How Sensor Technology Can Help Fight Fires

From smartphones and smartwatches to airplanes and health monitoring devices, sensor technology is all around us. Now researchers are figuring out how to use sensors when disasters such as forest fires strike.

By John McColgan — Edited by Fir0002 — taken by John McColgan, employed as a fire behavior analyst at the Forest Service, an agency of the U.S. Department of Agriculture, Public Domain,

Over the last three decades, various factors have contributed towards the increase in the destruction caused by wildfires in parts of the Western USA and Australia. As a result, it is estimated that an additional 4.2 million hectares of forest were destroyed by fires in the Western USA during 1984–2015.

Data in real-time might save lives

Fighting forest fires is a large management operation that requires cooperation and seamless communication between dozens or even hundreds of people led typically by an incident manager, who is often continuously monitoring the situation from a remote location.

In scenarios like this and in general in disaster management, the challenge for the incident manager is to continuously monitor the environment as well as updates regarding the positions and health conditions of the firefighters in real-time, to take decisions and coordinate the operations in a fast and efficient manner.

Sensors are evolving rapidly

Sensor technology is continuously evolving, and with the reducing costs, 26 billion devices containing sensors (excluding PC’s, tablets, smartphones) are expected to be installed and used by 2020.

These sensor devices can be leveraged in situations where groups have to work together to solve complex tasks (i.e., disaster management scenarios or even when analyzing the tactical behavior of sports teams). However, obtaining useful, filtered information from the large volumes of continuous raw sensor data and keeping track of the whole situation is challenging as this requires a continuous real-time analysis of the raw sensor data.

The StreamTeam project

To tackle this challenge, researchers at the Databases and Information Systems (DBIS) research group at University of Basel, led by Prof. Heiko Schuldt, are working on a project called StreamTeam. The researchers aim at monitoring, analyzing, and visualizing actions of teams (e.g., tactical compliance, collaboration) in real-time in highly dynamic environments, especially in the applications mentioned above.

As part of his Master Thesis project, Marko Obradovic, who studied Computer Science at the University of Basel, worked together with Lukas Probst, PhD student at the DBIS group in the StreamTeam project, on improving the forest fire management scenario. His goal: to reduce the risk for firefighters and to speed up the decision-making process.

M.Sc. Marko Obradovic and Dr. Heiko Schuldt, Databases and Information Systems Group. (Image: Marko Obradovic)
“Developing systems for Real-time Data Analysis has become increasingly important due to the increased availability and usage of sensors. As there are more and more applications that provide analysis results in real-time, we require systems that can process data while it is continuously collected,” explains Marko.

Simulating forest fires

Since there was no suitable data already available for this study, Marko had to simulate the data of a real forest fire management scenario. So Marko developed a highly parameterizable forest fire simulator to get the necessary realistic sensor data sets for analysis.

The parameters included GPS, body temperature, the temperature of the surrounding area, wind speed, wind direction, etc. These factors were then used to compute different models about the spread of a forest fire, change in temperature and the behavior of the firefighters.

Example of a real-time scenario where the incident manager gets real-time updates of the positions of each firefighter, temperature and environment conditions of each zone. (Image: Marko Obradovic)

Marko worked on nine different forest fire fighting scenarios. For each scenario, he varied the most relevant parameters, to evaluate the quality of the simulated data. This simulator was a major part of his project and will be crucial for further work in this field.

“The simulator can also be used as a standalone software. For instance, it can be used to reconstruct past forest fires to provide a deeper understanding. Moreover, it can be used to simulate or test new extinction tactics on different sorts of forest fires”, comments Marko.

“Generating the data was the unique and demanding part of my work. In fact, it was the like the playground for my project, which I enjoyed the most,” says Marko.

Proof of concept

The other two important phases of the study included analyzing the data streams emitted by each sensor and visualizing the results of this analysis while receiving signal alerts.

To present a proof of concept for the Real-time Data Analysis system, each firefighter was shown to be equipped with three types of sensors: a position sensor, a sensor to track the health conditions of the firefighter, and sensor devices for monitoring the surrounding environment.

“Performing Complex Team Analyses means to jointly analyze sensor data streams of all firefighters and produce more meaningful information,” says Marko.
Snapshot of an ongoing forest fire fighting scenario simulation. (Image: Marko Obradovic)

The Complex Team Analyses, done with the Analyzer developed, includes a live position map to track the current position of firefighters in the forest area, a live heat map tracking most common positions of firefighters during the fire extinction procedure and a live environment temperature map, tracking the environment temperatures continuously.

A potential dangerous situation which can be detected by the incident manager using information relayed from the different sensors. (Image: Marko Obradovic)

The developed Real-time Data Stream Analysis System can improve incident management of real-world scenarios and could potentially help save forests and lives.

In their current work, the DBIS group aims at automatically detecting dedicated events in the data streams and thus anticipating potentially dangerous situations. An example of a potentially dangerous situation could be a sudden gust of wind, for instance, capable of spreading the fire in a particular direction. The challenge here is to track the positions of all firefighters and to get those in the danger zone out.

To show your support for this post and recommend it to your followers, click on the clap icon 👏 below. Each users is allowed to clap up to 50 times to show how much they appreciated a story.

The University of Basel has an international reputation of outstanding achievements in research and teaching. Founded in 1460, the University of Basel is the oldest university in Switzerland and has a history of success going back over 550 years. Learn more