FireMan: Unmanned aerial systems based solutions for real-time management of wildfires

Wildfires are one of the major global environmental threats posed by climate change. The objective of the FireMan consortium is to develop novel, disruptive AI-based technology for the fast-paced detection of wildfires and for creating situational awareness during wildfire events using autonomous unmanned aerial systems (UASs, drones). The research objectives are related to two major areas “Autonomous flying” and “Situational awareness and decision support.”
Wildfire

Table of contents

Project duration
-
Core fields of research
Basic natural phenomena and mathematical thinking
Research areas
Spectral Imaging Laboratory
Computational Science
Co-operation
Finnish Geospatial Research Institute (FGI) in the National Land Survey of Finland, University of Oulu/CWC - Networks and Systems and VTT Technical Research Centre of Finland Ltd
Faculty
Faculty of Information Technology
Funding
Research Council of Finland

Project description

FireMan objectives and consortium

Wildfires are one of the major global environmental threats posed by climate change. The objective of the FireMan consortium is to develop novel, disruptive AI-based technology for the fast detection of wildfires and for creating situational awareness during wildfire events using unmanned aerial systems (UASs, drones). The research will consider aspects of autonomous flying using visual beyond line-of-sight drones and drone swarms, connectivity, and autonomous extraction of situational awareness using remote sensing and develop a DigitalTwin based decision support system for wildfire management.

The multidisciplinary consortium comprises researchers from the Finnish Geospatial Research Institute, Universities of Jyväskylä and Oulu and VTT and an extensive collaboration network. FireMan will create scientific breakthroughs and societal impacts by developing digital and low-emission technologies that will greatly support the objectives of adapting and mitigating climate change.

Research at the Ģֱ

Our researchers are developing methods for the early detection of wildfires, suitable for drone remote sensing, focusing on large areas beyond visual line of sight (BVLOS) techniques and high-altitude and large area coverages.  The methods are developed using data collected in the project's extensive experiments and Internet sources. We are currently testing different machine learning frameworks, e.g., YOLO, and pre-processing Boreal Forest data to be released as annotated data sets in the future. Our second objective is integrating fire spread modelling and AI to develop efficient tools for predicting fire spread.