The challenge
Mapping subsurface networks and unpredictable conditions
Complex underground settings present significant challenges for military and civilian first responders. Hazards vary drastically across domains, often degrading or changing over time, and are often too high-risk for personnel to enter.
DARPA Subterranean Challenge map rendering showing a representation of a tunnel , urban and cave environment. Features of DARPA applications are explained and include: Learn more at www.darpa.mil
To help resolve this challenge, 11 teams from around the world were invited to propose novel methods for tackling these time-critical scenarios, including CSIRO's Data61, the only Australian team.
Teams must develop and demonstrate physical systems that can function in three circuit events on physical, representative subterranean courses, and focus on advancing and evaluating novel physical solutions in realistic field environments.
Our response
Developing a team of autonomous robots
Our world-leading Robotics and Autonomous Systems group is one of seven teams to receive up to US$4.5 million in funding from DARPA to compete in the SubT Challenge. They are also one of 8 international teams to be selected for the final challenge, which will incorporate elements of the three previous circuits (tunnel, urban and cave).
The competition challenges teams to deploy an autonomous robotics team to successfully navigate and map complex environments, while identifying objects such as mannequins, backpacks, fire extinguishers and mobile phones throughout each course, analysing and communicating on each objects location and status to other vehicles in the team and humans at headquarters.
The CSIRO's Data61 team combines the expertise of our researchers, engineers and scientists with mapping and autonomy spinout Emesent, Australian robotics manufacturer BIAS, and US research partner Georgia Institute of Technology. Quadrapods, tracked vehicles, drones and hexapods are just so me of the robots on the team, which is adapted for each challenge.
Each bot is equipped with a perception pack called 'Wildcat', which enables a vehicle to map, analyse and navigate its surrounding environment without any human intervention. These packs also include a perception system ('Catpack') that provides real-time localisation and mapping data for autonomous operation and high-fidelity mapping.
The results
Team CSIRO's Data61 wins second place
Team CSIRO's Data61 beat teams from NASA JPL, MIT, California Institute of Technology, and Carnegie Mellon University to claim second place in the DARPA Subt Challenge.
Hosted in the Louisville Mega Cavern in Louisville, Kentucky, teams needed to rapidly map, navigate, and search complex underground environments such as human-made tunnel systems, urban undergrounds, and natural cave networks.
Leader of team CSIRO’s Data61 and CSIRO’s Robotics group leader, Dr Navinda Kottege said:
“This is an amazing result! We are the first Australian team to place in the top two at a DARPA robotics challenge.
“This cements CSIRO’s place as a world leader in robotics and puts Australia firmly on the map in this increasingly important area of science.
“I’d like to thank team partners Emesent and Georgia Institute of Technology for their exceptional research and development and contribution to this amazing result.”
The $US1 million ($AUD1.3) prize money will be reinvested into team CSIRO’s Data61's research and development of Australian technology.
The technology developed for the DARPA challenge has been applied to multiple scenarios across Australian industries, including agriculture, mining and defence.
An example of this would be the Wildcat perception system and the Early Adopters Program (EAP), makes low risk, affordable and customisable technologies available for Australian companies to trial and build upon, for use in potentially improving production efficiency and building new competitive advantages or markets.