mim_estimation

Introduction

This package contains estimation algorithm C++ implementations with their Python bindings and their Python prototypes.

Getting started

Installation

Dependencies

• Pinocchio (Required)
  • See this tutorial in order to install pinocchio or you can follow this tutorial to install related robotpkg software.
• Mim_Control (Optional, needed to run demos)
  • Internal github organization software see the next paragraph for the installation.
• Matplotlib (Optional, needed to run demos)
  • pip or apt installable software.
• BulletUtils (Optional, needed to run demos)
  • Internal github organization software see the next paragraph for the installation.
• Robot_Properties_Solo (Optional, needed to run demos)
  • Internal github organization software see the next paragraph for the installation.
• Robot_Properties_Bolt (Optional, needed to run demos)
  • Internal github organization software see the next paragraph for the installation.

Download the package.

External dependencies: See this tutorial in order to install pinocchio or you can follow this tutorial to install related robotpkg software. We use treep in order to manage the download of our code base. See treep_installation for the installation, and treep, and colcon for their usage.

Local and specific dependencies and the actual repo we need to compile:

mkdir devel
cd devel
pip install -U treep
git clone git@github.com:machines-in-motion/treep_machines_in_motion
treep --clone mim_estimation

For full installation of the package and related dependencies, use following:

treep --clone MIM_ESTIMATION

With this command all dependencies needed to run the demos are also downloaded, see this tutorial to install related dependencies for Mim_Control package.

Build the package

We use colcon to build this package:

cd devel/workspace
colcon build --cmake-args -DCMAKE_BUILD_TYPE=Release

Source the environment:

source ~/devel/workspace/install/setup.bash

Usage

Running Demos

To run the base EKF estimator on Solo12 with pre-defined contact schedule, follow the below steps:

source /opt/openrobots/setup.bash (source open robots)
cd demos
python3 demo_ekf_squatting_motion_cpp.py

License

BSD 3-Clause

Authors

Main authors:

• Shahram Khorshidi
• Ahmad Gazar

Secondary collaborators

• Maximilien Naveau (mnaveau@tue.mpg.de)
• Julian Viereck (jviereck@tue.mpg.de)
• Avadesh Meduri (ameduri@tue.mpg.de)
• Majid Kadhiv (mkadhiv@tue.mpg.de)
• Ludovic Righetti (lrighetti@tue.mpg.de)
• Nick Rotella
• Brahayam Ponton

Cite the work

Please if you use this repository cite the work: https://arxiv.org/abs/2202.12574

@misc{khorshidi2023use,
      title={On the Use of Torque Measurement in Centroidal State Estimation}, 
      author={Shahram Khorshidi and Ahmad Gazar and Nicholas Rotella and Maximilien Naveau and Ludovic Righetti and Maren Bennewitz and Majid Khadiv},
      year={2023},
      eprint={2202.12574},
      archivePrefix={arXiv},
      primaryClass={cs.RO}
}