AFRL Seeking ML, High Performance Computing Capabilities for Space Domain Awareness Improvement

The U.S. Air Force Research Laboratory is seeking machine learning and high-performance computing capabilities that could help improve the U.S. military’s space situational awareness and threat perception.

According to a broad agency announcement posted on SAM .gov, the capabilities must support data collection, processing, exploration, analysis and dissemination. The selected vendors will perform research, development, demonstration, integration, test and delivery tasks.

Specifically, AFRL needs algorithms and applications for several technical areas, including automated pattern learning and reasoning; anomaly detection, characterization and assessment; multi-source data exploration; data analysis and fusion; and high-fidelity tools for satellite modeling, classification and vulnerability assessment.

Total funding for the BAA is under $100 million, around $23.5 million of which is expected to be obligated in the fiscal year 2023. The BAA is open until Sept. 20, 2023, FedScoop reported.

Potential use cases for the ML and HPC capabilities include advanced orbit estimation and prediction, observation error characterization, and satellite identification and recognition.

In addition to space awareness improvements, AFRL is also looking for capabilities for 3D modeling and training set development for artificial intelligence inference engines and ML algorithms within deep neural network architectures.

The organization is also interested in HPC system accelerators that can project future needs for AI, ML and machine inference applications; radio frequency and radar applications; and data storage architecture for HPC system support.

AFRL expects to issue multiple contracts, grants, cooperative agreements or Other Transaction agreements. OTA awardees that come up with successful prototypes may be awarded a follow-on contract for production or transactions.