Visions of a safer, less-painful world seem to drive the researchers in the College’s applied science NDE laboratory. Whether creating the technology to scan prospective terrorists from afar using sound waves, perfecting the ability to interpret the sensor feedback necessary for the next generation of autonomous robots or simply working to replace reliance on hard-tipped dental probes with a gentler alternative, the graduate students, working under the direction of Mark Hinders, professor of applied science, have tomorrow well in hand.
Although these and other projects under investigation may appear disconnected, each relies on the manipulation of “ultrasonic waves,” according to Hinders.
“We are developing the artificial intelligence that allows us to make sense of these very complicated signals,” Hinders explained. “We are trying to build the sort of artificial brains that do what bats automatically do.”
During a recent tour of the laboratory located in the basement of the admissions building on Jamestown Road, Hinders displayed the high-tech speaker—a parametric array, he called it—that enables an operator to focus a beam of sound onto a subject and to interpret the echoes created. He explained how the same interpretive strategies are advancing abilities to inspect a variety of structural materials, including those used in aircraft, ships and even howitzer armaments.
For the most part, graduate students are conducting the individual experiments. They then spend a considerable portion of their time writing the algorithms that essentially interpret their results.
“A key part of these are the supercomputer animations that we create in order to understand the complex physics involved,” Hinders said. “So, part of it is analysis, part of it is experimental work in the laboratory or out in the field, and part of it is computer-simulated animations that allow us to very rapidly see how changes in the signals relate to changes in the materials.”
The creation of such animations is possible due to the College’s SciClone computer cluster. He calls it “the big iron.”
“It helps us understand what we are discovering,” he said, and then we write very efficient algorithms that can run on a modest computer that can be used in a dental office, for instance, or go out on a robot.”