Brain-computer interfaces sound sci-fi but are in fact a reality. There are many different methods used to detect and interpret electrical activity in the human brain so that it can be used to control a computer. Usually, this is the domain of university research departments and places like Google Labs, but a paper on the arXiv (opens in new tab) open access server, by Ildar Rakhmatulin and Sebastian Volkl, proposes doing away with hyper-expensive equipment and using a Raspberry Pi (opens in new tab) instead.
Rakhmatulin, who has form in this area having designed a laser turret (opens in new tab) for shooting down mosquitos using a Raspberry Pi 3 B+ (opens in new tab) and camera (opens in new tab), studies BCIs at Russia’s South Ural State University, while Volkl works in the fields of AI and neural networks. They came together to launch hackerBCI (opens in new tab), which aims to offer easy access to neuroscience projects and experiments.
Their Pi project, PiEEG, uses C, C++ and Python to read up to eight real-time electroencephalography (EEG) signals, gathered from the brain by electrodes placed in a cap worn by the subject. From this, the pair have posted on GitHub use cases including controlling robots and drones. The control of exoskeletons is also among the uses suggested.
The PiEEG board is shown fitted to a Raspberry Pi 3, judging by the full-size HDMI port, as a HAT (a Pi 4 (opens in new tab) can also be used) and connects through the GPIO pins. There's some stringent requirements to be hit before it will work, including isolation from the noise generated by mains power, but luckily there's plenty of open-source software available for signal processing, which is listed in the paper.
If you're keen to start reading your brain signals with your Raspberry Pi, hackerBCI will begin crowdfunding the PiEEG soon.
