/mppi-tf

Tensorflow implementation of Information theoretic MPC for Model-Based Reinforcement Learning

Primary LanguagePython

Torch implementation of MPPI

Dependencies:

similar as for the rnn repo

Running:

To run the model for one step, call the main file:

python run_controller.py

The controller works as follows:

  • Instanciate a model, a cost, a controller and a observer.
  • Call the controller on a fake state.