We release Evo-RL, the opensource real-world offline RL on So-101 and AgileX PiPER for easier reproduction.
Evo-RL is an open-source framework for real-world reinforcement learning on SO101 robots, providing pipelines for data collection, value function training, advantage computation, and iterative policy improvement.
How It Works
You find Evo-RL, an easy way to teach real robots new skills through trial and improvement.
Hook up your SO101 robot arm and leader controller so it follows your hand movements.
Guide the robot through tasks while it records your smooth motions to learn from.
The system learns to score how good each moment in your recordings is.
Value scores get added to recordings, highlighting the best parts for training.
Your robot learns policies that focus on high-value actions from the improved data.
Deploy the policy, collect better data, and repeat to make your robot a pro.
Star Growth
Repurpose is a Pro feature
Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.
Unlock RepurposeSimilar repos coming soon.