Watch a swarm of drones autonomously track a human through a dense forest

Abstraction: Ten-drone swarm autonomously navigates dense bamboo forest using shared depth sensing

Key points:

• Zhejiang University deployed 10 palm-sized drones through dense bamboo forest using only on-board depth-sensing cameras and inter-drone data sharing — no GPS, no prior mapping
• If one drone's path is blocked, it uses neighbors' sensor data to reroute; if a drone loses a tracked human, neighboring drones pick up the trail
• First demonstrated swarm flying autonomously outdoors in a fully unstructured environment (Swiss Federal Institute of Technology Lausanne researcher assessment)
• Two swarm programming paradigms: "insect" (fast reactive, low compute) vs "bird" (long smooth paths, more energy efficient); the Zhejiang team used the bird approach
• Military applications are explicit: multiple nations (US, China, Russia, Israel, UK) are developing drone swarms for surveillance, reconnaissance, and urban warfare
• Published in Science Robotics; proposed civil uses include disaster relief, wildfire reconnaissance, and ecological surveys

Connections: Zhejiang University · Drone Swarms · Autonomous Navigation · Multi Agent Systems

Source: https://www.theverge.com/2022/5/5/23058160/drone-swarm-autonomous-navigation-dense-forest-person-tracking