Many science fiction movies feature swarms of flying cars or drones. Now, a team of scientists from Zhejiang University (China) has been inspired by these movies and series to train swarms of drones to navigate through dense bamboo forests in China imbued with a new navigation system.
expert pilots
Although we are used to seeing drones fly in an adjustable formation and avoiding obstacles, their trajectories are usually controlled by a central computer that monitors their positions and issues commands. These drones are autonomous.
As we can see in the video distributed by the Chinese university, a swarm of 10 drones fly together without colliding with each other or obstacles, even in challenging places like a forest. Drones are capable of calculating their location and finding a way forward using a range of sensors.
The biggest advance was an intelligent algorithm that incorporates collision avoidance, flight efficiency, and coordination within the swarm.
Thanks to a new method that reduces the size and hardware requirements of a drone while maintaining its computing power, these small, palm-sized drones weighing 300 grams do not need to lug around heavy hardware to fly efficiently. Its battery barely weighs 100 grams and can keep it in the air for 11 minutes.
Characteristics of innovative drones
The new drones are autonomous, have the computing resources to quickly and effectively perceive and react to the environment through a dense and unknown forest, but are small and light enough that each one can move with agility. It’s a big step toward using swarms like this for fields like aerial surveying and disaster response.
Despite all their technological advances, drones are still limited by their flight times. So instead of flying a drone multiple times to perform a task, a swarm could quickly map an area or survey damage and improve response rates.
This swarm of drones does not rely on a global positioning system (GPS) or external guidance. Therefore, collision avoidance, swarm coordination, and flight efficiency are all encoded in this algorithm, which is quite a feat.
Referencia: Swarm of micro flying robots in the wild Xin Zhou, Xiangyong Wen Zhepei Wang Yuman Gao Haojia Li , Qianhao Wang Tiankai Yang Haojian Lu , Yanjun Cao , Chao Xu Fei Gao. Science Robotics 2022 PMID: 35507682 DOI: 10.1126/scirobotics.abm5954
