Researchers Ercan Erkalkan, Vedat Topuz, and Ayça Ak from the Department of Computer Engineering at Middle East Technical University in Ankara, Turkey, have developed a novel method for tracking wildfire perimeters using small, lightweight unmanned aerial vehicles (UAVs), or drones. Their work, published in the journal Sensors, focuses on improving the efficiency and reliability of wildfire monitoring in challenging environments with limited communication bandwidth.
The team’s approach combines thermal and RGB (color) imaging to create a more accurate and efficient perimeter tracking system for micro UAVs. Thermal images are used to identify hot regions, which are then refined using adaptive thresholding and morphological techniques. Meanwhile, RGB images provide edge cues and help suppress false detections caused by texture in the environment. By merging these two data sources, the system can better identify and track the boundaries of wildfires.
One of the key challenges in using UAVs for wildfire monitoring is the limited bandwidth available for communication. To address this, the researchers developed a rule-based merging strategy that selects boundary candidates and simplifies them using the Ramer-Douglas-Peucker algorithm. This approach reduces the amount of data that needs to be transmitted, making it more suitable for use in environments with limited communication capabilities.
The system also incorporates periodic beacons and an inertial feedback loop to maintain trajectory stability, even in the presence of GPS degradation. This is particularly important in wildfire environments, where GPS signals can be disrupted by smoke and other atmospheric conditions. The guidance loop is designed to achieve sub-50 ms latency on embedded System on Chip (SoC) platforms, ensuring that the UAVs can respond quickly to changes in the environment.
The researchers conducted small-scale simulations to evaluate the performance of their system. They found that it reduced average path length and boundary jitter compared to a pure edge tracking baseline, while maintaining environmental coverage. Battery consumption and computational utilization were also within feasible limits, allowing for forward motion speeds of 10-15 m/s on standard micro platforms.
This approach has significant practical applications for the energy sector, particularly in the context of wildfire management and prevention. By providing more accurate and efficient perimeter tracking, it can help energy companies to better monitor and manage wildfires in their operational areas. This can reduce the risk of wildfires causing damage to energy infrastructure, such as power lines and transmission towers, and help to ensure the safe and reliable delivery of energy to consumers.
In summary, the researchers from Middle East Technical University have developed a novel method for tracking wildfire perimeters using micro UAVs. Their approach combines thermal and RGB imaging, along with a rule-based merging strategy, to provide more accurate and efficient perimeter tracking in environments with limited communication bandwidth. This technology has significant potential applications for the energy sector, particularly in the context of wildfire management and prevention.
This article is based on research available at arXiv.