How does a gaming laptop turn overnight drone flights into morning fire maps?

The problem was not the drones

Thermal drones have flown over wildfires for years. Pilots spot heat the ground crews cannot see.

The bottleneck arrived after landing

Each pilot used different software, different file names, different formats. During one incident this year the system encountered 34 different versions of the same fire name. Data sat on memory cards or laptops until someone manually stitched it together. Eli Polsky, a wildland firefighter since 2012 and drone pilot since 2019, watched this happen repeatedly. He founded the UAS Wildland Fire Collaborative in May 2026 to fix it. A pilot uploads raw thermal images after a night flight. The automated pipeline identifies the files, matches them to the correct incident, and produces a standardized map set. Outputs include topographic sheets and aerial sheets. Google Earth files and GIS‑compatible datasets are also generated. Each map shows the heat perimeter, individual hot spots, areas of intense heat, and possible heat signatures that warrant closer inspection. The pilot reviews and approves before distribution. During the 2026 season the median time from upload to final approval was 18 minutes. That includes the pilot's own review.

Why the Legion hardware matters

Polsky built and tested the entire pipeline on a Lenovo Legion 5. The NVIDIA graphics card drove Agisoft Metashape, which stitches hundreds of thermal images into a geographic mosaic, and ArcGIS Pro, which handles GIS analysis and map layout. When the original card failed, a large ArcGIS Pro project took roughly two and a half minutes just to open on backup graphics. Cloud processing now handles stitching for most users. 195 thermal photographs from an October flight over West Fork Evans Creek became a finished map in about six minutes. Local development, LiDAR point clouds, and template testing still demand a capable workstation. Lenovo supplied a Legion Pro 7i with an Intel Core Ultra 9 processor and NVIDIA GeForce RTX 5070 Ti to restore that local speed.

Heat outside the perimeter

On July 21, 2026, Polsky flew two divisions of the East Evans Creek Fire in Jackson County, Oregon, shortly after midnight. The official daily perimeter showed one boundary. The thermal imagery revealed heat outside it. A log still burning under ash registered as high heat. Over the next three weeks the system produced 37 map sets across 15 nights for that fire. The busiest single map documented 3,327 individual heat sources. Mop‑up buffers grew from 100 feet to 400 feet as crews worked deeper. Weeks later at Woodrat Mountain a final survey showed almost no heat — just five possible sources. Polsky walked the area at 3 a.m. with the map on his phone. He found glowing embers at the bottom of a stump hole and faint heat deep inside another marked location. The system keeps "possible heat" on the map even when the signature is uncertain. Final interpretation stays with the people on the ground.

Adoption and archives

As of October 21, 2026, the public archive held 894 files representing 79 nights of delivered maps across 12 wildfires. Pilots from the Oregon Department of Forestry and its South Cascade District have used the system. Hummingbird Drones and Polsky's company Rogue Reconnaissance have also used it. Crewed‑aircraft data from ODF has been processed. The agency helped shape the symbology: drone‑observed heat appears in pink, other sources in red. ODF provided a letter of support recognizing the potential value of standardized UAS processing. The Collaborative can document maps delivered, fires supported, and processing speed. It does not yet have reliable data on firefighter hours saved or how operational decisions changed. Those outcomes remain unmeasured.

Off‑season work and the template problem

The new Legion Pro 7i will also handle off‑season mitigation mapping: drone‑based LiDAR and multispectral surveys to study vegetation, identify dead trees, and assess fuel conditions near communities. Those projects generate large point‑cloud datasets. Recent software changes let pilots view thermal imagery at full resolution without smoothing that can obscure detail. Each agency needs map templates that match its operational requirements and branding. Templates are built and tested in desktop GIS before entering the automated pipeline. A capable local workstation is essential for that development. Polsky's long‑term goal is a single free, standardized mapping system available to every wildland fire drone pilot. A pilot finishes a flight, uploads data, reviews maps, and makes them available before the morning briefing. Crews should not have to learn a different format depending on who flew the division. Expanding the Collaborative will require additional technical expertise and resources. The computing hardware helps make that possible.