General Atomics Aeronautical Systems (GA-ASI) has advanced its unmanned combat aircraft capabilities with a live demonstration of an MQ-20 Avenger operating under autonomous control via a Low Earth Orbit (LEO) satellite communication link. Conducted on April 6, 2023, the flight showcased collaborative maneuvers between human pilots and onboard artificial intelligence (AI) agents, integrating the company’s Live, Virtual, Constructive (LVC) collaborative combat aircraft ecosystem with beyond-line-of-sight (BLOS) connectivity.
The event marked the first time a LEO SATCOM provider’s connection was employed on an operationally relevant unmanned combat aerial vehicle platform. The datalink, operating over an IP-based architecture, enabled real-time tactical control and rapid AI model updates while the aircraft remained airborne. According to GA-ASI, AI pilots were retrained and redeployed within minutes, a capability that significantly compresses the adaptation cycle in dynamic mission environments.
Hardware integration was central to the test. Two L3Harris Technologies RASOR Multi-Functional Processors (MFPs) were installed—one hosting the transceiver card, the other managing the BLOS Active Electronically Scanned Array (AESA). The AESA system, provided by Ball Aerospace, supported full duplex operation, ensuring simultaneous transmission and reception of mission data. This combination of processors and antenna technology provided the high-throughput, low-latency link necessary for AI-driven maneuvering and operator oversight.
“The flight demonstrated GA-ASI’s unmatched ability to fly autonomy on real, tactically relevant, unmanned combat aerial vehicles,” said Michael Atwood, GA-ASI Senior Director of Advanced Programs. “It displayed effective BLOS Command and Control through the collaboration between three defense primes. This showcases our rapidly maturing CCA mission system suite and moves us one step closer to providing this revolutionary capability to the warfighter.”
The demonstration leveraged GA-ASI’s end-to-end CCA ecosystem, which integrates third-party capabilities with human-on-the-loop oversight. Operators issued commands via hands-on-throttle-and-stick (HOTAS) controls, which were transmitted through the LEO SATCOM link to AI pilots running reinforcement learning algorithms. These AI agents autonomously tracked and maneuvered around dynamic entities identified by the operator, while continuously relaying status updates to a cockpit heads-up display. As mission parameters evolved, operators could re-task the AI in real time through the same HOTAS interface.
A key element of the trial was the closed-loop AI improvement process. Performance data from the AI agents was transmitted to ground systems, where retraining occurred using updated mission conditions or tactical objectives. The revised AI models were then uplinked back to the aircraft over the LEO SATCOM connection, replacing the active control agents within minutes. This capability underscores a shift toward adaptive autonomy, where unmanned systems can evolve their decision-making logic mid-mission without returning to base.
The Avenger platform, designed for high-subsonic performance and capable of carrying a range of sensors and munitions, provided a relevant testbed for these technologies. Its integration with advanced datalinks and AI control agents reflects a broader industry trend toward collaborative combat aircraft concepts, where manned and unmanned systems operate in concert to expand mission reach and resilience.
GA-ASI has a history of iterative technology demonstrations funded through internal research and development, using flights like this to validate key concepts before they transition to operational programs. The company’s portfolio, which includes the Predator series and Lynx multi-mode radar, has accumulated over seven million flight hours, underscoring its experience in delivering long-endurance, mission-capable systems with integrated sensors and data links.
By combining LEO SATCOM’s low-latency global coverage with AI-driven autonomy and human-machine teaming, the April 6 flight demonstrated a pathway toward more responsive and adaptable unmanned combat operations. The collaboration between GA-ASI, L3Harris, and Ball Aerospace highlighted the role of cross-industry partnerships in accelerating the deployment of next-generation combat air capabilities.