rl_sar
- Software
- Humanoids and robot learning
- Framework
A framework for simulating and deploying reinforcement learning control policies on quadruped, wheeled, and humanoid robots, bridging simulation (Gazebo, MuJoCo) and real hardware.
Source on GitHub: github.com/fan-ziqi/rl_sar (opens in a new tab). Facts read at commit 376d42c (opens in a new tab).
Licenses
- Code
- Apache-2.0 (opens in a new tab)
Read by Robotheca from LICENSE (opens in a new tab) (Apache-2.0) at commit 376d42c (opens in a new tab), on 16 Oct 2026.
- Commercial use
- Yes
What it does
- Runs trained RL locomotion policies in Gazebo or MuJoCo simulators before deploying them on physical robots.
- Supports ROS-Noetic and ROS2 Foxy/Humble, and policies run through libtorch or onnxruntime.
- Provides gamepad and keyboard control, plus an experimental mobile web interface for operating robots.
- Includes drivers for real robots such as Unitree A1, Go2/Go2W/G1, Deeprobotics Lite3, and Agibot D1.
- Includes a tool for training an actuator network from logged motor data.
- Offers a Docker setup for simulation and deployment.
Written by Robotheca from the project's README (opens in a new tab) at commit 376d42c.
Getting started
- Difficulty
- Advanced
- How to start
- The project's README (opens in a new tab)
Maintenance
- Status
- Active: a commit in the last 90 days
- Newest commit
- 16 Aug 2026 on main
- Latest release
- v4.1.0 (19 Mar 2026)
Checked on GitHub on 16 Oct 2026.
Links
Robots that can run it
No robot on Robotheca is linked to it yet. Links come from a robot's robot.yaml, from its dependency files read as text, or from Robotheca's editors.
Robotheca lists software as a reference: nothing is hosted or run here, and every fact links to where it was read.