huangwl18/geometry-dex

PyTorch Code for "Generalization in Dexterous Manipulation via Geometry-Aware Multi-Task Learning"

39
/ 100
Emerging

This project helps robotics researchers develop and test policies for dexterous robot hands to manipulate a wide range of objects. It takes in 3D object geometry (point clouds) and outputs a robot control policy that can perform in-hand manipulation on over 100 diverse real-world objects, even generalizing to new objects with different shapes or sizes. Robotics researchers and engineers working on advanced manipulation tasks would find this useful.

No commits in the last 6 months.

Use this if you are a robotics researcher working on developing generalizable policies for robot dexterous manipulation across a large and varied set of objects.

Not ideal if you are looking for a pre-trained, production-ready solution for a specific industrial manipulation task without requiring policy development or research.

robotics dexterous-manipulation robot-learning reinforcement-learning robot-control
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 16 / 25
Community 15 / 25

How are scores calculated?

Stars

64

Forks

10

Language

Python

License

MIT

Last pushed

Feb 15, 2022

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/huangwl18/geometry-dex"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.