The Air Force Research Laboratory (AFRL) has awarded BAE Systems a $7.8m contract to develop machine learning software that will enable advanced situational awareness and automatic target recognition (ATR).

The tightly integrated machine learning software will be part of the Multi-Sensor Exploitation for Tactical Autonomy (META) programme.

Under the terms of the award, BAE Systems’ FAST Labs research and development organisation will provide Environmentally Adaptive Geospatial Learning and Exploitation, an innovative suite of machine learning and fusion algorithms.

The system integrates multiple elements of the company’s extensive autonomy portfolio to provide high confidence detection, tracking, identification, and intent understanding for critical mobile targets in contested environments, including targets under camouflage, concealment, and deception.

“With the addition of environmentally adaptive processing, this solution bridges a critical gap in machine learning,” said Mark Kolba, programme manager for BAE Systems’ FAST Labs. “Providing local situational awareness and human-like context of environmental conditions will greatly reduce false alarms.”

BAE Systems develops autonomous data exploitation capabilities for systems and sensors for potential use in US military operations across air and space domains.

The programme builds on the success of the company’s Multi-intelligence ATR for Geospatial Intelligence Capabilities technology that was successfully developed through AFRL’s Study of Emerging Exploitation Developments program.

www.baesystems.com