Documentation Guide
GitHub Repositories and Documentation Entry
The table below summarizes the wuji-mjlab repository, use case, and recommended reading entry.
| Repository | Version | Use Case | Applies To |
|---|---|---|---|
| wuji-mjlab |  | Reinforcement learning training and real-robot deployment for the hand | Wuji Hand |
wuji-mjlab currently targets the Wuji Hand (first generation), and the deploy bridge wraps the Wuji Hand SDK. Read in order: introduction, installation, training, hardware setup, deployment, and architecture.
Available Documentation
Changelog
Version updates and changelog for Wuji MJLab
Introduction
Project overview, task, demo, and released assets
Installation
pixi environment setup, verification, and dev configuration
Training and Evaluation
Training configs, replay, evaluation, and ONNX export
Hardware Setup
Camera, MVS SDK, cube, wrist tag, hand jig, and calibration
Sim-to-real Deployment
Software pipeline, pose tuning, closed-loop run, and smoke test
Architecture
Three-layer stack, reorient task and MDP design, domain randomization
Technical Support
For technical support or any questions, please contact us: support@wuji.tech