The NAT Brain Drone is a general purpose BCI classification controller that is specifically tailored to control a drone. This controller is fundamentally designed to be expandable and general purpose so that it can realistically control whatever you can connected to the computer. For more information about this project, visit its website or view the submission video on youtube.
- A 16 channel OpenBCI (Cyton + Daisy)
- A DJI Tello drone
- A MacOS machine (should also work on Linux and Windows, though it has not been tested)
- python >= 3.6.5
To open terminal, press command + space and type terminal then press enter
Navigate to your desired install location using
cd [directory] (For example, cd Desktop)
Clone this repo
git clone https://github.com/neuralbertatech/openComp2020
Navigate into this repo
cd openComp2020
Install homebrew
ruby -e "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install)"
Install venv
brew install venv
Install pyenv
brew install pyenv
Configure a virtual environment to contain dependencies
pyenv install 3.6.5
pyenv local 3.6.5
python3 -m venv venv
Activate the environment
source venv/bin/activate
Update Pip installer
pip install --upgrade pip
To install the required software run
pip install -r requirements.txt
Not yet tested
Not yet tested
Activate the virtual environment with
pyenv local 3.6.5
and
source venv/bin/activate
Then finally, run the command
python masterController.py
and follow the onscreen instructions.
If you are running the program after a training session, and have not yet moved the OpenBCI from where it was when the baseline was collected, you can reuse this data and skip the lengthy baseline collection process by calling
python masterController.py nocol
By default, the previous session will be saved to ~/TrainingData/masterControllerSessions. If you would like to use data in another directory, you can call
python masterController.py nocol [directory relative to this file]