Team Chiron Competes in DARPA Triage Challenge Finals

Using robots, drones, sensors and advanced artificial intelligence techniques, Carnegie Mellon University’sTeam Chironhopes to help emergency responders save more lives and treat more patients during disasters like earthquakes or industrial accidents. The team, a collaboration among CMU’s Kenneth C. Griffin School of Computer Science, the University of Pittsburgh and Lockheed Martin, is developing autonomous robotic systems that can locate injured people, assess their conditions and help determine who needs care most urgently during a disaster. Team Chiron’s system employs drones that fly above the scene and use cameras to find patients. They then send initial reports back to ground robots that can deploy to gather more information. “In a mass casualty event, the hardest question a medic faces is, ‘Who do I treat first?’ and today they answer it on foot, one patient at a time, with almost no information,” said Kimberly Elenberg, a principal project scientist in CMU’s Robotics Institute (RI). “Finding the injured and reading their vital signs from a distance is only the start. The goal is to turn everything the robots see into a picture of who needs care first, so medics can make better, faster decisions in the minutes that matter most.” Team Chiron is participating in the Defense Advanced Research Projects Agency’s (DARPA)Triage Challenge, a competition spanning three years designed to develop and test research in robotics, AI and sensing to help emergency responders. The competition’s finals started on Nov. 5 and will run through Nov. 13 at Disaster City, a disaster response training facility near College Station, Texas. In a mass casualty incident, such as an earthquake or major accident, emergency responders may have to assess dozens of injured people in a short amount of time. Team Chiron aims to help them work more efficiently by using robots and drones equipped with advanced algorithms and sensors to gather information about casualties without requiring physical contact. Team Chiron tests their technology at the Allegheny County Fire Academy. The team combines drones and ground robots with AI systems that analyze information collected from a distance. Sensors can measure physiological indicators, including heart and breathing rates, while cameras and other sensors help identify people, assess their conditions and locate them within a chaotic scene. The information can then be relayed to emergency responders to support critical triage decisions. “In a disaster, you rarely have a complete picture of what’s happening. Someone may be hidden from view, the lighting may be poor or a sensor may not give you a clear reading,” said Sebastian Scherer, an associate research professor in the RI. “The challenge is to pull together those pieces of information and make them useful to the people responding on the ground. That’s where combining robotics, sensing and AI can make a real difference.” Lockheed Martin members participated in the team’s experiments and are helping demonstrate the system during the finals, contributing their expertise in operating the high-altitude drones used to survey disaster scenes. These drones fly above the scene and use cameras to find patients and send initial reports back to ground robots that deploy to gather more information. The technology Team Chiron has developed is intended to support emergency medical personnel, not replace them. By surveying a scene and gathering preliminary information, autonomous systems could help responders identify where their attention is needed most, especially in dangerous or difficult-to-access environments. Team Chiron has advanced through the challenge’s first two events, each of which tested the technology under increasingly difficult conditions. Inevent 1, held in September 2024, the team used aerial and ground robots to navigate simulated disaster scenes and assess actors and mannequins representing injured people. The robots had to evaluate casualties from a distance, without touching them, and transmit their findings to a central command station. Forevent 2, in September 2025, the team further developed their autonomous systems to perform at night. To prepare, they tested their work at CMU’s Robotics Innovation Center in Hazelwood Green. Team members simulated casualties in a grassy field while quadruped robots searched for them in the dark. The robots assessed vitals such as heart rate, breathing rate and alertness, then relayed the information to a remote operator. Event 2 challenged the system to navigate uneven terrain, operate in low light and communicate reliably despite intermittent connections. The robots needed to do more than find people: They had to gather useful information about their conditions and deliver it to the humans making decisions. “Team Chiron is combining aerial drones, ground robots and AI to build a more complete picture of a disaster scene,” said John Galleotti, a senior systems scientist in the RI. “Each platform has different capabilities, and getting them to work together autonomously is a major engineering challenge. But that’s exactly what we’re working to solve. We’re uniting these systems to navigate difficult terrain, locate casualties, gather useful information and communicate it to human responders, even when conditions are far from ideal.” The team also brings a medical perspective to its work through its collaboration with Dr. Lenny Weiss, an associate professor of emergency medicine at the University of Pittsburgh, which helps connect the engineering research with the realities of assessing and treating patients during emergencies. Weiss is also the medical director of the City of Pittsburgh Department of Public Safety and the Allegheny County Emergency Services and 911 Communications. “I am so proud of the collaboration that we have put together between our universities, experts in science and medicine, and public safety. This is really something unique to Pittsburgh,” Weiss said. “The novel resources we are developing will make a significant impact in saving lives here at home and in any austere environment. I have a true dedication to and respect for all our first responders and military personnel, but we need ways to both keep them safe and enhance their capabilities. We are doing that right now. The future is here.” The Triage Challengefinalspush the technology even further, requiring teams to demonstrate how their systems perform incomplex conditions, including darkness, smoke, fire, varied terrain and rubble. For Team Chiron, the competition is an opportunity to show how advances in autonomous navigation, sensing and AI can work together in a practical emergency response setting. The team’s larger goal extends beyond the competition: advancing autonomous systems that could eventually help emergency responders assess casualties more quickly, make better-informed decisions and direct limited medical resources to the people who need them most. For More Information:Aaron Aupperlee | 412-268-9068 | aaupperlee@cmu.edu