Restricted Research - Award List, Note/Discussion Page
Fiscal Year: 2023
2512 The University of Texas at San Antonio (144400)
Principal Investigator: Houpt, Joseph
Total Amount of Contract, Award, or Gift (Annual before 2011): $ 29,640
Exceeds $250,000 (Is it flagged?): No
Start and End Dates: 9/1/22 - 12/31/22
Restricted Research: YES
Academic Discipline: none
Department, Center, School, or Institute: Matrix AI Sponsorships
Title of Contract, Award, or Gift: RTX Application of Artificial Intelligence to Defense Systems Human Machine Teaming (RAAIDS-HMT) Phase II
Name of Granting or Contracting Agency/Entity:
Raytheon Company
CFDA: 0
Program Title: none
Note:
SAMS 1.1.1; The RAAIDS project is developing cognitive aides that can be deployed within reconnaissance, surveillance, and target acquisition (RSTA) sensing systems to speed the observe-orient-decide-act (OODA) loop and shorten the target kill chain. It is demonstrating how increasing the autonomous operations of sensors can improve on-board processing, exploitation, and dissemination (PED) of collected data. In particular, RAAIDS is developing AI/ML-enabled algorithms that can autonomously control operations of human-machine teamed sensors to achieve greater automated detection, classification, and tracking of objects in the field of regard, optimizing both within an individual RSTA system and across multiple systems and platforms that are sensing cooperatively. By increasing sensor autonomy in this way, RAAIDS will enhance sensor systems in two key ways: 1) The system will be able to actively interrogate potential targets and the environment by autonomous manipulation of its controllable parameters or via cooperative sensing across platforms. This will enable dynamic tuning of sensory data collection within and across platforms to improve Automatic Target Recognition (ATR) performance, ameliorating performance limitations that arise due to nuisance factors such as obscuration, clutter, and variations in view, scale, and illumination. 2) The system will be able to surreptitiously and opportunistically collect and exploit sensor data on-board the PED platform. Potential threats or items of interest will be detected, classified, and tracked, proactively providing the human operator with relevant information without being directly tasked and without further adding to the operator cognitive load.
Discussion: No discussion notes