Only released in EOL distros:
Package Summary
The vigir_step_control package
- Maintainer status: maintained
- Maintainer: Alexander Stumpf <stumpf AT sim.tu-darmstadt DOT de>
- Author: Alexander Stumpf <stumpf AT sim.tu-darmstadt DOT de>
- License: GPLv3
- Source: git https://github.com/team-vigir/vigir_step_control.git (branch: master)
Package Summary
The vigir_step_control package
- Maintainer status: maintained
- Maintainer: Alexander Stumpf <stumpf AT sim.tu-darmstadt DOT de>
- Author: Alexander Stumpf <stumpf AT sim.tu-darmstadt DOT de>
- License: GPLv3
- Source: git https://github.com/team-vigir/vigir_step_control.git (branch: master)
Package Summary
The vigir_step_control package
- Maintainer status: maintained
- Maintainer: Alexander Stumpf <stumpf AT sim.tu-darmstadt DOT de>
- Author: Alexander Stumpf <stumpf AT sim.tu-darmstadt DOT de>
- License: GPLv3
- Source: git https://github.com/team-vigir/vigir_step_control.git (branch: master)
Contents
Overview
The vigir_step_control provides a step queue management system and a state machine to keep track of current step plan execution. It is able to feed the low-level motion layer with the proper number of steps required for seamless execution while continuously merging incoming step plans provided by the footstep planner. This allows even for continuous walking using our footstep planning stack. This package provides following features:
- Step Queue Management:
- Seamless updates, merges, and stitches of multiple step plans
- Convenient step spooler
- Easy integration:
- Pre-implemented state machine simplifies deployment
- Comprehensive interface plugin as conveyor between footstep planning and motion execution layer
For detailed information about the system please take a look at the tutorial page.
Getting Started
Installation
The vigir_step_control packages can be downloaded from GitHub:
Furthermore you may need to download the dependencies as well (if not already done):
After a recompile of you workspace everything should be ready to use.
Example Startup
In order to run this example. you will need to install the vigir_footstep_planning stack as well. An install instruction can be found here.
After compiled your workspace you start up the system (each line in a seperate terminal):
roscore roslaunch vigir_footstep_planning footstep_planner_test.launch roslaunch vigir_step_control step_controller_test.launch roslaunch vigir_footstep_planning step_interface_rqt.launch
In the step_interface_widget you have first to select a footstep planner parameter set. Afterwards the direction commands become available which allows to generate simple pattern of footsteps. As soon a footstep plan has been generated the Execute button becomes active as well. When you click on this button the previously generated footsteps will be executed virtually by the StepControllerTestPlugin which fakes an footstep execution. During the fake execution, you will be able to follow the fake execution in the step_interface_widget by the progress bar.
Note: If you would like to see the system running with a real robot, then you can get in touch with it using the THORMANG3 in simulation. The full install instruction can be found here.
Tutorial
A Tutorial will be added step by step here soon. In the meantime feel free to contact us, if you have further questions!
Sample Applications
THORMANG3
The THORMANG3 provides a nice real world example how to migrate the software into an exisiting setup here.
During the Humanoids@RoboCup Demo 2016 at Leipzig, the step controller was used.
THORMANG3 Demo
If you would like to try it out in simulation by yourself, then just follow the instructions here.