- November 18, 2026
- By Robert Herschbach
When a tornado hits a neighborhood or an adversary strikes on the battlefield, first responders can face a difficult imbalance: few medics and many casualties. A team of University of Maryland engineers, pilots and students that developed a network of drones and robot “dogs” to locate and assess the wounded in these situations won the top prize of $1.5 million in a rescue competition hosted by the Defense Advanced Research Projects Agency (DARPA).
In addition to Team RoboScout’s victory on Friday in the autonomous systems competition held at “Disaster City” in College Station, Texas, the Terps won an additional $12,500 award for their accurate detection of vital signs to conclude the three-year DARPA Triage Challenge, which encouraged innovation in helping emergency workers locate victims, assess injuries and prioritize lifesaving care.
Willis H. Young Professor Derek Paley credited RoboScout’s “culture of experimentation and field testing” with helping to give it the edge, noting that the team partnered with the U.S. Air Force’s Center for the Sustainment of Trauma and Readiness Skills to run several events that tested the system, including a mass casualty simulation at the Fairfax County (Va.) Urban Search and Rescue Training Facility earlier this year.
He also said that UMD and other university teams were able to “do more with less” during the Triage Challenge, giving them an advantage over the private companies and nonprofit research organizations such as Battelle, with its Drone-Assisted Rapid Triage Team, which RoboScout edged out from behind to secure the top spot.
The team’s efforts were put to the test in a series of simulated scenarios involving actors, manikins and trained medics. The scenarios included a military compound, the aftermath of a tornado in a residential area and an attack on critical infrastructure.
Team RoboScout employs a phalanx of drones and robotic dogs equipped with color and thermal cameras, radar and microphones to search for victims and identify conditions such as injury, blood loss, respiratory distress, trauma or amputation, while measuring heart rates, alertness, respiratory rate and other vitals. That information is then relayed to human first responders, with color-coded virtual “pushpins” appearing on their tablets to allow them to rush aid to those in greatest need of care.
RoboScout, Paley said, came with a technological edge because of its core IP, dubbed Triage AI.
“It's basically an inference engine for robotic assessment of patient status using both historical and live observational data,” he said. “The robots and drones are the current embodiment of this core IP. They are the mobile sensor platforms that feed the observations into our inference engine and from there we can make deductions about both observable and unobservable health states including triage status and evacuation priority.”
The dramatic win caps a highly successful year for teams led by Paley, with the Autonomous Micro Air Vehicle team having won a $500,000 prize for Most Revolutionary Aerodynamic Design at the August 2026 DARPA Lift Challenge in Dayton, Ohio.
In addition to student participants, the team drew faculty and personnel from the UMD Department of Aerospace Engineering, the University of Maryland School of Medicine in Baltimore and the UMD UAS Research and Operations Center (UROC), a Southern Maryland-based hub for designing and testing autonomous aircraft. Joshua Gaus, UROC’s lead UAS engineer, is a team member.
A key member of the RoboScout team was Rain Technology Solutions, which contributed the medic interface that provides situational awareness for first responders.
"We've decisively proven that human machine teams outperform every other approachable problem in some of the most rigorous, large scale, complex environments that you've ever seen," said DARPA Director Stephen Winchell. "Some of the teams perform really well on the identification of victims. Some of the teams perform really well on the triage and lifesaving or the evacuation pieces. So there's a lot that we can learn from each other in this community, and I just want to say thank you for coming out here."