Gepetto / aig

Stand alone computation of inverse geometry for biped robots or robotic arms.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

build token unit test token

AIG

Pipeline status Coverage report Code style: black pre-commit.ci status

Stand alone computation of inverse geometry for biped robots or robotic arms.

Here inverse geometry refer to function that compute the robot joint configuration from SE3 poses of it's end-effectors. Much similar to the classical inverse kinematics though at the position level and not at the velocity level.

This code is decomposed in 3 parts:

  • Leg inverse geometry.
  • Arm inverse geometry.
  • Biped inverse geometry. This part uses the 2 previous ones for computing the whole body inverse geometry assuming a constant level arm between the center of mass of the robot and its base.

1/ :penguin: Installation

1.1/ Standard dependencies

This code relies essentially on the C++ std and on the Rigid Body Dynamics implemented in Pinocchio.

For the unit-test it relies on the example-robot-data package as well as the boost unit test framework.

In order to install those 📦 from Debian / Ubuntu packages, you can use with robotpkg

  1. If you have never added robotpkg's software repository, do it now:

    sudo tee /etc/apt/sources.list.d/robotpkg.list <<EOF
    deb [arch=amd64] http://robotpkg.openrobots.org/packages/debian/pub $(lsb_release -sc) robotpkg
    EOF
    
    curl http://robotpkg.openrobots.org/packages/debian/robotpkg.key | sudo apt-key add -
    sudo apt update
  2. installation of pinocchio and example-robot-data and their python utils:

    sudo apt install robotpkg-py3\*-example-robot-data robotpkg-py3\*-pinocchio

1.2/ Build and install this package:

  1. 🐢 With ROS

    Just clone it (with --recursive) into a catkin or a colcon workspace.

  2. 📁 From source

    Clone it (with --recursive), create a build directory inside, and:

    cmake .. && make && make install

2/ Usage

2.1/ UnitTests/Examples

In this packages the unit-testsare based on the Talos robot from PAL Robotics. Please read the unit-test to see how to use the code.

2.2/ API documentation

How to build the API html doc. Where to find the last built doc on the internet.

3/ License and Copyrights

License BSD-2-Clause Copyright (c) 2022, CNRS, Gepetto

About

Stand alone computation of inverse geometry for biped robots or robotic arms.

License:BSD 2-Clause "Simplified" License


Languages

Language:C++ 75.6%Language:Python 21.3%Language:CMake 3.1%