The Czech Republic claims significant success in the global race in drone development. The means by which the geopolitical situation has opened the future and the arms of investors are, according to Fortune Business Insights, to create a market worth 210.26 billion dollars by 2034. And the transfer of technology from the Czech Technical University's Faculty of Engineering and Technology through the spin-off company Fly4Future has now come up with a new product that has been awarded in the Fully Autonomous Flight 2.0 competition of the German Federal Agency for Breakthrough Innovation SPRIND.
The current drone market is worth around 100 billion dollars globally. Europe holds only a fraction, around seven billion dollars. Although attention is focused on their development mainly due to ongoing military conflicts, they are also changing other fields in addition to the battlefield. These include construction and inspection or precision agriculture.
The innovation that scientists have now brought is controlling a drone using speech, without the need for GPS navigation and manual control. The technological basis of the Czech system is four cameras, 3D LiDAR and a powerful on-board computer. Through them, the drone locates itself without GPS, maps the surroundings and recognizes objects necessary for the completion of the assigned mission.
Detected buildings, people, vehicles or other objects are inserted into a continuously created spatial map. Above it, a graph structure is created that captures not only individual objects, but also the relationships between them - for example, that a bicycle is in front of a specific house or that a person is standing in its window. The system can then "ask" the map where the object or even a complex situation corresponding to the human task is located, and plan further movement based on the answer.
A stand-alone drone that creates its own map
However, the biggest shift does not lie in one specific sensor or algorithm, but in the interconnection of the entire chain. The customer's instruction is transformed into a mission plan without the intervention of a technician, the drone creates a map of the unknown environment, finds the required objects and decides on its own further course of action.
“The simple connection of the end user and the robotic system, which does not require the involvement of a robotics expert or programmer, is one of the last technologies that prevents aerial robotics from massively expanding in many areas of human activity,” commented the importance of the challenge being solved by the leader of the F4F & CTU SkyGraph team, professor at the Faculty of Electrical Engineering of the Czech Technical University and founder of Fly4Future Martin Saska.
The capabilities are best demonstrated by the recent Fully Autonomous Flight 2.0 competition, in which seven European technology teams participated. It itself lasted 14 months and each of the finalists received half a million euros for system development. The final was then held at the Fliegerhorst Erding air base in Bavaria. Each team had to demonstrate its capabilities in four disciplines.
It is clear that no one was without error in the tests. Not even the Czech team, whose technology proved to be the best of all in the final challenge called Helping an Injured Person.
How to find an injured person
The drone was supposed to find the “injured person in the red shirt” and deliver a medical package to him. The Czech system independently discovered the extra, who was lying partially hidden behind a building and leaning against a tree. However, it did not find enough space immediately next to him for a safe landing, and therefore refused the maneuver.
In another attempt, it landed on a nearby roof – geographically close, but practically out of reach of the injured person. The test thus showed that the robot needs to know not only the formal goal, but also the broader meaning of the mission. However, despite these complications, the Czech team once again presented the best solution in this task. Moreover, during the demonstration to investors on the final day of the competition, it was the only team to solve the assigned task repeatedly and completely flawlessly.
“I was pleased that we won this task, because many Fly4Future clients are asking for a similar system, and in addition, we recently won a large national project focused on solving the problem of fires in buildings, where finding and rescuing people is always a priority,” commented Tomáš Báča, technical team leader, researcher at the Faculty of Electrical Engineering and Mechanical Engineering of Czech Technical University and co-owner of Fly4Future, on the competition round.
Ready for production
The result of the competition is not just a one-off prototype prepared for four specific tests. Fly4Future is already gradually integrating the achieved technologies into its robotic platforms and preparing them for commercial use.
According to its creators, a generally usable drone controlled via natural language could help rescuers search for people, firefighters locate fires and control outbreaks, security dispatch centers monitor large areas, or industrial companies inspect hard-to-reach places. It can also be used in parcel delivery, smart agriculture, nature conservation or mapping areas affected by an emergency.
The key business benefit lies in the reduction of operator requirements. One more versatile system can gradually take over different types of missions without having to create a separate drone and new control software for each deployment. Natural language also allows the technology to be operated by people who know the real problem – such as dispatchers, firefighters, inspectors or rescue team members – without the need for detailed knowledge of robotics and programming.