Neurogrid, a neuron-like computer chip consisting of millions of silicon neurons
The human brain is a powerful information processing system. It is composed of about 10 billion nerve cells, or neurons and each neuron is connected to other neurons through about 10,000 synapses. The brain's complex network of neurons forms a massively parallel information processing system as contrasted to conventional computers, in which a single processor executes a single series of instructions.
The brain's neural network is seemingly described as a 'chaos' because the communication patterns among neurons vary much according to specific situation. This 'chaos' creates electrical impulses which when simulated by a robot will require at least 10 megawatts to operate, the same amount of energy produced by a small hydroelectric plant.
However, a computer scientist from Stanford University, Kwabena Boahen, developed a neuron-like computer chips which will require only at least 20 watts, close to the actual human brain's energy consumption. This microchip which initially consisted one million silicon neurons is called Neurogrid, but can accommodate more of up to 64 million to simulate the brain of a mouse.
Boahen and the team of scientists developed neurogrid primarily to provide a solution in brain simulation with an efficient energy usage.
Click here to read the full story.
