After forest fires dominated the headlines nationwide this summer, early detection methods are coming into focus: In addition to long-term forest conversion, Saxony-Anhalt also wants to keep an eye on and evaluate monitoring with drones and satellite technology for early detection of forest fires. The aim is to “keep the time interval between the outbreak of the fire and the report to the fire brigade control centers as short as possible,” explained the Forestry Ministry in Magdeburg when asked by the German Press Agency. The country is currently relying on “a functioning, AI-supported camera system at 16 locations that takes this requirement into account”.
As early as 2023, researchers carried out test flights with AI-supported drones in the Harz Mountains in a pilot project. Simon Gehringer was involved in the tests as team leader of Evolonic, a research group at the University of Erlangen-Nuremberg at the Fraunhofer Institute for Integrated Systems and Device Technology. According to Gehringer, the test in the Harz region was successful. “We recognized the forest fire back then, relatively quickly, despite the fact that it was so small,” he told the German Press Agency. The burning area at that time was a maximum of around one meter. She was recognized from two to three kilometers away.
Researchers: Drones provide important data on fire behavior
Researchers from the Otto von Guericke University Magdeburg and the Federal Institute for Materials Research and Testing published the research results of this and other experiments in Saxony-Anhalt and Brandenburg in the specialist journal “Fire and Materials” in January: The data obtained provided important insights into the course of the fire, smoke development and the behavior of forest fires for future fire management models, they concluded.
According to drone expert Gehringer, experience has shown that the technology can detect fires within a few minutes under favorable conditions. In the best case scenario, targeted deletion can occur much more quickly. The researchers are also currently working on a fuel cell drive for their drones, which will enable largely automated flights lasting eight hours.
At the same time, the researchers are currently testing dog-like walking robots that could explore the forest on the ground during existing fires in areas contaminated with ammunition. However, the technology has hardly been used systematically in forest fire prevention so far, says Gehringer. Instead, there is a “wild proliferation” of drone technology that fire departments are purchasing on their own. However, this is usually only used in the event of a fire in order to be able to assess the situation for extinguishing work.
Criticism: Municipalities are more likely to invest in classic equipment
Gehringer believes that the reasons lie less in the acquisition costs. The researcher estimates that around 30,000 euros would be needed to purchase one of the specialized drones. The necessary flight permits, costs for training and know-how of the pilots were particularly important, as the devices are significantly larger and heavier than the hobby drones that are commonly sold.
In addition, there is a need for permanently qualified personnel who would at least have to supervise the flights. In addition, municipalities are more likely to invest in classic equipment such as fire engines than in new systems, the benefits of which are difficult to quantify in individual cases.
An exception is a cross-border Bavarian-Czech project: As part of EU Interreg funding, the use of automated drones and artificial intelligence for early forest fire detection is currently being tested in the neighboring districts of Cham and Pilsen in the Bavarian and Bohemian Forest.
