Many of the outstanding challenges to the development and deployment of tactical autonomy are related to operations in the real world. Through AIR, DARPA is looking to prioritize fully integrated sensors, scalability and adaptability in the face of “open-world problems,” as well as deceptive effects and the learning of predictive models “that capture uncertainty and automatically improve with data,” according to a document posted with the listing.
AIR will focus on previously avoided dimensions that must be addressed to enable tactical autonomy in operationally relevant combat: fully-integrated sensors, scalability to larger engagements, adaptability to changing conditions in open-world problems, and the ability to learn predictive models that incorporate uncertain knowledge of adversary and self, as well as deceptive effects.
AIR will pair existing, maturing, and emerging algorithmic approaches with expert human feedback to rapidly evolve the cooperative autonomous behaviors that solve previously avoided challenges.
AIR will address two technical areas:
1. Creating fast and accurate models that capture uncertainty and automatically improve with more data.
2. Developing AI-driven algorithmic approaches which enable real-time distributed autonomous tactical execution within uncertain, dynamic, and complex operational environments.
The AIR program will also develop the processes needed to rapidly design, test, and implement future iterations of AIR software products.
The Defense Advanced Research Projects Agency (DARPA) Tactical Technology Office (TTO) sponsored a Proposers Day in Nov 2022 to provide information to potential proposers on the objectives of the AIR program in advance of a planned Broad Agency Announcement (BAA).