Flying robots might not appear to have much in common with energy grids. To function smoothly, however, both use hybrid systems that combine stop-start action with continuous movement. Developing algorithms for these systems is a specialty of Ricardo Sanfelice, associate professor of computer engineering.
“I enjoy solving applications that combine physics and computational components,” said Sanfelice.
One of those applications enables “smart grids” to link off-and-on sources of renewable energy, such as solar power, with the continuous electric currents of standard electricity. A patent is pending for that hybrid control algorithm.
With a $432,000 National Science Foundation grant, Sanfelice and a colleague from the University of Arizona are now developing a system to enable groups of unmanned vehicles to work together under uncertain conditions.
A key problem is that computer calculations could take longer to make than actual movements—potentially putting vehicles and people at risk. Introducing a metric, called uncontrollable divergence, allows the computer to calculate tradeoffs between accuracy and speed to determine its best trajectory. Sanfelice’s team described this “computationally aware cyber-physical system” in the journal _Autonomous Robots. _
Sanfelice hopes to gather an interdisciplinary group of his campus colleagues to advance the development of these complex cyber-physical systems.