SRC Inc., in collaboration with the Air Force Research Lab (AFRL), and General Atomics Aeronautical Systems, Inc., successfully demonstrated the Agile Condor® high-performance embedded computing (HPEC) system’s onboard, autonomous, real-time processing capabilities in a recent flight test.
SRC together with AFRL, developed the Agile Condor system to be a scalable, flexible, low cost, size, weight and power (low-CSWaP) hardware architecture capable of, among other things, enabling machine learning for upstream data processing at the sensor.
During the flight test, the Agile Condor system was rapidly integrated on an MQ-9 remotely piloted aircraft (RPA) and leveraged a variety of machine learning algorithms to analyze gigabytes of synthetic aperture radar (SAR) and electro-optical/infrared (EO/IR) full motion video data in real-time. While in flight, the system autonomously detected, tracked and classified urban convoys and was able to quickly identify targets of interest, passing on actionable intelligence to the operator.
“We are proud to be supporting the Air Force with the advanced machine learning capabilities of the Agile Condor system,” said Kevin Hair, president and CEO of SRC, Inc. “Bringing advanced processing power to the edge will be critical to providing our warfighters with the actionable intelligence they need to remain safe and complete their missions.”
During the test, the Agile Condor system’s on-board processing resulted in providing a significant reduction in the amount of data needed to deliver actionable intelligence to a ground operator, freeing up limited ground-link bandwidth for other uses. The autonomous operation of the system at-the-edge allows for mission continuation in contested environments, even when communications with the ground station are lost. It also reduces the mental load on operators and provides enhanced, near real-time, situational awareness to support rapid decision making.
Commentaires