vhutchison

NFPA and University of New Mexico explore use cases for sensors that will improve fire ground safety

Blog Post created by vhutchison Employee on May 7, 2019


Despite advances in various areas of science and technology, the act of firefighting is inherently dangerous. Over the past 10 years, an average of 81 U.S. firefighters have died each year in the line of duty, with nearly 30,000 injured during that span.

 

Dangers on the fire ground are a constant challenge. After being notified of an incident and dispatched, the incident commander and emergency responders make rapid fire ground decisions, based on the best information that they have at the time. They work to control the fire and ensure the life safety of building occupants; however, the tactics and strategies implemented rely on the following:

 

  1. The experience and judgement of the incident commander; 
  2. available, current, and accurate data (visually observed or determined from the operating environment); 
  3. and the standard operating procedures/guidelines of the responding fire department.

 

Situational awareness on the fire ground is paramount. Since the conditions of the fire scene are continually changing, it is common to lack some of the critical information that is needed to make optimal decisions about the stability of the structure, the health and status of firefighters, the location of victims, changing conditions on the fire scene, etc.

But what if emergency responders could be better informed via more accurate data? Can we improve firefighter safety by leveraging data captured thought sensor technologies?

 

To address this question, a research project led by the University of New Mexico in collaboration with the Fire Protection Research Foundation, the research affiliate of NFPA, is exploring novel use cases for sensors that will improve the safety of firefighters on the fire ground. Funding for this effort is through a multi-year grant from the National Science Foundation.

 

The goal of this research project is to make fundamental technical and algorithmic advances courtesy of connected and smart fire fighting technology. The proposed system will augment existing systems used by first responders by adding hardware and software components to the fire fighters’ existing equipment. This initiative will provide predictive modeling capabilities to support incident command evaluation of best approaches, based on experience and available resources.

 

This project addresses the following five key topic areas:

 

  1. Fire ground PAN/LAN Data Communication System. Establishment of a practical Personal-Area Network (PAN) using a PPE Sensor Network, and a Local-Area-Network (LAN) involving a Fire ground Local Area Data Communication System. The backbone of this project will consist of a mesh structure for communications that, based on the experimental approach, will exist in Wifi communications and can be extended to other communication methods. This will provide an important baseline structure that supports other key topic areas.
  2. Fire Ground Sound Discrimination. The capture and identification of critical fire ground sounds (e.g., PASS device or “Mayday”), with discrimination and filtering of these sounds from other fire ground noise will be considered. Algorithms will help support machine learning and help to implement specific fire ground actions.
  3. Prediction of Firefighter Exhaustion. Speech features will be included, identified, captured, and processed through the central computer in order to determine the level of stress and exhaustion of firefighters. This combined with respiration estimation procedures, will be used for actionable measures, such as assessing the remaining quantities of SCBA air or to supplement other physiological indicators.
  4. Human/Object/Event Recognition with Thermal Imaging. Algorithms will identify specific target entities using thermal imaging. With support from machine learning, the recognized objects will be transformed into knowledge-based actions for firefighters.
  5. Navigational Image Search Techniques. The imaging techniques within supported by machine learning will also be adapted to support firefighter locator navigation. This will ultimately benefit key fire ground activities dependent on locator technology such as search and rescue or RIT.

 

The proposed smart mesh communications structure combined with situational awareness will provide enhanced location and search capabilities. The communications backbone, in addition to the voice channel, will be enhanced and extended to enable increased data flow from various sensors collected locally but not yet fully integrated into the command infrastructure.

 

This project is on schedule to be completed in 2019, and we are looking forward to enabling new technology that supports firefighter situational awareness and ultimately improves the safety of our first responders.

Outcomes