Visual Dexterity: In-Hand Reorientation of
Novel and Complex Object Shapes

Tao Chen     Megha Tippur     Siyang Wu     Vikash Kumar     Edward Adelson     Pulkit Agrawal
Accepted by Science Robotics
Improbable AI Lab, CSAIL
Massachusetts Institute of Technology


In-hand object reorientation is necessary for performing many dexterous manipulation tasks, such as tool use in less structured environments that remain beyond the reach of current robots. Prior works built reorientation systems assuming one or many of the following: reorienting only specific objects with simple shapes, limited range of reorientation, slow or quasistatic manipulation, simulation-only results, the need for specialized and costly sensor suites, and other constraints which make the system infeasible for real-world deployment. We present a general object reorientation controller that does not make these assumptions. It uses readings from a single commodity depth camera to dynamically reorient complex and new object shapes by any rotation in real-time, with the median reorientation time being close to seven seconds. The controller is trained using reinforcement learning in simulation and evaluated in the real world on new object shapes not used for training, including the most challenging scenario of reorienting objects held in the air by a downward-facing hand that must counteract gravity during reorientation. Our hardware platform only uses open-source components that cost less than five thousand dollars. Although we demonstrate the ability to overcome assumptions in prior work, there is ample scope for improving absolute performance. For instance, the challenging duck-shaped object not used for training was dropped in 56 percent of the trials. When it was not dropped, our controller reoriented the object within 0.4 radians (23 degrees) 75 percent of the time.


Visual Dexterity: In-hand Dexterous Manipulation from Depth
Tao Chen, Megha Tippur, Siyang Wu, Vikash Kumar, Edward Adelson, Pulkit Agrawal
Science Robotics / arXiv / project page / code / bibtex


Continous Reorientation in the Air

Once a goal orientation is reached, and the hand stops, we give it a new goal orientation.


Video Clips

A System for General In-Hand Object Re-Orientation
Tao Chen, Jie Xu, Pulkit Agrawal
Conference on Robot Learning (CoRL), 2021 (Best Paper Award)
paper / arXiv / code / project page / bibtex

Vegetable Peeling: A Case Study in Constrained Dexterous Manipulation
Tao Chen, Eric Cousineau, Naveen Kuppuswamy, Pulkit Agrawal
project page