search

tud-airlab / mppi-isaac

Model Predictive Path Integral Control using isaacgym for rollouts, gpu-accelerated
178 stars 21 forks source link

Moved all configs outside of python files to address #1 #3

Closed c-salmi closed 1 year ago

c-salmi commented 1 year ago

Moved all configs outside of python files, uses the hydra library to manage configs.

The test in the examples directory should still pass (at least it does for me locally).

This PR uses the hydra library which I think is very useful in managing different configurations for the mppi.

Some examples:

cd examples

# runs everything with defaults as set inside the respective dataclasses 
# and the default config file in conf/config.yaml
python point_robot.py

# runs with different dt and horizon by overriding from the command line
python point_robot.py isaacsim.dt=0.01 mppi.horizon=40

# runs with isaacsim configurations as in conf/isaacsim/fast.yaml
python point_robot.py isaacsim=fast