A technology was developed to produce optical devices that can instantaneously recognize objects without additional computer processing. The technology could ultimately be useful for robots, autonomous vehicles, and other applications.
The optical neural network, a “maze” for beams of light, has a series of specially designed translucent wafers made of different materials such as plastic or glass. The wafers have rough surfaces — similar to frosted glass — that are designed at a scale smaller than the wavelength of light to split up light beams into various directions as they travel through the maze.
Because of how the light interacts with the wafers, each sub-band of light exiting the maze is precisely directed to a desired spot on a screen at the end of the array. The device can simultaneously process many wavelengths of light, in contrast to one that could only use a single wavelength. Those earlier devices were able to identify handwritten numbers and clothing items, which are commonly used in tests of artificial intelligence systems. The more wavelengths a network can see, the more it increases the amount of information it can process.
The beam of light directed into the maze was composed of many wavelengths in the terahertz part of the light spectrum. The network was designed using a branch of artificial intelligence called deep learning, in which computer programs “learn” and adjust their responses accordingly based on numerous repetitions of an action or experiment.