A University of Technology Sydney (UTS) project has produced a brain-computer interface that uses human brainwaves to control machines.
The project is a collaboration between Professor Francesca Iacopi and Distinguished Professor Chin-Teng Lin in UTS’ Faculty of Engineering and IT.
The research is funded by $1.2 million from the Defence Innovation Hub. It has three main components: a biosensor to detect electrical signals from the brain, a circuit to amplify the signals, and an artificial intelligence decoder to translate them into instructions – stop, turn right, turn left – that the machine can understand.
“I see this technology as the next generation of human-computer interfaces,” Professor Lin said.
Professor Iacopi, a leading researcher in nanotechnology, is leading the development of the biosensor, which is worn on the head. The sensor is made of epitaxial graphene – essentially multiple layers of very thin, very strong carbon – grown directly onto a silicon carbide on silicon substrate.
The result is a novel biosensor that overcomes three major challenges of graphene-based biosensing: corrosion, durability, and skin contact resistance, where non-optimal contact between the sensor and skin impedes the detection of electrical signals from the brain.
“We’ve been able to combine the best of graphene, which is very biocompatible and very conductive, with the best of silicon technology, which makes our biosensor very resilient and robust to use,” says Professor Iacopi.
Professor Lin is working on AI brain decoding technology. He said that his team have achieved two major breakthroughs in their work so far.
The first was figuring out how to minimise the noise created by the body or the surrounding environment so that the technology can be used in real-world settings. The second was increasing the number of commands that the decoder can deliver within a fixed period of time.
“Current BCI technology can issue only two or three commands such as to turn left or right or go forward,” Professor Lin said.
“Our technology can issue at least nine commands in two seconds. This means we have nine different kinds of commands and the operator can select one from those nine within that time period.”