force_feedback_mpc

Optimal control toolbox to achieve force feedback control in MPC. This library is basically an extension of the Crocoddyl optimal control library: it implements custom action models in C++ with unittests and python bindings. In particular, it contains the core classes used in MPC experiments of the following papers

• S. Kleff, et. al, "Introducing Force Feedback in Model-Predictive Control", IROSS 2022. PDF
• S. Kleff, et. al, "Force Feedback in Model-Predictive Control: A Soft Contact Approach" PDF (under review)

The code to reproduce our experiments (i.e. real-time implementation of the force-feedback MPC), along with our experimental data are available in this separate repository.

Dependencies

• Crocoddyl (>=3.0)
• Pinocchio
• boost
• eigenpy (>=2.7.10)
• [Optional] OpenMP if Crocoddyl was built from source with the multi-threading option.

For the Python demo scripts

• croco_mpc_utils
• mim_robots
• PyBullet
• PyYAML
• importlib_resources
• matplotlib

Installation

git clone --recursive https://github.com/machines-in-motion/force_feedback_mpc.git
mkdir build && cd build
cmake .. 
make -j6 && sudo make install

How to use it

Simply prototype your OCP using Crocoddyl as you would normally do, but use the custom integrated action models provided in this library (IntegratedActionModelLPF and IAMSoftContactAugmented). Example python scripts can be found in the demos directory (simulated force and polishing tasks).

In the python directory, OCP utilities are implemented. These are simplified interfaces to Crocoddyl's API ; they allow the quick proto-typing of OCPs from templated YAML files (i.e. this is an extension of croco_mpc_utils to force-feedback OCPs).

The real-time controllers used on the real robot for the paper experiments are implemented in a separate repository, force_feedback_dgh. In this separate repo, you will also find the experimental data used to generate the paper figures.

Citing this work

@unpublished{kleff:hal-04572399,
  TITLE = {{Force Feedback in Model-Predictive Control: A Soft Contact Approach}},
  AUTHOR = {Kleff, S{\'e}bastien and Jordana, Armand and Khorrambakht, Rooholla and Mansard, Nicolas and Righetti, Ludovic},
  URL = {https://hal.science/hal-04572399},
  NOTE = {working paper or preprint},
  HAL_LOCAL_REFERENCE = {Rapport LAAS n{\textdegree} 24093},
  YEAR = {2025},
  MONTH = Jun,
  PDF = {https://hal.science/hal-04572399v2/file/force_feedback_article_second_submission.pdf},
  HAL_ID = {hal-04572399},
  HAL_VERSION = {v2},
}