Sequential Dexterity: Chaining Dexterous Policies for Long-Horizon Manipulation

Yuanpei Chen, Chen Wang, Li Fei-Fei, Karen Liu
Proceedings of The 7th Conference on Robot Learning, PMLR 229:3809-3829, 2023.

Abstract

Many real-world manipulation tasks consist of a series of subtasks that are significantly different from one another. Such long-horizon, complex tasks highlight the potential of dexterous hands, which possess adaptability and versatility, capable of seamlessly transitioning between different modes of functionality without the need for re-grasping or external tools. However, the challenges arise due to the high-dimensional action space of dexterous hand and complex compositional dynamics of the long-horizon tasks. We present Sequential Dexterity, a general system based on reinforcement learning (RL) that chains multiple dexterous policies for achieving long-horizon task goals. The core of the system is a transition feasibility function that progressively finetunes the sub-policies for enhancing chaining success rate, while also enables autonomous policy-switching for recovery from failures and bypassing redundant stages. Despite being trained only in simulation with a few task objects, our system demonstrates generalization capability to novel object shapes and is able to zero-shot transfer to a real-world robot equipped with a dexterous hand. Code and videos are available at https://sequential-dexterity.github.io.

Cite this Paper

BibTeX
@InProceedings{pmlr-v229-chen23e, title = {Sequential Dexterity: Chaining Dexterous Policies for Long-Horizon Manipulation}, author = {Chen, Yuanpei and Wang, Chen and Fei-Fei, Li and Liu, Karen}, booktitle = {Proceedings of The 7th Conference on Robot Learning}, pages = {3809--3829}, year = {2023}, editor = {Tan, Jie and Toussaint, Marc and Darvish, Kourosh}, volume = {229}, series = {Proceedings of Machine Learning Research}, month = {06--09 Nov}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v229/chen23e/chen23e.pdf}, url = {https://proceedings.mlr.press/v229/chen23e.html}, abstract = {Many real-world manipulation tasks consist of a series of subtasks that are significantly different from one another. Such long-horizon, complex tasks highlight the potential of dexterous hands, which possess adaptability and versatility, capable of seamlessly transitioning between different modes of functionality without the need for re-grasping or external tools. However, the challenges arise due to the high-dimensional action space of dexterous hand and complex compositional dynamics of the long-horizon tasks. We present Sequential Dexterity, a general system based on reinforcement learning (RL) that chains multiple dexterous policies for achieving long-horizon task goals. The core of the system is a transition feasibility function that progressively finetunes the sub-policies for enhancing chaining success rate, while also enables autonomous policy-switching for recovery from failures and bypassing redundant stages. Despite being trained only in simulation with a few task objects, our system demonstrates generalization capability to novel object shapes and is able to zero-shot transfer to a real-world robot equipped with a dexterous hand. Code and videos are available at https://sequential-dexterity.github.io.} }
Endnote
%0 Conference Paper %T Sequential Dexterity: Chaining Dexterous Policies for Long-Horizon Manipulation %A Yuanpei Chen %A Chen Wang %A Li Fei-Fei %A Karen Liu %B Proceedings of The 7th Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2023 %E Jie Tan %E Marc Toussaint %E Kourosh Darvish %F pmlr-v229-chen23e %I PMLR %P 3809--3829 %U https://proceedings.mlr.press/v229/chen23e.html %V 229 %X Many real-world manipulation tasks consist of a series of subtasks that are significantly different from one another. Such long-horizon, complex tasks highlight the potential of dexterous hands, which possess adaptability and versatility, capable of seamlessly transitioning between different modes of functionality without the need for re-grasping or external tools. However, the challenges arise due to the high-dimensional action space of dexterous hand and complex compositional dynamics of the long-horizon tasks. We present Sequential Dexterity, a general system based on reinforcement learning (RL) that chains multiple dexterous policies for achieving long-horizon task goals. The core of the system is a transition feasibility function that progressively finetunes the sub-policies for enhancing chaining success rate, while also enables autonomous policy-switching for recovery from failures and bypassing redundant stages. Despite being trained only in simulation with a few task objects, our system demonstrates generalization capability to novel object shapes and is able to zero-shot transfer to a real-world robot equipped with a dexterous hand. Code and videos are available at https://sequential-dexterity.github.io.
APA
Chen, Y., Wang, C., Fei-Fei, L. & Liu, K.. (2023). Sequential Dexterity: Chaining Dexterous Policies for Long-Horizon Manipulation. Proceedings of The 7th Conference on Robot Learning, in Proceedings of Machine Learning Research 229:3809-3829 Available from https://proceedings.mlr.press/v229/chen23e.html.

Related Material