Rofunc

🎉 This is the second official release version of `rofunc` package. We provide a reinforcement learning baseline framework (`RofuncRL`) that performs well in robot tasks 🦾, and several state-of-the-art online reinforcement learning algorithms (PPO, SAC, and TD3) have been completed.🥳

👉**Next: We plan to combine planning algorithms for motion generation in the next major release (v0.0.3) and use it to train offline reinforcement learning as well as robot Large Models.**

**After more than 400 commits, we successfully ushered in the first official release version of `rofunc`.**

- [x] **Devices**: support Xsens, optitrack, Zed 2i (multi), Delsys EMG, Manus gloves and multimodal
- [x] **Learning algorithms**: TP-GMM, TP-GMR, RL baselines (RLlib, ElegantRL, SKRL with IsaacGym)
- [x] **Planning and control**: LQT variants, iLQR variants
- [x] **Simulator (IsaacGym-based)**: CURI, Ubtech walker, Franka, Baxter, Sawyer, CURI-mini

New Features:

1. Pipeline
2. readthedoc
3. TP-GMM
4. TP-GMR
5. LQT
6. Franka simulator
7. CURI simulator (bimanual humanoid robot)

**Full Changelog**: https://github.com/Skylark0924/Rofunc/compare/v0.0.0.7.1...v0.0.0.9

New Features:
1. ZED export
2. Parallel computing