A revolutionary advancement in drone technology is transforming how unmanned aerial vehicles navigate unknown environments. Professor Fu Zhang and his team at the University of Hong Kong have developed a system that enables drones to fly with bird-like agility through complex spaces at remarkable speeds without relying on GPS or pre-mapped routes.
The technology, called SUPER (Safety-assured high-speed navigation for MAVs), was unveiled on June 7, 2025, and represents a significant leap forward in autonomous flight capabilities. The system utilizes a lightweight 3D LIDAR sensor that can detect obstacles up to 70 meters away with pinpoint accuracy. What makes SUPER truly innovative is its advanced planning framework that simultaneously generates two flight trajectories—one optimizing for speed by exploring unknown spaces, and another prioritizing safety by remaining within known, obstacle-free zones.
"Picture a 'Robot Bird' swiftly maneuvering through the forest, effortlessly dodging branches and obstacles at high speeds," explains Professor Zhang. "It's like giving the drone the reflexes of a bird, enabling it to dodge obstacles in real-time while racing toward its goal." In tests, the system has demonstrated the ability to navigate at speeds exceeding 45 mph (20 meters per second), even through dense forests.
Concurrently, researchers at Texas A&M University are addressing another critical challenge for small drones: energy efficiency. Dr. Suin Yi's team is developing neuromorphic computing systems that mimic the human brain's processing methods. Their approach uses conducting polymer thin films as artificial neurons that activate only when necessary, dramatically reducing power consumption.
"Small drones don't have an engine, so their energy budget is very small," notes Dr. Yi. "Adding AI to battery-powered drones can reduce flight time from 46 minutes to just four minutes." The neuromorphic computing solution could allow drones to maintain their full flight duration while performing complex AI tasks like object recognition and autonomous navigation.
The applications for these technologies span numerous fields, including search and rescue operations, forest monitoring, power line inspection, and autonomous delivery. In disaster scenarios, drones equipped with SUPER technology could navigate collapsed buildings or dense forests day and night, locating survivors or delivering critical supplies to remote areas with unprecedented speed and reliability.