Engineers create circuit boards that mimic human brain

Even though scientists have made leaps and bounds in computer innovations, we may be able to admit that no computer is equivalent to the human brain.

So, engineers from Stanford University decided to use the human brain as a starting point when developing a new type of circuit board.

 
The “Neurogrid”. (Image via Stanford University, YouTube)

Not only are PCs slower than human brains, but it also takes 40,000 times more power to run one.

Associate Professor Kwabena Boahen and his bioengineering team at Stanford took these facts and created the Neurogrid, a circuit board consisting of 16 custom-designed “Neurocore” chips. These 16 chips can simulate one million neurons and billions of synaptic connections.
 
“From a pure energy perspective, the brain is hard to match,” says Boahen.

These advancements could have potential significant impact on the robotics and computing industries.

While the team was designing the chips, they made sure to keep power efficiency in mind. Their strategy was to enable certain synapses to share hardware circuits and what they came up with was a $40,000 device prototype about the size of an iPad with speeds and low-power characteristics that make it ideal for more than just modeling the human brain.

Why we care about this circuit board: Future impact

Next, Boahen and team will try to lower the costs and create compiler software that would enable engineers and computer scientists with no knowledge of neuroscience to solve problems such as controlling a humanoid robot using Neurogrid.

“Right now, you have to know how the brain works to program one of these. We want to create a neurocompiler so that you would not need to know anything about synapses and neurons to able to use one of these” said Boahen.

Boahen is also working to develop prosthetic limbs that could be controlled by a Neurocore-like chip. A chip as fast and efficient as the human brain could drive these prosthetic limbs with the speed and complexity of our own human actions without being tied to a power source.

Working on prosthetic limbs in the lab. (Image via Stanford University, YouTube)

Power is still an issue that Boahen and his team face, though. Even though the human brain has 80,000 times more neurons than the Neurogrid, it only consumes three times as much power. Achieving this kind of energy efficiency is a major challenge engineer’s face.

Stanford offers a much deeper look into the team’s research. Read about it here.