Scientists from China have created a swarm of drones that can navigate through a dense bamboo forest.

A group of 10 palm-sized drones communicate with one another to stay in formation, sharing data collected by on-board cameras to map their surroundings. If the path in front of one drone is blocked, it can use information collected by its neighbors to plot a new route. The technique can be used by the swarm to track a human walking through the same environment. The others are able to pick up the trail if one of the drones loses sight of the target.

In the future, scientists will write in a paper that drones could be used for disaster relief and ecological surveys.

Drones share data to plan routes and track targets

In natural disasters, a swarm of drones can search, guide, and deliver emergency supplies to trapped people.

The work has military potential according to experts. The US, China, Russia, Israel, and the UK are all developing drones that could be used in war. The same technology could be used to track and attack both combatants and civilians.

An illustration from the paper showing how multiple drones can be used to track a target even if the view from one drone is blocked.
Image: Science Robotics / Xin Zhou et al

The use of drones in combat is one of the specialisms of a senior lecturer at Queen Mary University of London.

The ability to navigate cluttered environments is desirable for a range of military purposes.

“The capability to navigate cluttered environments is desirable for a range of military purposes”

The recent war between Russia and Ukraine has shown how quickly drones can be adapted for use on the battlefield. Cheap consumer drones are being used by both sides in the conflict. One method involves using drones. A recent video shows Ukrainian troops dropping a grenade through the roof of a car that was being driven by Russian soldiers.

drones are potentially more dangerous than lone machines because of their autonomy. Military planners are more likely to use this kind of system in war if it can be off-programmed to an automated system.

Drones in the swarm are capable of navigating through gaps as small as 30 centimeters.
Image: Science Robotics / Xin Zhou et al

In their application, drones are limited. Light shows are the most common real world use-case. In these scenarios, drones are following preset trajectory in open spaces, using tracking technology to locate themselves.

This is the first time that a swarm of drones successfully flying outside. Soria said that the work was impressive.

In their paper, the scientists note that approaches to drone swarms tend to follow one of two programming paradigms. Both methods have tradeoffs, as planning like a bird is more energy efficient than thinking like an insect. Programming bird-like behavior has become more accessible as the computing capacity of hardware improves.

Although the focus in such drone swarm research is often on these technological achievements, this can obscure the trickier questions of how such work should be deployed. She points to the work of 20th century US mathematician Norbert Wiener, who laid the foundations for artificial intelligence.

In the 1960s, weiner said that there was a disastrous focus on and obsession with know-how.