China-developed AI drones navigate forests autonomously

A fleet of AI-powered miniature drones developed by Chinese engineers has proven capable of autonomously traversing forests, potentially paving the way for a revolution in disaster response.

The drones, developed by researchers at Zhejiang University in east China, can navigate tight spaces and complex environments without human input.

"While traditional drones require skilled operators to control their flight with a remote controller, we have chosen to let drones make their own decisions, granting them the ability to perceive and observe the environment, make decisions and exercise control without human intervention," said Zhou Xin, a PhD student at the College of Control Science and Engineering at Zhejiang University.

Rather than relying on satellite navigation systems, the drone fleets communicate with each other to autonomously avoid obstacles and coordinate movement.

"With the help of AI algorithms, drones can acquire the environmental information with their sensors acting as their 'eyes' and generate a 3D map of the surroundings. Leveraging this 3D mapping, drones can navigate freely and ensure collision-free movement. Moreover, collaborative algorithms have been developed to facilitate recognition and understanding among multiple drones, allowing for effective control of the entire fleet," said Zhou.

According to their developers, these drones could have a wide range of applications in various scenarios, from tracking the health of forests to emergency response.

"For example, the drones can assist post-earthquake search and rescue efforts or firefighting operations. They can even accurately locate each tree in the forest environment and evaluate their growth status," said Gao Fei, a PhD supervisor from the college.