Australia’s Ai-search prototype enters second development phase

22 April 2020 (Last Updated April 22nd, 2020 16:28)

The Australian Defence Force’s artificial intelligence (AI) prototype to transform airborne search and rescue, known as Ai-Search, has entered the second phase of development.

Australia’s Ai-search prototype enters second development phase
Warfare Innovation Navy conduct AI- search and rescue training mission off the coast of Stradbroke Island on board a C-27J Spartan. Credit: CPL Jessica de Rouw/ © Commonwealth of Australia, Department of Defence.

The Australian Defence Force’s artificial intelligence (AI) prototype to transform airborne search and rescue, known as Ai-Search, has entered the second phase of development.

The prototype is the Royal Australian Air Force’s (RAAF) Plan Jericho collaboration with Warfare Innovation Navy Branch and Air Mobility Group’s 35 Squadron and the University of Tasmania’s Australian Maritime College.

The project’s future phases will see the involvement of more partners, including Defence Science and Technology.

Plan Jericho AI lead wing commander Michael Gan said: “The next phase will involve testing different sensor and processor combinations in a range of environmental conditions, with the potential of testing on a range of aircraft, including UAS.”

Combined with a sensor and processor, the portable system has the capability to transform any aircraft, vehicles or vessels into a search and rescue platform.

Earlier this month, trials were conducted to test the potential of the AI system to save lives at sea.

The algorithm was recently evaluated to check its ability to identify a life raft and other waterborne vessels.

The evaluation was done by a C-27J Spartan sortie from RAAF Base Amberley, with the assistance of the Australian Volunteer Coast Guard.

Warfare Innovation Navy Branch machine learning expert lieutenant Harry Hubbert said: “During the sortie, we had a few GoPro sensors rigged up to detect a life raft and two algorithmic approaches working together to increase accuracy and the likelihood of a detection.”

“This sortie was pretty challenging as the life raft was upside down, making it harder to see for both the human eye and the Ai-Search sensors.”

Hubbert added: “The sensors are trained to detect an orange top, rather than a black top, but the Ai-Search still had a 70% detection rate, compared to the human detection rate of around 50%.

“The 30% Ai-Search non-detections happened when there was low contrast between dark water and the black underside of the life raft, and the good news is that we had no false positives.”