim-to-real transfer with three domains: two simulators of increasing fidelityand a real robot system. Using a dataset of demonstration data gathered throughtwo teleoperation systems: a virtualized control environment, and humanshadowing, we assess the testbed with behavior cloning, offline RL, and RL fromscratch.
@InProceedings{caleb_chuck_AirHockeyICRA_2024,
author = {Caleb Chuck and Carl Qi and Michael J. Munje and Shuozhe Li and Max Rudolph and Chang Shi and Siddhant Agarwal and Harshit Sikchi and Abhinav Peri and Sarthak Dayal and Evan Kuo and Kavan Mehta and Anthony Wang and Peter Stone and Amy Zhang and Scott Niekum},
title = {Robot Air Hockey: A Manipulation Testbed for Robot Learning with Reinforcement Learning},
booktitle = {ICRA Workshop on Manipulation Skills},
year = {2024},
month = {May},
location = {Yokohama, Japan},
abstract = {Reinforcement Learning is a promising tool for learning complex policies even in
fast-moving and object-interactive domains where human teleoperation or
hard-coded policies might fail. To effectively reflect this challenging category
of tasks, we introduce a dynamic, interactive RL testbed based on robot air
hockey. By augmenting air hockey with a large family of tasks ranging from easy
tasks like reaching, to challenging ones like pushing a block by hitting it with
a puck, as well as goal-based and human-interactive tasks, our testbed allows a
varied assessment of RL capabilities. The robot air hockey testbed also supports
sim-to-real transfer with three domains: two simulators of increasing fidelity
and a real robot system. Using a dataset of demonstration data gathered through
two teleoperation systems: a virtualized control environment, and human
shadowing, we assess the testbed with behavior cloning, offline RL, and RL from
scratch.
},
}
Generated by bib2html.pl (written by Patrick Riley ) on Tue Nov 19, 2024 10:24:46