All research lines

SmartChair

A powered wheelchair driven by brainwaves

A powered wheelchair controlled by brain activity, read through a Brain Computer Interface.

Started:

Phase 4/6 Prototype

  1. Idea
  2. Literature review
  3. Design
  4. Prototype
  5. Testing
  6. Results

Research question

Can a person with a severe motor disability drive a powered wheelchair safely and independently using only their brainwaves?

Context

For people who cannot use a joystick, not even with their mouth or chin, driving a powered wheelchair on their own is often impossible. A Brain Computer Interface (BCI) turns the electrical activity of the brain, measured with electroencephalography (EEG), into commands for a device, without using any muscles.

Goal

SmartChair aims to connect a BCI to a powered wheelchair, so that the person can decide where to go using their own brainwaves, with safety first.

The story of the project

SmartChair is not new. I started it in Bari in early 2014, in my spare time, with the help of my friend Michele Galletta, who took care of the electrical connections. On 2 January 2014 the first wheelchair to build the system on arrived. I had bought it specifically for the project, and every day I sat in it for a while to understand how it feels for the people who really use one.

The starting point was ThinkUino (2013), an open source project of mine that connected an Arduino board to a NeuroSky MindWave Mobile EEG headset. With ThinkUino you could switch a device on just by concentrating.

The idea behind SmartChair was broader: to bring together, in a single low-cost architecture based on open software, assistive tools that usually stay separate. The 2014 prototype was built like this:

Component Role in the prototype
NeuroSky MindWave Mobile EEG headset Reads the brainwaves; an eye blink changes the selected device
Android smartphone on the armrest The hub of the system, talking to the other devices over Bluetooth
Arduino Drives the electronic board of the powered wheelchair
Cupersafety NewDom home automation unit (ZigBee) Switches lights and household devices on and off
Always-listening speech recognition Gives voice commands without touching the screen
GlassUp glasses and Pebble smartwatch Show notifications and the state of the house

In the 2014 demonstrations the prototype switched the lights of a room on and off by voice and by thought, and drove a remote-controlled toy car. The wheelchair part was still being tested, waiting for the right tools. Back then I was already noting the main limits: about 1 s of delay between the thought and the command, data security still to be solved, and a “bare” system with no attention to its appearance.

Why it is now a prototype

Since that prototype exists and is documented, the research restarts from the Prototype phase rather than from scratch. From here on I will review it in the light of what is known today about BCIs and smart wheelchairs.

Warning. SmartChair is an experimental project and is not a certified medical device. A consumer EEG headset like the 2014 one is not enough to drive a real wheelchair safely. Any test on people will require independent stop systems, informed consent and, where needed, the approval of an ethics committee.

Where I am

I have a 2014 prototype to restart from. I will document every step in the journal.

Press coverage and resources

All the sources about SmartChair I have found online, in chronological order. The articles are in Italian.

Articles and reports

Events

Demo videos

Original 2014 materials

  • Project website on blesciasw.it (Web Archive copy from 2 January 2015): architecture, devices, videos and press coverage.
  • Project blog (January 2014): the arrival of the first wheelchair and the app’s settings screen.

Projects SmartChair grew out of