A palm-sized drone using bat-style echolocation achieved 72-100% success in fog and darkness. The engineering is in the efficiency, not the biomimicry.
Researchers at Worcester Polytechnic Institute have built a drone that navigates using echolocation — the same technique bats have used for millions of years.
Published in Science Robotics, the project equipped a palm-sized X-shaped quadrotor with ultrasound sensors and an acoustic shield to dampen propeller noise. The onboard AI analyses weak ultrasound echo patterns to detect obstacles in real time — navigating without cameras, lidar, or GPS.
The test conditions were deliberately hostile: dense fog, total darkness, simulated snow. Thin and transparent obstacles that would fool optical sensors. Over 180 tests, the drone achieved a success rate between 72 and 100 per cent.
"Bats that weigh less than two paper clips can accurately navigate in dark, damp, and dusty caves by sending out short chirps and listening to the weak echoes with a limited number of neurons," said research lead Nitin J. Sanket, assistant professor at WPI's Department of Robotics Engineering.
The engineering insight is not the biomimicry — it is the efficiency. A system that achieves reliable navigation with passive acoustics, minimal power, and onboard computation has genuine applications in disaster response, confined-space inspection, and underground operations.