sizhe-li/neural-jacobian-field
Controlling diverse robots by inferring jacobian fields with deep networks! Let's make robots understand their bodies!
This project helps roboticists understand and control complex robot movements by learning how different parts of a robot's body interact. You input multiview videos of a robot performing various actions, and the system outputs a "Jacobian Field," which is a representation of the robot's physical capabilities. Roboticists and control engineers can use this to develop more agile and adaptive robot control systems.
203 stars.
Use this if you need a flexible way to model the mechanics of diverse robots directly from visual data, without needing detailed CAD models or physical parameters.
Not ideal if you require real-time, ultra-low-latency control on very simple robots, or if you already have precise analytical models and prefer traditional control methods.
Stars
203
Forks
31
Language
Jupyter Notebook
License
—
Category
Last pushed
Dec 09, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/sizhe-li/neural-jacobian-field"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
enactic/openarm
A fully open-source humanoid arm for physical AI research and deployment in contact-rich environments.
Shuijing725/awesome-robot-social-navigation
A curated list of robot social navigation.
thomashiemstra/fred
This my 3d printed robot arm project
jstmn/ikflow
Open source implementation to the paper "IKFlow: Generating Diverse Inverse Kinematics Solutions"
TimSchneider42/tactile-mnist
The Tactile-MNIST Active Perception Benchmark