Project combines noninvasive brain imaging techniques, animal behavior experiments, & signal processing models to gain a better understanding of how animals use echolocation to gather information about their environment.
Dolphins and porpoises exploit their natural sonar – called echolocation – to move in complicated underwater environments, even in low visibility conditions. They transmit short high frequency clicks, then listen for the returning echoes to avoid obstacles, find their prey, and sometimes even return sunglasses that human trainers have lost in the water. “The animals often find objects faster and more reliably than manmade sonars, suggesting that they extract information from sound more efficiently than engineering systems,” says Chancellor Professor of Electrical and Computer Engineering Professor John R. Buck.
Dr. Buck is part of a five-university team led by Carnegie Mellon University that recently received a Multidisciplinary University Research Initiative (MURI) grant from the Office of Naval Research to study the echolocation strategies of marine mammals. Buck’s $434,000 portion of the grant will create acoustic signal processing models for the information in marine mammal sonars.
“The project combines noninvasive brain imaging techniques, animal behavior experiments, and signal processing models to gain a better understanding of how the animals use echolocation to gather information about their environment,” Buck says. The project also investigates how maximizing the information guides the animals’ search strategies.
Abigail Keith ‘20, who earned her BS in electrical and computer engineering recently defended her MS thesis, worked with Buck in his lab at UMassD. Keith implemented biologically-inspired search algorithms on the iRobot Create, the educational version of the popular Roomba cleaning robot. While conducting research experiments with the robot, she demonstrated that the dolphin-inspired algorithms are faster on average than other common search algorithms when searching for difficult to detect objects.
Keith says independent research as a graduate research assistant allowed her to explore topics she was interested in through hands-on experience and in-depth study. “I found it immensely valuable to have a thesis advisor guide me through my research.” Keith, who is an Electrical Engineer 2 at Raytheon Technologies, had the opportunity to have my mentor from Raytheon serve on my thesis committee. “Conducting research as a graduate student truly has helped me to feel confident and capable in my knowledge and abilities as I begin my career as an engineer,” Keith says.
“The insights gained from this research will ultimately lead to better manmade sonars to allow underwater robots to navigate more safely and locate objects in difficult conditions, such as around oil platforms or crowded harbors,” Buck says.