RSL-RL
An open-source PPO implementation from the Robotic Systems Lab at ETH Zurich, optimized for large-scale parallel simulation.
Timeline (last 90 days)
A game animator used MyoSuite's MyoLeg model (80 leg muscles) in MuJoCo/MuJoCo Warp to train an NPC physical walking policy with the PPO algorithm (calling RSL-RL through mjlab) on a laptop GPU.
RSL-RL is a PPO implementation open-sourced by ETH Zurich's Robotic Systems Lab, optimized for large-scale parallel simulation.
mjlab is a training framework that connects MuJoCo (and its GPU-accelerated version MuJoCo Warp) with RSL-RL, enabling developers to run thousands of environment instances in parallel on a single GPU.
RSL-RL is a lightweight reinforcement learning framework developed by ETH Zurich's Robotic Systems Lab specifically for legged robots, and is deeply integrated with Isaac Gym.
All Facts (4)
A game animator used MyoSuite's MyoLeg model (80 leg muscles) in MuJoCo/MuJoCo Warp to train an NPC physical walking policy with the PPO algorithm (calling RSL-RL through mjlab) on a laptop GPU.
50%UnverifiedRSL-RL is a PPO implementation open-sourced by ETH Zurich's Robotic Systems Lab, optimized for large-scale parallel simulation.
50%Unverifiedmjlab is a training framework that connects MuJoCo (and its GPU-accelerated version MuJoCo Warp) with RSL-RL, enabling developers to run thousands of environment instances in parallel on a single GPU.
50%UnverifiedRSL-RL is a lightweight reinforcement learning framework developed by ETH Zurich's Robotic Systems Lab specifically for legged robots, and is deeply integrated with Isaac Gym.
50%