A U.S. Air Force C-17 Globemaster III from Joint Base Charleston deploys flares as part of a training event over the Atlantic Ocean in a military operating area (MOA) outside Charleston, South Carolina on June 5, 2021 (U.S. Air Force Photo)
By Frank Wolfe |
1 year ago |
01/31/2023
The Air Force Research Laboratory’s (AFRL) information directorate in Rome, N.Y., is seeking the online submission of new, artificial intelligence/machine learning algorithms to speed the planning and execution of airlift operations.
The directorate is the nerve center for Air Force research and development on command and control, commmunications, computers, and intelligence.
The AFRL Airlift Challenge began on Jan. 23 and will run through Feb. 27.
Andre Beckus, an AFRL machine learning researcher, said in an AFRL statement last month that “a large demand and tight deadlines make airlift operations difficult to plan even under ideal conditions.”
“Unexpected disruptions only further complicate the problem, potentially introducing major delays and stressing planning software to its limits,” he said.
Such “unexpected disruptions” were a feature of the U.S. airlift operation to withdraw its troops and Afghan civilians from Afghanistan in August 2021.
“To identify new solutions, AFRL is inviting international researchers to apply their skills and develop unprecedented, new algorithms that achieve on-time deliveries while improving efficiencies,” the lab said last month regarding its Airlift Challenge. “Students are also invited to participate and showcase their talents.”
Beckus said that the Airlift Challenge “provides a simulation environment in which AI agents can interact.”
“Agents will be tested against a series of scenarios of ever-increasing complexity with evaluation ending when a time limit is reached or when the agent misses too many deliveries,” he said.
AFRL is to choose three Airlift Challenge winners to present a joint paper at the Society of Photographic Instrumentation Engineers (SPIE) defense and commercial sensing conference in Orlando on Apr. 30-May 4.
About a month after the U.S. withdrawal from Afghanistan in 2021, the Air Force’s Air Mobilty Command issued a request for information on automated, beyond line of sight (BLOS) communications for its fleet of aircraft to fit into the Pentagon’s planned Joint All Domain Command and Control (JADC2) architecture (Defense Daily, Oct. 4, 2021).

AMC is “seeking information from interested companies to provide an operationally effective, affordable, resilient, and secure Beyond Line of Sight (BLOS) Integrator/Manager/Controller [IMCs] for AMC’s fleet of C-5, C-17, C-130, KC-135, and KC-46 aircraft,” AMC said then in the RFI.

AMC envisioned the possible delivery of 900 IMCs for the airlift and tanker fleets.

“Within the framework of JADC2, the need for full-spectrum, seamless and resilient communications and adaptable data pathways out to the tactical edge is critical,” the RFI said. “AMC intends to develop an automated digital PACE (Primary, Alternate, Contingency, and Emergency) plan for BLOS communications for each platform to achieve resiliency and adaptability necessary to remain relevant across the spectrum of competition.  This capability will support roaming between service providers, which may require handoffs between satellites, gateways, and networks to provide rapid recovery of service, rapid (re)assignment of resources in the event of a surge or reduction of capacity.”

AMC cited the example of a Boeing [BA] C-17 using one PACE profile with the minimum required bandwidth, then switching to a new PACE profile with much higher bandwidth to perform an aeromedical evacuation.

“The same BLOS PACE profile will also enable a variety of tactical missions across the spectrum of competition, from Humanitarian Assistance to Palletized Munitions,” AMC said at the time.