Advances in chip-making technology, which lead to enhanced performance, translate directly to improved performance in autonomous robots. And yesterday brought news of significant chip-production technologies from two U.S. universities.
“Computer engineers at North Carolina State University have developed hardware that allows programs to operate more efficiently by significantly boosting the speed at which the ‘cores’ on a computer chip communicate with each other,” says a statement from the school.
“Our technology is more efficient because it provides a single instruction to send data to another core, which is six times faster than the best state-of-the-art software we could find,” says James Tuck, an assistant professor of electrical and computer engineering at NC State and co-author of a paper describing the research.
Meanwhile, “University of Maryland researchers have made a breakthrough in the use of visible light for making tiny integrated circuits,” a news release from the school notes. If available for commercialization within a decade as researchers hope, the work, described in an article in Nature Chemistry, might allow chipmakers to continue pushing the limits of chip performance and cost, without resorting to more exotic methods.
Connecting the dots here, improved communication speed, together with the ability to create higher performance chips economically, could lead to new generations of dedicated processors in an area such as robotics vision. Such chips would lead to significant improvements in power consumptions and speed over the off-the-shelf processors commonly used today.
One person working in this area is Eugenio Culurciello, associate professor of electrical engineering at Yale University. As Culurciello notes on his website, “we are developing NeuFlow: a dataflow computer for bio-inspired visual processing aimed at speeding up the computations of state-of-the-art vision algorithms and holistic vision systems. NeuFlow is a custom microprocessor with arrays of computing elements suited to accelerate computations on videos and streaming data (data flow).”
NeuFlow is especially suited to neural applications. And in addition to vision, the technology might lead to robotic and intelligent systems’ applications in areas such as “security, monitoring, and also in posture recognition, assisted living, and remote care of elderly patients.”