Students at Switzerland’s ETH Zurich have developed a robot which works 30 times faster than the quickest human.
The Institute of Neuroinformatics’ Sensors Group at the ETH Zurich has used a brain-inspired neural network and camera to develop ‘Dextra — a robotic hand’ which essentially reads your mind anticipating your hand gestures to defeat you in the rock-paper-scissors game.
Digital neural network
“Dextra sees the world in super slow motion. It is so fast at seeing the symbol you are about to throw that it is able to both determine your next move and execute a winning symbol 30 times faster than the quickest human,” said ETH Zurich Professor, Tobi Delbruck. “Unlike conventional artificial intelligence (AI) vision systems based on image frames, Dextra wins by being frame-free.”
Conventional robotic or computer systems must continually process frames at a very high rate in order to react quickly. High processing rates in the camera and the computing system sap energy and power.
“Dextra’s AI vision technology drives the motion in the scene by using a silicon retina Dynamic Vision Sensor (DVS) camera and a custom neural network accelerator called, ‘NullHop’.
“The robotic AI computation occurs only when necessary enabling the system to always react quickly, while at the same time, optimising energy use. It is a digital convolutional neural network (ConvNet) accelerator that determines which symbol the human game player displays,” he added.
Comments
Comments have to be in English, and in full sentences. They cannot be abusive or personal. Please abide by our community guidelines for posting your comments.
We have migrated to a new commenting platform. If you are already a registered user of TheHindu Businessline and logged in, you may continue to engage with our articles. If you do not have an account please register and login to post comments. Users can access their older comments by logging into their accounts on Vuukle.