Decentralized Drone Control Without Mobile Radio How Mesh Networks Help to Control Swarms of Drones

Swarms of drones can monitor progress on large construction sites, monitor remote offshore wind farms and search for survivors of natural disasters. If several drones are used, they have so far had to be controlled centrally via mobile radio. Researchers at the Fraunhofer Institute for Integrated Circuits (IIS) are working on a decentralized, less interference-prone and more flexible solution.

Drones operating in a swarm can perform different tasks and exchange information with each other. Until now, civilian drone swarms have often been controlled centrally via a mobile phone connection. With regard to autonomous missions, the individual drones can therefore only communicate with each other in a star shape via a central node—usually a mobile phone mast. However, if this hub is damaged, for example as a result of a natural disaster, it is no longer possible for the swarm to communicate with each other. This makes it more difficult to search for survivors, for example, as the swarm can no longer work cooperatively.

Mesh Networks Enable Information Exchange in A Swarm

Researchers at the Fraunhofer Institute for Integrated Circuits IIS in Erlangen are working on a solution: "We are developing a mobile ad-hoc mesh network in which several drones are connected to each other decentrally and communicate via Bluetooth, for example," explains Dr. Manuel Schrauth, group leader in the Communication Systems department at Fraunhofer IIS. As a result, the exchange of information in the swarm is no longer star-shaped, but completely decentralize— i.e. across several nodes if necessary. "The individual drones can still exchange information with each other even if there is no external communication connection—a decisive advantage when you consider that only a fraction of the entire earth's surface has good mobile connecitivity.

So far, highly dynamic, decentralized networks have mainly been used in the military sector, where the technology is expensive, large and only suitable for heavy drones. There are still no suitable mesh networks in the ultra-low-power sector. "As the connection can run via Bluetooth, for example, our technology is very inexpensive and energy-saving," says Manuel Schrauth. "It is therefore suitable for small, lightweight drones, such as those used in the civilian sector."

The idea for the development of extremely energy-efficient ad-hoc mesh networks originated in projects of the so-called GAIA initiative: Schrauth and his team are working with an interdisciplinary consortium on satellite-based animal transmitters for vultures in Namibia that can communicate with each other in a decentralized manner. This enables the research team to analyze the animals' behaviour—and draw conclusions about environmental changes, among other things.

Constant Exchange of Information—Even Without Mobile Communications

The team is now building on this knowledge to network swarms of drones: The researchers are developing a table-based system for this purpose. The individual drones, which now act as nodes, send out so-called heartbeats, also known as advertisements in technical jargon, at regular intervals. This is comparable to Bluetooth headphones, which flash and send out the signal "I'm here" as they search for devices ready to be paired, says Schrauth.

As soon as two nodes have registered each other, they complete their respective tables: There they record which other devices have been sighted. If this process takes place every few seconds, an almost complete knowledge of the entire network can be built up. Although each individual node only has incomplete information locally, a data packet can be spread out successively, says Schrauth: "If node number one is looking for node number seven, for example, it can pass the information on to the nearest point, node two, which in turn passes it on to node three and so on. A data packet therefore winds its way through several nodes until it reaches its destination; the necessary information is assembled iteratively."

Task Sharing Thanks to Decentralized Communication

According to Schrauth, communication in the swarm will later feel like a one-to-one connection for the users. In practice, thanks to the new technology, several drones could automatically take off from a suitcase in the future, form a given formation in the mesh network, thereby increasing the range of the swarm and using their different technical strengths to create a three-dimensional map, for example.

The researchers have already shown that the technology works in a simulation, and now they want to transfer it to commercial hardware. Studies are currently underway in the laboratory with Bluetooth devices that process data across multiple nodes.

The project group is expected to present a prototype of the technology next year, and Manuel Schrauth and his team will be showing initial results at this year's ILA.(sg)

Smart-Factory

Access solutions for the Smart Factory - Industry 4.0