FroboMind

This guide describes how to install FroboMind based on ROS Kinetic and Ubuntu 16.04 LTS on either a standard PC or a Raspberry Pi 3 model B.

Please notice that this version is considered the FroboMind current stable version. If you want to install another version please visit the FroboMind installation page.

The first step of installing FroboMind is to install Ubuntu 16.04.

For PC please click the link Ubuntu Desktop 16.04 LTS (Xenial Xerus) to download an installation image.

For Raspberry Pi please click the link Ubuntu Mate 16.04 LTS to download an installation image.

FroboMind will also work on Raspbian, however ROS install on Raspbian requires wet compilation of all required ROS packages. The result is a nice barebone system with only the software you need, however the processes of installation and subsequent addition of ROS packages are significantly more complicated.

The next step is to install ROS on top of your Linux installation.

ROS is an abbreviation for Robot Operating System which provides libraries and tools to help software developers create robot applications. It provides hardware abstraction, device drivers, libraries, visualizers, message-passing, package management, and more.

For information about how to install ROS on Ubuntu please go to this page: http://www.ros.org/wiki/kinetic/Installation/Ubuntu

For PC we recommend that you install the ROS-Desktop-Full configuration.

For Raspberry Pi we recommend that you install the ROS-Base configuration to minimize the installation.

If you have installed the ROS-Desktop-Full configuration, there are currently no required dependencies.

If you have installed the ROS-Base configuration you will need to install the software packages listed below:

sudo apt install git
sudo apt install catkin
sudo apt install python-matplotlib
sudo apt install ros-kinetic-tf ros-kinetic-stage ros-kinetic-stage-ros

If you wish to run the MarkerLocator software, you will also need to install the software packages listed below:

sudo apt install python-opencv
sudo apt install ros-kinetic-cv-bridge

Run these commands to create a ROS workspace directory under your user home dir. If you are new to the linux, please notice that the shortcut to paste in the terminal is CTRL-SHIFT-V.

mkdir -p ~/roswork/src

Clone the FroboMind repository and the Frobit demo application to the ROS source directory:

cd ~/roswork/src
git clone http://github.com/frobolab/frobomind.git
git clone http://github.com/frobolab/frobit_demo.git

ROS Groovy is compiled using catkin, however in FroboMind a script in the ~/roswork/src/frobomind directory handles compilation of the components you need for a particular application, so you don't have to deal with catkin directly. Run these commands to compile and run the Frobit demo application:

cd ~/roswork/src
frobomind/frobomind_make /frobit_demo

During the compilation process catkin has created two new directories, so now your roswork directory contains:

~/roswork/src
~/roswork/devel
~/roswork/build

To configure ROS for this workspace run:

echo "source ~/roswork/devel/setup.bash" >> ~/.bashrc
source ~/.bashrc

The first line above configures your current shell, the second line configures all new shells that you open.

To be able to run the frobomind_make script from everywhere run:

echo 'PATH="$PATH:~/roswork/src"' >> ~/.bashrc

When you have completed the installation above you are ready to begin using FroboMind. This is where it gets really exiting. We recommend that you:

• Try out our Frobit demo to test your installaion.
• Familiarize yourself with the FroboMind architecture.
• Learn how to create a new FroboMind application and write a new FroboMind component using the Tutorials.