AFRL updates ATA-AI BAA

On September 23, the Air Force Research Laboratory posted an update to the Broad Agency Announcement (BAA) for Advanced Tracking Architectures using Artificial Intelligence (AI) (ATA-AI). The due date for FY21 funding has been extended to 5:00 p.m. Eastern on October 16.

AFRL is soliciting white papers under this BAA for research, design, development, test, evaluation and experimentation of innovative technologies and techniques for Next Generation Target Tracking architectures, which exploit a wide array of both open data sources and traditional data sources and leverage the power of Artificial Intelligence (AI), Machine Learning (ML), and Machine Inferencing (MI) algorithms in a High Performance Computing (HPC) enabled framework.

Mission success within the Intelligence Community (IC) and Department of Defense (DoD) is critically dependent on information sharing and the ability to manage, discover, retrieve and access information across the organizational IC and DoD domains.  The objective of this BAA is to research, develop, integrate, test and deliver technologies/techniques associated with providing state-of-the-art solutions that could inform future, upcoming mission needs or requirements.  It is anticipated that Hardware  (HW) and Software (SW) solutions will be designed for Artificial Intelligence (AI), Machine Learning (ML), and Machine Inferencing (MI) applications.

This BAA seeks new algorithms, methods, novel techniques and applications for the five following technical areas (TA):

TA1: 3D pixel, vector, and point cloud processing and accelerations on a High Performance Computing (HPC) system.

TA2: Deep Neural Networks (DNNs) developed for measurement of social reactions in a population based on social media data.  This TA could include DNNs capable of classification and inference of social reactions in specific geographic regions and at varying levels of intensity across the full range of societal behaviors.

TA3: Deep Neural Networks (DNNs) developed for identification, classification and inference of information from imagery.

­ TA4: Ingestion and processing of Geographical Positioning System (GPS), non-GPS, Inertial Navigation System (INS), Radio Frequency Identification (RFID) trackers, or telematic based data into traffic tracks that can measure utilization of lines of communication.

­ TA5: Processing of cellphone Geographical Positioning System (GPS) and non-GPS data (Inertial Navigation System (INS), accelerometers, altimeters, personal fit devices, etc.) into and display Graphical Information System (GIS) data that can assist first responders in identifying the location of individuals in disaster areas based on last known location from their personal devices.

Full information is available here.

Source: SAM

­