LIPM Walking Controller

Source code of the walking and stair climbing controller used in the experiments of Stair Climbing Stabilization of the HRP-4 Humanoid Robot using Whole-body Admittance Control, as well as in an industrial demonstrator at the Airbus Saint-Nazaire factory.

Installation

Compilation has been tested on Ubuntu 14.04 (gcc/clang) with ROS Indigo and Ubuntu 16.04 (gcc) with ROS Kinetic.

Dependencies

Compilation requires:

• ROS with a working catkin workspace
• SpaceVecAlg: spatial vector algebra
• RBDyn: rigid body dynamics
• eigen-lssol: quadratic programming (if you have the LSSOL licence ask us this library)
• eigen-qld: quadratic programming
• sch-core: collision detection
• Tasks: inverse kinematics
• mc_rbdyn_urdf: robot model loader
• copra: linear model predictive control

The following dependencies are not publicly released but available upon request to Pierre Gergondet:

• mc_rtc: robot controller library (includes mc_control, mc_rbdyn, mc_solver and mc_tasks)
• mc_rtc_ros: ROS tools for mc_rtc
• mc_rtc_ros_data: ROS environment and object descriptions for mc_rtc

Building from source on Linux

Link lipm_walking_controller from the source folder of your catkin workspace, then follow the standard catkin workflow:

catkin_make -DCMAKE_BUILD_TYPE=RelWithDebInfo && catkin_make install

To avoid a sudo at catkin_make install you can change ownership or permissions of the /usr/local/lib/mc_controller folder.

Usage

Launch RViz for the HRP-4 model by:

roslaunch lipm_walking_controller display.launch robot:=hrp4

Enable the controller in your mc_rtc configuration:

{
  "MainRobot": "HRP4",
  "Enabled": ["LIPMWalking"]
}

Finally, start the controller from your mc_rtc interface. Here is the example of a local Choreonoid simulation using the mc_udp interface:

cd /usr/local/share/hrpsys/samples/HRP4LIRMM
choreonoid --start-simulation HRP4LIRMM_udp.cnoid  # in one terminal
MCUDPControl -h localhost  # in another terminal

You should end up with the following windows: See the Graphical user interface page of the wiki for further instructions on how to use this GUI.

Thanks

To @gergondet for developing and helping with the mc_rtc framework.