The more information a firefighter has on a situation, the better they'll be as a first responder. Unfortunately, observing and processing that information quickly is a daunting task, especially in a high pressure scenario. NASA's Jet Propulsion Lab (JPL) and the Department of Homeland Security (DHS) are working together to take some of the load off their shoulders, by developing a new machine learning system that collects data on the emergency environment, and feeds relevant information and recommendations to the firefighting team.
Equipment to help first responders safely and effectively save lives and minimize damage is constantly improving, like devices designed to improve navigation in smoke-filled buildings through vibrations or thermal imaging. Created as part of the DHS's Next Generation First Responders (NGFR) program, JPL's artificial intelligence (AI) system was designed to coordinate those kinds of technologies.
NEW ATLAS NEEDS YOUR SUPPORT
Upgrade to a Plus subscription today, and read the site without ads.
It's just US$19 a year.UPGRADE NOW
The Assistant for Understanding Data through Reasoning, Extraction and sYnthesis, or AUDREY, is much more than a clunky acronym. Integrated with the Internet of Things (IoT), the AI is designed to connect with wearable devices and sensors in the responders' uniforms, analyze the data and send relevant information to individual members of the team to expand their situational awareness and aid communication.
Locations of each responder can be tracked through GPS, or other devices like the digital positioning shoe, sensors in the uniforms could track the concentration of dangerous chemicals and gases in the air and the temperature in different parts of the building, while satellite images of a location could help the team find their way around. And all of this information can be relayed to the responders through mobile devices or heads-up displays, similar to the SINTEF jacket.
"When first responders are connected to all these sensors, the AUDREY agent becomes their guardian angel," says Edward Chow, the research's program manager. "Because of all this data the sensor sees, firefighters won't run into the next room where the floor will collapse."
For that to work, massive amounts of data need to be processed in real time, a task that no human is up to, but as a cloud-based, machine learning system, AUDREY can handle it, specializing which information it passes onto individual team members based on their role. The system also learns over time to make its future predictions and suggestions more accurate.
"Most AI projects are rule-based – if this, then that," says Chow. "But what if you're only getting part of the information? We use complex reasoning to simulate how humans think. That allows us to provide more useful info to firefighters than a traditional AI system."
The result of several technologies developed by NASA and the Department of Defense, AUDREY has been in development for around nine months taking advantage of several technologies that have been in development for the past five years. Its ability to make recommendations was tested in a simulator in June, where AUDREY was given data from a variety of sensors, asked to make safety recommendations and then transmit those to a mobile device. The team believes the system will be ready for field demonstrations within the next year.
Source: JPL NASAView gallery - 2 images