Imagine a drone that can dodge tree branches at night like a bird. Now stop imagining—because scientists in Hong Kong have just built it.
Key Points at a Glance
- HKU researchers developed SUPER, an ultra-agile autonomous drone
- It avoids obstacles as small as 2.5 mm and flies over 20 m/s
- The system runs entirely on onboard sensors and computing
- SUPER uses LIDAR and dual real-time trajectory planning
- Applications include search and rescue, exploration, and delivery
While most drones still stumble over complex terrains or need pre-mapped paths, a new invention from the University of Hong Kong could change everything. Inspired by nature, and engineered for the future, the SUPER drone—short for Safety-Assured High-Speed Aerial Robot—navigates like a bird on adrenaline, weaving through forests, ducking under power lines, and flying blind in total darkness.
Developed by Professor Fu Zhang and his team at the HKU Faculty of Engineering, SUPER combines advanced autonomy with high-speed maneuverability. Its key breakthrough? Fully onboard navigation and obstacle avoidance that works even when visibility is zero and danger is everywhere.
“This is a significant leap in autonomous flight,” says Zhang. “SUPER gives micro air vehicles reflexes more like those of a bird than a robot. It can navigate tight and dangerous spaces in real time without external help.”
The drone’s performance is staggering. With a wingspan of just 280 mm and weighing only 1.5 kg, it can fly over 20 meters per second while avoiding obstacles as narrow as 2.5 mm—things like bare wires, antennae, or twigs. What makes it truly groundbreaking is its complete independence from GPS or centralized computing. Everything happens onboard.
SUPER uses a compact LIDAR sensor capable of detecting objects up to 70 meters away, feeding raw point cloud data directly into its ultra-efficient onboard processor. This enables the drone to generate two simultaneous flight paths: one fast and adventurous, venturing into unmapped space, and one conservative and safe, sticking to known territory. Depending on conditions, it dynamically switches between the two—balancing risk and speed like an aerial chess master.
Lead researcher Yunfan Ren emphasizes the practical potential: “Because it can see in the dark and avoid even the thinnest obstacles, SUPER is perfect for disaster zones. It could navigate collapsed buildings, find survivors, or deliver aid where no human or vehicle could reach.”
The drone has already been field-tested in indoor and outdoor environments, including dense forests and heritage sites. Unlike traditional drones, SUPER doesn’t need a map. It builds one on the fly—literally.
This makes the system ideal not just for emergencies, but for commercial and environmental use as well. From inspecting high-voltage lines to mapping forests or ancient ruins, SUPER introduces a new generation of MAVs that are faster, smarter, and vastly more adaptable.
The core idea behind SUPER isn’t just speed. It’s safety at speed. In robotics, those two goals usually conflict. Speed tends to mean risk; safety, hesitation. SUPER’s innovation lies in a real-time planning algorithm that reconciles both. It embraces the unknown—without crashing into it.
Professor Zhang envisions a future where drones routinely navigate dangerous environments that humans can’t enter. “We’ve equipped machines with instincts. Now, they don’t just fly—they react,” he said.
As we push the boundaries of what autonomous systems can do, SUPER may become the new benchmark for aerial robots—particularly those that must make life-saving decisions in milliseconds. In the race between nature and machine, this bird-like drone has just taken the lead.
Source: The University of Hong Kong
Enjoying our articles?
We don’t show ads — so you can focus entirely on the story, without pop-ups or distractions. We don’t do sponsored content either, because we want to stay objective and only write about what truly fascinates us. If you’d like to help us keep going — buy us a coffee. It’s a small gesture that means a lot. Click here – Thank You!
