BAE Systems announced that its FAST Labs research and development organization has received a $4 million contract under Phase 1 of the Defense Advanced Research Projects Agency’s Artificial Intelligence Reinforcement, or AIR, program to develop autonomy solutions in air combat.
Under the contract, FAST Labs will develop machine learning-based simulation models of sensors, weapons and electronic warfare systems currently used in air combat missions, BAE said Tuesday.
The contract’s work also includes FAST Labs using ML to establish combat aerial physics for input to the processes required in the rapid development and delivery of future AIR software products. The work will be performed at BAE Systems facilities in Arlington, Virginia, and Burlington, Massachusetts.
Michael Planer, FAST Labs scientist and principal investigator, pointed to the vast data and complex testing cycles needed for air combat performance to be reliable and consistent. “Using machine learning, we will train the models used to make dynamic decisions – ensuring that the artificial intelligence pilot is tested and trusted by human pilots,” he said.
The AIR program, which DARPA launched with a proposal solicitation in November 2022, is divided into two technical areas: model development and multi-agent AI agent training. Each area has a Phase 1 as the base period and a Phase 2 option for system design, development and testing.
In June, the agency awarded EpiSys Science and PhysicsAI an 18-month, $6 million Phase 1 AIR contract to develop tactical algorithms for team-based air combat missions.
Other industry players under contract in the program include Lockheed Martin, which received a $4.6 million award in July, and Northrop Grumman, which secured a $28.9 million contract in February.