Breadcrumb

  1. Home
  2. Research
  3. Challenges
  4. DARPA Triage Challenge
  5. Challenge Events
  6. Systems Competition | Event 2

Systems Competition | Event 2

Smoke, Darkness, and Obfuscation
 

This video contains graphic depictions of simulated injuries. Viewer discretion is advised.
In this immersive overview of Challenge Event 2, guided by DARPA Triage Challenge Program Manager retired Army Col. Jeremy C. Pamplin, M.D., you'll experience how teams of innovators, engineers, and DARPA are redefining the future of combat casualty care. Check out competition runs, behind-the-scenes of what it takes to put on a DARPA Challenge, and glimpses into the future of lifesaving care.
Be sure to look all around! Use your mouse to look up, down - even behind you! Open the video in a full window for best viewing.

 

Courses
 

Night Ambush

dtc-darpa-event2-systems-night.

LiDAR-equipped UGV exploring the Night Ambush course at Workshop 2 event. DARPA | Tom Shortridge

C130 Crash

dtc-darpa-event2-systems-ugv

Quadruped UGV surveying course for C130 crash at Workshop 2 event. DARPA | Paul Flacks

 

From Casualty Mapping to Responder-ready Tools

dtc-competitions-systems

In Challenge Event 2, teams were required to operate with full autonomy as well as accurately geolocate casualties. Degraded sensing, dynamic physiological changes and walking casualties will significantly added difficulty to the challenge. 

To win a prize, teams passed casualty localization and triage accuracy thresholds. Teams also developed a user interface designed for use by first responders.

 

Technical Challenge Elements

null
Degraded Sensing
The courses included elements that ranged from constrained passages to large fields, lighted areas to complete darkness, and wet to dusty conditions. Sensors had the dynamic range to operate reliably in these environments. Dust, fog, mist, smoke, talking, flashing light, hot spots, and gunshot and explosion sounds were within scope of this challenge element. Extreme temperatures, fire, tremors, and hazardous materials were not within scope.
 
null
Obscuring Obstacles
Casualties were fully visible to partially obscured to completely obscured, such as buried under a shallow layer of rubble. Sensor modalities capable of penetrating rubble had an advantage in such situations. Casualties were grouped with limbs overlapping, or may be interacting with live responders.
 
null
Terrain Obstacles
The competition scenarios were held in realistic environments that may include natural or human-made materials; structured or unstructured clutter; and intact or collapsed structures and debris. Robotic mobility was not the focus of this challenge and it was expected that widely available outdoor robotic platforms will be capable of navigating the competition environments. In some cases, more robust platforms benefitted from more direct navigation paths or closer approach paths to casualties.
 
null
Dynamic Obstacles
Live responders, ambulatory wounded, or other physical changes to the environment were e included to test the agility of the system autonomy to identify and assess casualties.
 
null
Dynamic Casualties
Some treatable injuries may rapidly become fatal, and delays in finding and assessing casualties may result in missing the window for effective LSI. While competitors were not expected to re-evaluate casualties for changes in status, casualties who were not evaluated within an appropriate timescale may have had a change in status (for example, progression of untreated hemorrhage or airway injury).
 
null
Endurance Limits
Individual scenarios ran <30 minutes. Teams were permitted to replace batteries during their run; teams considered the implications of returning to the original launch location and redeploying their systems Real-world systems used stand-off sensors on platforms to locate and triage casualties in the field.
 
Accordion item
Teams
Body

Qualified funded and unfunded teams participated in the competitions, striving to advance the use of autonomous systems for rapid, stand-off casualty assessment to facilitate timely and accurate medical triage.

  • AIR TAGS
  • Chiron
  • Coordinated Robotics
  • DART
  • PRONTO
  • RoboScout
  • Robotika
  • TTT
  • UAS-DTU

See all teams | Team qualifications

Evaluation criteria
Body

Casualty Report

FieldValuesScoring criteria
LocationLatitude, Longitude+5 if within d meters of ground truth (GT) casualty location
Severe HemorrhagePresent, Absent+4 if correct
–2 if incorrect
Respiratory Distress [t] +Present, Absent+4 if correct
–2 if incorrect 
Heart Rate Beats per minute (BPM)+1 if within 5 BPM of GT 
–0 otherwise 
Respiratory RateBreaths per minute (BrPM)+1 if within 3 BrPM of GT 
–0 otherwise
Trauma  Head: Normal, Wound, Not Testable (NT) 
Torso: [Normal, Wound, NT 
Upper Ext.: Normal, Wound, Amputation, NT
Lower Ext.: Normal, Wound, Amputation, NT
For each field: 
+1 if correct 
–1 if incorrect 
Alertness Ocular: Open, Closed, Not Testable (NT) Verbal: Normal, Abnormal, Absent, NT Motor: Normal, Abnormal, Absent, NTFor each field: +2 if correct –2 if incorrect


Golden Window

FieldValuesScoring criteria
Severe HemorrhagePresent, Absent+2 if correct
Respiratory DistressPresent, Absent+2 if correct
VitalsHeart Rate +/- 3BPM
Respiratory Rate +/- 5BrPM 
+1 if either is correct


Scoring Casualty Reassessment

FieldValuesScoring criteria
Location1Latitude, Longitude+5 if within d meters of ground truth (GT) casualty location
Severe Hemorrhage1Present, Absent+4 if correct
–2 if incorrect
Respiratory Distress1Present, Absent+4 if correct
–2 if incorrect 
Heart Rate1Beats per minute (BPM)+1 if within 5 BPM of GT 
–0 otherwise 
Respiratory Rate1Breaths per minute (BrPM)+1 if within 3 BrPM of GT 
–0 otherwise
Alertness1Ocular: Open, Closed, Not Testable (NT) 
Verbal: Normal, Abnormal, Absent, NT
For each field: 
+2 if correct
 Motor: Normal, Abnormal, Absent, NT–2 if incorrect

[1] Points only received if initial assessment was correct and ground truth at reassessment time has since changed.


Rapid Casualty Location Bonus

Bonus pointsTime of reporting
+30-5 minutes after run start
+25-10 minutes after run start
+110-15 minutes after run start
+015+ minutes after run start

 

Timeline
Body
Challenge Event 2
2025
Workshops
SystemsMarch 10
DataMarch 17
Competitions
Systems, DataSept. 28
Awards
CeremonyOct. 4
 Artist's concept: Using data captured by stand-off sensors onboard autonomous systems, uncrewed aerial and ground vehicles assess injuries during a mass casualty event to determine treatment priorities, allowing medical care professionals to quickly focus on the most urgent casualties.
Artist's concept: Using data captured by stand-off sensors onboard autonomous systems, uncrewed aerial and ground vehicles assess injuries during a mass casualty event to determine treatment priorities, allowing medical care professionals to quickly focus on the most urgent casualties. Source: DARPA | Jahyra Catala

Leaderboard

Prizes

Second

UAS-DTU
$150,000

First

Coordinated Robotics
$300,000

No other self-funded teams achieved the minimum benchmark (prizes are only for self-funded teams in year two).

Scoring

Run: Calculated as the sum of all correctly identified injury components minus penalties for wrong submissions.

Event: Sum of the two runs.

Results
Run
Event
DART*
728
536
192
RoboScout*
630
383
247
Coordinated Robotics
546
267
279
PRONTO*
483
321
162
UAS-DTU
387
109
278
Chiron*
357
176
181
AIR TAGS*
143
33
110
Robotika
36
36
TTT
0
0
 

C130 Attack

 

Night Ambush

*DARPA-Funded Team

Overview
 

Challenge Event 1

2024

Competitions
Systems  |  Data |  Virtual

Challenge Event 2

2025

Competitions
Systems  |  Data

Challenge Event 3

2026

Competitions
Systems  |  Data

 

DTC heartbeat right
DARPA Triage Challenge 
Resources  |   FAQs  |  Terms and Conditions  |  YouTube Playlist  |  Media Gallery  |  Contact

Contact