This repository hold code that runs on the main computing board of an OpenAg food computer (usually a Raspberry Pi). It runs on top on ROS and uses CouchDB for data storage.

There is a pre-built Docker image for running this codebase on ARM machines. Most people should use this image. To set it up, follow the instructions in this repo.

If you plan to modify the code in this repository, it might make more sense to install the software directly on your machine instead of running everything in a Docker container. For Raspberry Pi, this can be done by following the instructions in this repo.

If neither the Docker image nor the install script work for you, for whatever reason, you will have to install each component of the system yourself. These instructions describe how to do so.

First, install ROS Indigo on your machine. There are installation instructions here for doing so. For Raspberry Pi, there are instructions here instead.

Next, install an instance of CouchDB on your machine. There are installation instructions here for doing so. In newer version of Ubuntu (13.10 and up), this can be done via sudo apt-get install couchdb.

Create a catkin workspace as described here and install the code from this repository in the workspace.

cd ~/catkin_ws/src
git clone https://github.com/OpenAgInitiative/openag_brain.git -b ros
cd ..
catkin_make
catkin_make install
rosdep install -i openag_brain

To run any of the openag commands, you must first activate the catkin workspace as follows:

source ~/catkin_ws/devel/setup.bash

Finally, you must install platformio. platformio and rosserial use different versions of pyseial, so installing them both on the same machine will break things. Instead, platformio should be installed in a python virtual environment. There is a script in openag_brain that will do this for you:

rosrun openag_brain install_pio

With everything installed (not through Docker) and with the catkin workspace active, the project can be run as follows:

rosrun openag_brain main