image: The tiny rescue aerial robots, created by WPI researchers, can fly through low light and hard to reach spaces.
view more
Credit: Worcester Polytechnic Institute
When Worcester Polytechnic Institute (WPI) Professor Nitin Sanket watches birds weave through forests or bats fly effortlessly in the dark, he sees more than nature’s marvels—he sees the future of robotics. That vision has now earned him a major milestone: his first National Science Foundation (NSF) grant, awarded through the highly competitive Foundational Research in Robotics (FRR) program.
Sanket’s project, “Sound Navigation: Enabling Tiny Robots to Find Their Way Through Smoke, Dust, and Darkness,” has been granted $704,908 over three years to develop sound-based navigation systems for small aerial robots that can operate in environments where cameras and light-based sensors fail.
“This is one of NSF’s toughest programs, with extremely high standards,” said Sanket, assistant professor of robotics engineering and computer science. “I was determined to pursue it independently because the project fit so perfectly. Receiving this grant, I feel very accomplished and re-energized to push the boundaries of bio-inspired robot perception forward.”
For more than a decade, Sanket’s work has focused on vision-based autonomy for aerial robots—an approach that mirrors how humans rely on sight. But in conditions like fog, smoke, or darkness, light sensors struggle. Inspired by how bats use ultrasound to navigate, Sanket is developing bio-inspired echolocation systems that allow drones to “see” using sound.
The project will focus on tiny aerial robots—smaller than 100 millimeters and weighing under 100 grams—that navigate independently using sound instead of cameras. To achieve this, Sanket’s team will tackle several key challenges:
- Hardware design: Developing metamaterials that minimize noise interference from drone propellers.
- Software innovation: Using physics-informed deep learning to interpret complex ultrasonic signals.
- Sensor fusion: Combining sound with inertial data to improve navigation reliability.
- Learning systems: Creating a reinforcement learning framework that teaches drones to reach goals while avoiding obstacles.
By integrating these technologies, Sanket aims to build low-cost, energy-efficient drones that can succeed where current systems cannot. Potential applications include search and rescue missions, disaster response, hazardous environment monitoring, and environmental protection.
“This work will enable rapid deployment of robots in challenging environments such as disaster zones or smoke-filled areas,” Sanket said. “It’s about creating tools that support protection, prevention, and preservation in a cost-effective, scalable, and deployable way.”
The broader principles of sound-based navigation may also advance fields such as self-driving cars, coral reef monitoring, and volcanic exploration—any setting where traditional vision systems are limited.
Beyond the technology, Sanket says his greatest motivation comes from mentoring students and exploring uncharted research territory. “The most rewarding part of this work is collaborating with students, discovering new ideas, and solving hard problems in creative ways,” he said. His advice to early-career researchers: “Go for a grant you truly believe in—your passion will show in your writing. Don’t let anyone tell you something is impossible.”
As he looks to the future, Sanket envisions a world where aerial robots are not just machines, but partners in improving human life—saving lives, protecting ecosystems, and expanding our reach into places too dangerous or inaccessible for people.
“I’m excited to keep learning from nature,” he said. “The more we understand how the natural world solves problems, the better we can build robots that make life safer and better for everyone.”