RL + Model-based Control: Using On-demand Optimal Control to Learn Versatile Legged Locomotion

no code implementations • 29 May 2023 • Dongho Kang, Jin Cheng, Miguel Zamora, Fatemeh Zargarbashi, Stelian Coros

These reference motions serve as targets for the RL policy to imitate, leading to the development of robust control policies that can be learned with reliability.

Reinforcement Learning (RL)

Gradient-Based Trajectory Optimization With Learned Dynamics

no code implementations • 9 Apr 2022 • Bhavya Sukhija, Nathanael Köhler, Miguel Zamora, Simon Zimmermann, Sebastian Curi, Andreas Krause, Stelian Coros

In our hardware experiments, we demonstrate that our learned model can represent complex dynamics for both the Spot and Radio-controlled (RC) car, and gives good performance in combination with trajectory optimization methods.

Learning Solution Manifolds for Control Problems via Energy Minimization

no code implementations • 7 Mar 2022 • Miguel Zamora, Roi Poranne, Stelian Coros

We formulate the learning of solution manifolds as a minimization of the energy terms of a control objective integrated over the space of problems of interest.

Imitation Learning Model Predictive Control