⛇☃︎

experiments & fun

• CPG locomotion lab
  simple, interactive CPG (central pattern generator) experiments to understand how to build movement from oscillations
• MPC explorer
  interactive MPC (model predictive control) sandbox for experimenting with control horizons, constraints, and simple dynamics
• PPO explorer
  reinforcement learning basics: watch a square learn to reach a reward circle with PPO (clipped policy gradients + gradient descent)
• VMC explorer
  2D VMC (virtual model control) sandbox: task-space springs/dampers mapped to joint torques (arm + quadruped) and thrust/moment (drone)