UMD-led Team Selected for DARPA Triage Challenge

Dinesh Manocha, a co-PI of the challenge, contributes expertise in perception-based machine learning and multi-agent coordination.
Descriptive image for UMD-led Team Selected for DARPA Triage Challenge

A multi-institutional team led by the University of Maryland (UMD) has been selected for the DARPA Triage Challenge, in which participants compete to develop novel methods of detecting injuries, particularly in mass casualty incidents, so that medical personnel can respond more quickly, efficiently, and precisely.

Dubbed RoboScout DTC, the UMD team will be part of the Systems Competition, one of three competitions in the Triage Challenge, and will focus its efforts on primary triage, in which medical personnel seek to identify and treat those most urgently in need of care.

In mass casualty incidents, whether civilian or military, triage needs often outstrip resources and place enormous strains on medical personnel. DARPA hopes to ease this burden by equipping uncrewed aerial vehicles (UAV), or drones, with stand-off sensors—that is, sensors capable of detecting subtle changes in heat or energy from a distance—and feeding the data to machines trained with perceptual algorithms.

RoboScout DTC is bringing together experts on robotics and automation, UAV ops, machine learning, and medical sensing to help make this vision a reality. One of only three university teams selected for funding by DARPA, it is led by Derek Paley, who is the Willis H. Young. Jr. professor of aerospace engineering education at UMD and also directs the Maryland Robotics Center (MRC), a research hub housed in the Institute for Systems Research.

UMD Distinguished University Professor of Computer Science Dinesh Manocha is also a co-PI, bringing his long-standing expertise on perception-based machine learning and multi-agent coordination to the endeavor. Manocha will train perception algorithms to integrate data from multiple stand-off sensor streams to generate assessment of injuries in real-time. 

Click HERE to read the full article 

The Department welcomes comments, suggestions and corrections.  Send email to editor [-at-] cs [dot] umd [dot] edu.