Abstract
In this letter, we propose a novel framework for tactile-based dexterous manipulation learning with a blind anthropomorphic robotic hand, i.e. without visual sensing. First, object-related states were extracted from the raw tactile signals by a graph-based perception model - TacGNN. The resulting tactile features were then utilized in the policy learning of an in-hand manipulation task in the second stage. This method was examined by a Baoding ball task - simultaneously manipulating two spheres around each other by 180 degrees in hand. We conducted experiments on object states prediction and in-hand manipulation using a reinforcement learning algorithm (PPO). Results show that TacGNN is effective in predicting object-related states during manipulation by decreasing the RMSE of prediction to 0.096 cm comparing to other methods, such as MLP, CNN, and GCN. Finally, the robot hand could finish an in-hand manipulation task solely relying on the robotic own perception - tactile sensing and proprioception. In addition, our methods are tested on three tasks with different difficulty levels and transferred to the real robot without further training.
Links
BibTeX (Download)
@article{Yang2023TacGNN,
title = {TacGNN: Learning Tactile-based In-hand Manipulation with a Blind Robot using Hierarchical Graph Neural Network},
author = {Linhan Yang and Bidan Huang and Qingbiao Li and Ya-Yen Tsai and Wang Wei Lee and Chaoyang Song and Jia Pan},
doi = {10.1109/LRA.2023.3264759},
year = {2023},
date = {2023-04-05},
urldate = {2023-04-05},
journal = {IEEE Robotics and Automation Letters},
volume = {8},
number = {6},
issue = {June},
pages = {3605-3612},
abstract = {In this letter, we propose a novel framework for tactile-based dexterous manipulation learning with a blind anthropomorphic robotic hand, i.e. without visual sensing. First, object-related states were extracted from the raw tactile signals by a graph-based perception model - TacGNN. The resulting tactile features were then utilized in the policy learning of an in-hand manipulation task in the second stage. This method was examined by a Baoding ball task - simultaneously manipulating two spheres around each other by 180 degrees in hand. We conducted experiments on object states prediction and in-hand manipulation using a reinforcement learning algorithm (PPO). Results show that TacGNN is effective in predicting object-related states during manipulation by decreasing the RMSE of prediction to 0.096 cm comparing to other methods, such as MLP, CNN, and GCN. Finally, the robot hand could finish an in-hand manipulation task solely relying on the robotic own perception - tactile sensing and proprioception. In addition, our methods are tested on three tasks with different difficulty levels and transferred to the real robot without further training.},
keywords = {Co-Author, Dual-Track, IEEE Robot. Autom. Lett. (RA-L), JCR Q2},
pubstate = {published},
tppubtype = {article}
}
