Palm-sized drones with sensors mimic bats to navigate blind conditions

Researchers at Worcester Polytechnic Institute (WPI) have developed palm-sized drones that mimic bat echolocation to navigate through challenging conditions such as fog, smoke, darkness, and cluttered spaces. These X-shaped quadrotor drones, about six inches wide and weighing roughly one pound, use ultra-light ultrasound sensors combined with AI and deep learning to interpret echo patterns, enabling obstacle detection and autonomous flight with minimal onboard computation. An acoustic shield reduces noise from the drone’s propellers, improving the clarity of ultrasound echoes. This approach offers a low-cost, energy-efficient alternative to traditional navigation systems like lidar, radar, or cameras, which tend to be heavier, more power-hungry, and less effective in visually obscured environments. The drones were tested indoors and outdoors, successfully navigating through complex obstacle courses including transparent and metal poles, as well as simulated environmental hazards like fog and snow. In 180 trials, the drones achieved success rates between 72% and 100%, demonstrating strong capabilities in visually impaired conditions