The Hottest Technology in AI

There are approximately half a million building fires in the United States every year according to the National Fire Protection Association. Tragically, these blazes are responsible for over 3,600 deaths, 15,000 injuries, and over eleven billion dollars in property damage. The first priority of rescue crews during building fires is to get people who are either trapped or incapacitated out of harm’s way before it is too late. This can be quite a challenge in the heat of the moment, especially when having to contend with darkness and thick smoke. Fires often move very fast, and time spent searching a building without specific information about where to look is not an efficient method of finding those in need of rescue.

Providing firefighters with better information is one way that the loss of life associated with fires could be reduced. Help may be on the way on this front thanks to a prototype AI-powered firefighter helmet recently developed by a team of researchers at the University of Edinburgh. This helmet utilizes cutting-edge technology to give rescuers the ability to see through darkness and smoke to quickly locate people. Chris Xiaoxuan Lu from the research team said that their goal was to give firefighters a “superhero ability” to help them in the difficult work that they do.

The device is built on top of a standard-issue firefighter’s helmet. Mounted on the helmet are a slew of affordable off-the-shelf sensors and processing units to give rescuers the “superhero” abilities that the team was working towards. Among the sensors are thermal imaging cameras, inertial sensors, and a radar unit, all of which can see through smoke and darkness to give firefighters a clear view of what is otherwise invisible to the naked eye. Sensor data is fed into a machine learning algorithm that has been trained to recognize people, which can draw the attention of a rescuer to exactly where anyone in need of rescue may be before they even have time to scan the room.

Field trials of the smart helmet were conducted with the Scottish Fire and Rescue Service, where it was noted that the onboard technology gave rescuers a distinct advantage and allowed them to locate people very quickly. Watch Commander Glen Macaffer noted, “We can scan a room a lot better. We can take five to ten seconds compared to probably a couple of minutes when we wouldn’t have that technology. So for us to have that would be a massive game changer.”

Weighing in at a bit over two pounds, the helmet offers many advantages to wearers without being cumbersome. As they continue to test their system in hopes that it could one day be used in real-world rescue situations, the team is also working to improve their AI-powered helmet. Presently they are working towards building 3D maps of the structure to help with navigation, and they would also like to include built-in display screens. Based on how well this helmet was received by those that would use them, we hope that this technology can be rapidly refined for adoption in fire departments around the world.