krohling/nl-act

Integrating Natural Language Instructions into the Action Chunking Transformer for Multi-Task Robotic Manipulation

29
/ 100
Experimental

This project helps roboticists program industrial or research robots to perform complex manipulation tasks by simply providing natural language instructions. You input descriptions like "grasp the red cube" or "stack the blocks," and the system outputs the precise control commands needed for the robot to execute those actions. It's designed for robotics engineers or researchers who need to quickly teach robots new multi-step tasks without extensive coding for each scenario.

No commits in the last 6 months.

Use this if you need to enable a robot to understand and execute varied manipulation tasks based on diverse natural language commands, without retraining the robot for every new phrasing or slight task variation.

Not ideal if you are looking for a plug-and-play solution for consumer robotics or if you do not have access to a GPU-enabled computing environment for training.

robotic-manipulation robot-task-automation industrial-robotics robot-programming human-robot-interaction
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 16 / 25
Community 8 / 25

How are scores calculated?

Stars

9

Forks

1

Language

Python

License

MIT

Last pushed

Dec 23, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/krohling/nl-act"

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