AI-Powered Drone from Yugra Can Detect Fires Burning Underground
Neironiks, a resident company of the Yugra Technopark, has developed an intelligent system for the early detection of peat fires. The high-tech monitoring platform combines artificial intelligence, thermal imaging, and unmanned aerial vehicles to help identify hidden wildfire threats before they escalate.
Unlike crown forest fires, peat burns beneath the surface, and the process can continue for years. These underground hotspots have earned the nickname "zombie fires" because they survive beneath snow through the winter before reigniting in spring, threatening nearby communities as well as oil pipelines, gas pipelines, power transmission lines, and other critical infrastructure.
Catching "Zombie Fires"
Conventional monitoring methods fail to detect up to 40% of peat fires in Yugra during their early stages. At the same time, more than 70% of the region's licensed hydrocarbon production areas are located on peatlands. Underground fires destroy forests, fill the air with toxic smoke, and endanger pipelines, power transmission infrastructure, and other industrial assets. That gives the new system a dual mission: protecting both the environment and critical infrastructure.
The platform's biggest advantage is its ability to detect heat beneath the ground surface or dense vegetation. The drone carries a high-sensitivity thermal imaging camera and an onboard neural network that processes data during flight. The system continues operating even where reliable cellular coverage is unavailable, making it well suited for Yugra's remote territories. It can detect hidden thermal anomalies as small as 30 centimeters, allowing it to identify smoldering peat at the earliest stage. The coordinates are transmitted directly to dispatchers, while any data collected during temporary signal loss remains stored onboard for later analysis.
Moving From Response to Prevention
Artificial intelligence is playing an increasingly important role in forest protection. Rosleskhoz (Federal Forestry Agency) considers AI a promising tool for wildfire forecasting and has emphasized the need to shift from responding to fires after they begin toward predicting where they are likely to emerge. Neironiks' project fits naturally into that broader digital ecosystem. In 2024, Russian engineers equipped drones with AI-powered cameras capable of detecting smoke. Systems developed by Innopolis University and Delidron successfully identified both white and black smoke while surveying up to 200,000 hectares during a single flight. Those platforms, however, detected visible signs of fire. The Yugra system is designed to locate hidden underground combustion before smoke appears.
Two years ago, Yugra began using drones for wildfire surveillance. They now serve as the region's primary tool for identifying new fire outbreaks at an early stage. As a result, the number of wildfires recorded last year fell by one and a half times, while the total burned area was reduced by nearly half.
Yugra already operates the Lesookhranitel (Forest Guardian) hardware and software platform with fixed monitoring cameras, alongside satellite-based forest surveillance. In practice, wildfire prevention could become even more effective through a unified monitoring network that combines data from permanent observation stations with information collected by mobile drones.
Protecting Nature and Saving Lives
According to Rosleskhoz, as of June 29, 2026, firefighters across Russia were battling 218 active forest fires covering more than 277,000 hectares. Every additional day of firefighting requires substantial financial resources and personnel. Detecting hidden underground hotspots early could reduce the cost of suppressing large wildfires many times over while protecting both public health and forest ecosystems.
Peat fires are a challenge for many Russian regions. The new platform is expected to be used not only by government forestry agencies and EMERCOM, but also by major industrial companies, pipeline operators, and electric grid operators responsible for protecting critical infrastructure.
