This post is also available in: עברית (Hebrew)
Operational, intelligent swarming aircraft autonomy will be integrated into US Air Force unmanned platforms. The combat-proven Hivemind autonomy software will advance the autonomous collaboration capabilities of the Air Force Special Operations Command’s UAVs enabling them to team with Air Commandos in complex environments.
Aviation autonomy company Shield AI uses self-driving software to enable collaborative teams of unmanned aircraft to operate in GPS- and RF-denied environments.
The company has been recently awarded the maximum amount through the US Air Force’s Strategic Funding Increase (STRATFI) Program, in conjunction with Air Force Special Operations Command.
Hivemind is an autonomy and artificial intelligence stack for military and commercial aircraft, designed from the ground up to be multi-role, multi-vehicle. On teams of aircraft, Hivemind enables full autonomy and is designed to run fully on the edge in high threat, GPS and communication degraded environments. It uses state-of-the-art path-planning, mapping, state-estimation, and computer vision algorithms, combined with reinforcement learning (RL – a machine learning field) and simulations, to train unmanned systems to execute a variety of missions – from infantry clearance operations to breaching integrated air defense systems with unmanned aircraft.
The advancement of Hivemind supports the concept of AI for Small Unit Maneuver (AISUM), as outlined in AFSOC’s “Big 5” innovation initiatives.
The integration of Hivemind will initially be onto a commercial eVTOL aircraft – Beta Technologies’ ALIA, an Air Force Agility Prime platform. The integration will be in simulation to build the foundation for a follow-on full platform integration, according to the company’s announcement.