search

iit-DLSLab / Quadruped-PyMPC

A model predictive controller for quadruped robots based on the single rigid body model and written in python. Gradient-based (acados) or Sampling-based (jax).
216 stars 18 forks source link

deal with rough terrain #14

Closed zhxiaoyu218 closed 4 months ago

zhxiaoyu218 commented 4 months ago

I interested in if this work could deal with rough terrain, e.g. wave or uneven

giulioturrisi commented 4 months ago

Yes it can. Just change the string here https://github.com/iit-DLSLab/Quadruped-PyMPC/blob/4d6437c67e66eecdb29bfac70732edeee412b363/quadruped_pympc/config.py#L198 to modify the terrain. For the stairs, the swing gain in the same config file maybe are too big, and should be reduced