seekdepth/ppo-LunarLander-v2

PPO Agent playing LunarLander-v2

This is a trained model of a PPO agent playing LunarLander-v2 using the stable-baselines3 library.

Interactive browser demo

Try this policy in the Lunar Lander Sandbox. Choose Human Pilot for direct keyboard control, AI Pilot to let the policy choose every action, or Autopilot with Manual Override to let the policy fly until you intervene. 4 AI Simulations runs four independent policy-controlled flights together, with separate rewards and outcomes.

The browser runs an ONNX export of this checkpoint with ONNX Runtime Web. Its Sensor Fault setting (OFF by default) can substitute the previous physics-step observation for an AI decision's current input. Thruster Fault (also OFF by default) can drop a requested thruster action or fire an unrequested thruster when the controller chose coast. Thruster faults affect human and AI commands; sensor faults affect policy inputs. These are browser-side experiments and do not change the trained checkpoint or the Stable-Baselines3 evaluation.

The browser environment follows the LunarLander-v2 observation and reward structure, but uses PyPyBox2D's older physics solver and a settled-contact landing completion rule. Its individual flight rewards and outcomes can differ from the self-reported evaluation below and from the course evaluation.

Project Links

• Live browser demo
• Source code and implementation notes
• Hugging Face Deep RL Course: Unit 1 hands-on

Run the browser demo locally

git clone https://github.com/jpeckenpaugh/rl-livive-net.git
cd rl-livive-net
bash install.sh
bash run.sh

Open http://localhost:8787. The build downloads the published ONNX policy from this model repository if it is not present in the checkout, so the first build needs an internet connection.