FIRE-VLM: A Vision-Language-Driven Reinforcement Learning Framework for UAV Wildfire Tracking in a Physics-Grounded Fire Digital Twin

Chris Webb, Mobin Habibpour, Mayamin Hamid Raha, Ali Reza Tavakkoli, Janice Coen, Fatemeh Afghah · 2026

Read FIRE-VLM: A Vision-Language-Driven Reinforcement Learning Framework for UAV Wildfire Tracking in a Physics-Grounded Fire Digital Twin by Chris Webb, Mobin Habibpour, Mayamin Hamid Raha, Ali Reza Tavakkoli, Janice Coen, Fatemeh Afghah (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-03449.txt.

Description

Chris Webb and five co-authors build a training world for drones that track wildfires, because real fire data is scarce and dangerous to gather. Their system, FIRE-VLM, learns inside a digital twin, a detailed computer copy of real terrain and fire behavior. The twin uses federal elevation and fuel maps plus fire spread solvers that capture wind, smoke, and fuel use.

Inside that world, a learning agent flies a drone with two camera views. A vision-language model scores how well the view matches fire and smoke, and those scores shape the agent's rewards. The setup links map data to simulation in one pipeline. It offers a safer way to train fire-tracking drones before real flights.

Catalog record

CREATORChris Webb, Mobin Habibpour, Mayamin Hamid Raha, Ali Reza Tavakkoli, Janice Coen, Fatemeh Afghah
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE48 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2601-03449

Browse the full library · About the Tower of Babel