Loading video...

Video Failed to Load

Go Home

How to chain multiple dexterous skills to tackle complex long-horizon manipulation tasks? Imagine retrieving a LEGO block from a pile, rotating it in-hand, and inserting it at the desired location to build a structure. Introducing our new work - Sequential Dexterity 🧵👇

159,809 views • 2 years ago •via X (Twitter)

11 Comments

Chen Wang's profile picture
Chen Wang2 years ago

The core of the system is a learning-based transition feasibility function that progressively finetunes the sub-policies (learned with RL) for enhancing chaining success, which can also be used during skill selection for re-planning from failures and bypassing redundant stages.

Chen Wang's profile picture
Chen Wang2 years ago

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.

Chen Wang's profile picture
Chen Wang2 years ago

We hope Sequential Dexterity paves the path for future research on long-horizon dexterous manipulation. Feel free to check out our code! Website & Paper: Code: Work done w/ Yuanpei Chen, @drfeifei, and Karen Liu at @StanfordAILab.

Karol Hausman's profile picture
Karol Hausman2 years ago

Congrats, great work!

Chen Wang's profile picture
Chen Wang2 years ago

Thanks Karol!

Deepak Pathak's profile picture
Deepak Pathak2 years ago

very cool task and results, congrats!!

Chen Wang's profile picture
Chen Wang2 years ago

Thanks Deepak! Very interested in LEAP hand and functional grasp

Lucy Shi's profile picture
Lucy Shi2 years ago

super cool work, @chenwang_j!

Chen Wang's profile picture
Chen Wang2 years ago

Thanks Lucy!

Kevin Zakka's profile picture
Kevin Zakka2 years ago

Very inspiring, congrats!

Chen Wang's profile picture
Chen Wang2 years ago

Thanks Kevin! huge fan of RoboPianist!

Related Videos