Quadruped Locomotion using Model Predictive Control and Reinforcement Learning

Quadruped Locomotion using Model Predictive Control and Reinforcement Learning 🤖
We implemented the convex Model-Predictive Control (MPC) approach described in the "Dynamic Locomotion in the MIT Cheetah 3" paper within the Quad-SDK framework.
For the RL a Proximal Policy Optimization algorithm was used in NVidia's Isaac Sim.
MPC code: github.com/iol...
RL code: github.com/dyu...
#robotics #MPC #quadruped #locomotion #CentraleNantes #ROS