January, 2024 – Forest fires are one of the phenomena we hear about more and more every summer. Due to climate change, they are occurring more and more abundantly all over the world and their consequences are extremely destructive.
In a new research project of the Ministry of Defense of the Slovak Republic, under the name FalcoGuard*, experts from YMS and Military Forests and Assets investigated the use of tethered drones with sensors for flexible monitoring of the risk of fires in currently threatened locations.
In Slovakia, the highest risk of forest fires is only for a few critical weeks of the year, but in different territories. “The risk depends on the meteorological situation, the moisture content of the “fuel” under the trees (stump, litter, dead wood, crevices, etc.), the number of consecutive sunny days, and, significantly, also the activity of people. Moreover, in the territories of military forests, the risk is increased, for example, even after a military exercise or other activity. Once a fire starts, the speed of detection and immediate intervention is critical, because in the forest, fire spreads at an extreme speed through the crowns of trees, on the ground, and through the root system,” says prof. B.Sc. Ing. Andrea Majlingová, MSc. PhD., external researcher of the project.
Due to this relatively short risk period, the dynamically changing conditions on the territories, the hilliness of the forest terrain, and the potentially high destructive effect of an outbreak of fire, other than traditional stationary detection methods were investigated. “General monitoring is usually done by physical inspections in the field, or special rotating cameras on masts (ASDS system). However, the stationary camera approach has its limitations. Among others, these are high costs for installations, operation, as well as low flexibility. These reasons led us to investigate more flexible approaches, specifically the potential of tethered drones with specific sensors,” explains Miroslav Čongrády, head of the IT Department of Military Forests and Assets, š.p.
In combination with special dispatching, drones prove to be a practical and usable model. “Drones can be placed independently in any area with a high fire risk. The tested models fly to a height of 100-150 m, which does not cause problems for flight and operational services, and they scan the territory from different angles. In practice, multiple drones would be deployed on site to have good, overlapping visibility. If they detect a fire from several angles, we can use mathematical calculations in the linked software to determine the exact location and intervene immediately, but even if the fire is detected by only one drone, the local foresters can determine the likely location of the fire from the calculated azimuth,” continues Miroslav Čongrády in describing the specifically investigated monitoring in terrain.
The model has a perspective not only in the rapid detection of fires, but also in prevention. “We have software at our disposal that can automatically process a large amount of data from forest monitoring, and we are aiming to be able to calculate the risk of fires in specific areas using the most modern procedures of machine learning, automatic image processing and artificial intelligence. In other words, to determine with high precision in advance locations where fires are a real threat and to dynamically adjust the intensity of autonomous monitoring,” adds Radovan Hilbert, director of business development at YMS, possibilities that experts are counting on for the future.
________________
This work was supported by the Ministry of Defense of the Slovak Republic.
* Researchers creatively derived the name of the FalcoGuard project from the flight and hunting of the falcon, which observes from the sky, or it guards (English to guard) the surface of the Earth and when it notices the prey – movement, it aims its sight and focuses on the selected target.
The purpose of the FalcoGuard project was to identify and verify the possibilities of building a system of mobile monitoring devices intended for the early detection of forest fires, which will consist of three basic components: carriers, sensor equipment, and a system for processing monitored data.
YMS regularly collaborates with foresters on technologically advanced research projects. Monitoring of lycophagous pests with the help of automatic analysis of data from satellites won the award of the Minister of Economy for the Innovative Act of 2020. We also successfully verified the procedure for cleaning infested soil using ecological methods with an algorithm for detecting the spread of toxic substances in space.