Researchers develop sonar-based navigation system for small drones inspired by bat echolocation, reducing power by 1,000 times compared to current solutions.

Researchers at Worcester Polytechnic Institute developed an ultrasound-based perception system for aerial robots that uses sonar (sound-based sensing) inspired by bat echolocation. The system includes a physical acoustic shield that reduces propeller noise and a neural network called Saranga that recovers weak echo signals from noisy measurements.

The drones can estimate obstacle locations in 3D and navigate safely in low-visibility environments—smoke, fog, dust, snow, or complete darkness—where cameras and lidar (light detection and ranging) sensors fail. The system operates using milliwatt-level sensing power.

According to the researchers, the work reduces power by 1,000 times, weight by 10 times, and cost by 100 times compared to current solutions. The drones are intended for search and rescue, cave exploration, environmental monitoring, and other humanitarian applications in environments where human or larger helicopter access is limited.