An efficient acyclic contact planner for multiped robots: installation and technical details

These instructions explain how to install the specific version of the planner used for the TRO paper, as well as the optional tools that were used to validate the approach.

Motion planner installation on ubuntu-14.04 64 bit

Installation instructions

These instructions use permanent links to the libraries used to generate the examples demonstrated in the original T-RO paper. To get the latest version of hpp-rbprm, please refer to the github repository.
We strongly advise using the latest version of rbprm, as the present one is no longer supported, and has been dramatically improved!
To install all the packages on ubuntu 14.04 LTS 64 bit, you should do the following steps:

1. install ROS-indigo: follow steps 1.1 to 1.3 of the ROS installation website

2. install by apt-get:

autoconf
g++
cmake
libboost-dev
liburdfdom-dev
libassimp-dev
ros-indigo-xacro
ros-indigo-kdl-parser
ros-indigo-common-msgs
ros-indigo-tf
ros-indigo-tf-conversions
ros-indigo-libccd
ros-indigo-octomap
ros-indigo-resource-retriever
ros-indigo-urdfdom-py
ros-indigo-srdfdom
ros-indigo-pr2-robot
flex
bison
asciidoc
source-highlight
git
libomniorb4-dev
omniorb-nameserver
omniidl
omniidl-python
libltdl-dev
python-matplotlib
libtinyxml2-dev
liblog4cxx10-dev
libltdl-dev
qt4-dev-tools
libqt4-opengl-dev
libqtgui4
oxygen-icon-theme
libcdd-dev

sudo apt-get install autoconf g++ cmake libboost-dev liburdfdom-dev libassimp-dev ros-indigo-xacro ros-indigo-kdl-parser ros-indigo-common-msgs ros-indigo-tf ros-indigo-tf-conversions ros-indigo-libccd ros-indigo-octomap ros-indigo-resource-retriever ros-indigo-srdfdom ros-indigo-pr2-robot flex bison asciidoc source-highlight git libomniorb4-dev omniorb-nameserver omniidl omniidl-python libltdl-dev python-matplotlib libtinyxml2-dev liblog4cxx10-dev libltdl-dev qt4-dev-tools libqt4-opengl-dev libqtgui4 oxygen-icon-theme libcdd-dev

install dependencies of openscenegraph:
```
sudo apt-get build-dep openscenegraph
```
Choose a directory on your file system and define the environment variable DEVEL_DIR with the full path to this directory. the packages will be cloned into $DEVEL_DIR/src, the packages will be installed in $DEVEL_DIR/install. It is recommended to set variable DEVEL_DIR in your .bashrc for future use.

Copy Config and Makefile

wget -O $DEVEL_DIR/config.sh https://raw.githubusercontent.com/humanoid-path-planner/hpp-doc/rbprm/doc/config.sh

wget -O $DEVEL_DIR/src/Makefile https://raw.githubusercontent.com/humanoid-path-planner/hpp-doc/rbprm/doc/Makefile

type

cd ${DEVEL_DIR}

source config.sh

cd ${DEVEL_DIR}/src

make robot_state_chain_publisher.install;

source ../config.sh;

make all;

make hpp-rbprm-corba.install

Documentation

Open $DEVEL_DIR/install/share/doc/hpp-doc/index.html in a web brower and you will have access to the documentation of most packages.

Source code for the trajectory generation solver used to generate dynamically feasible Center Of Mass trajectories

The source code of the version used with the TRO paper can be found on github, or downloaded from the present website here. To get the latest version of the solver, please refer to the github repository. The readme file contains information to get started.

Source code for the dynamic simulator used for validating the contact plans

A permanent tag pointing to the version used with the TRO paper can be found on github, or downloaded from the present website here. To get the latest version of the simulator, please refer to the github repository. The readme file contains information to get started.

Details on the decimate tool

To reduce the complexity of a mesh, we used the decimate tool from the version 2.75 of blender. A permanent link to this version along with the documentation can be found here. However, similar results can be obtained with most 3D authoring tools available, which all propose a similar feature. Another option is to implement your own decimation tool, following the instructions of the paper "Surface Simplification Using Quadric Error Metrics", by Garland and Heckbert, for which you can find the complete reference here. Using Blender, for HRP-2 and HyQ, we set a decimation ratio of 0.06 to obtain the meshes shown in the paper.