AI-Driven Mission-Contingency Management for Autonomous UAVs

Lockheed Martin’s notoriously renowned Skunk Works division, depending on what side you find yourself on, has advanced significantly towards fully autonomous unmanned operations by successfully demonstrating artificial intelligence-driven mission-contingency management (AI/MCM). This was achieved using a Stalker XE Block 25 Unmanned Aerial Vehicle and a modified Alta X 2.0 drone. The live demonstration, conducted on December 4, 2025, in Fort Worth, Texas, highlighted how AI can quickly and intelligently handle unforeseen mission disruptions, enhancing the operational maturity of autonomous systems in both air and ground domains.

Stalker XE Block 25 Unmanned Aerial Vehicle

Central to the demonstration was the capability of an artificial intelligence agent to identify mission-critical anomalies and instantly create recovery options without needing operator input. In this scenario, various fuel-related contingencies were simulated. Once abnormal conditions were introduced, the AI-powered command-and-control (C2) system evaluated the issue, generated multiple re-planning paths within seconds, and displayed them to the operator via an intuitive interface.

Once the operator selected the preferred course of action, the AI took over to execute the recovery. The mission tasks of the Stalker UAV were seamlessly transferred to the Alta X platform, and Stalker was directed to return to base before its fuel levels became critical. This allowed the operator to focus on other mission priorities, avoiding the cognitive overload that can occur when managing multiple unmanned assets simultaneously.

Alta X 2.0 drone

The swift and smooth transfer of responsibilities between unmanned aircraft is a major advantage of AI-enhanced autonomy. In operational settings, particularly those that are contested, unforeseen events like weather changes, fuel irregularities, GPS degradation or system failures can occur within moments. Typically, these challenges demand the complete focus of a human operator. Lockheed Martin's AI/MCM demonstration illustrates how future unmanned systems can help shoulder this load, enhancing resilience, safety, and mission success.

The demonstration also showcased the growing ecosystem of unmanned air and ground (UxV) integration. The Stalker UAV gathered mission data and sent it to a unified C2 node, which simultaneously managed an Unmanned Ground Vehicle (UGV) operating in Kansas. This network was further enhanced by UAVs from Australian manufacturer Fulcrum Robotics, contributing to the creation of a geographically distributed yet centrally coordinated mesh of autonomous systems.

This capability, employing a single mobile node to manage multiple platforms across distant theatres, has been a longstanding objective for modern forces aiming for adaptable, swiftly deployable command solutions. Skunk Works' demonstration showed that one operator can oversee an air-to-ground unmanned network operating in mounted, dismounted and low-signature setups.

OJ Sanchez, vice president and general manager of Lockheed Martin Skunk Works, said the demonstration underscored AI's readiness to make a real impact in frontline operations. “This demonstration proves AI can move from the lab to the battlefield, delivering a multitude of capabilities ranging from autonomous decision-making to rapid data flow between unmanned vehicles across air, ground and synthetic environments,” Sanchez said. “By fusing AI-enabled UAV replanning with UGV capabilities, we give warfighters the safety, speed and confidence they need to act first in contested environments.”

A crucial factor in the demonstration was Lockheed Martin’s STAR.SDK™, which is a component of the company's expansive STAR.OS™ architecture. STAR.OS is crafted to enable developers to swiftly create, integrate, and implement AI-driven applications, whether they are for autonomy, data fusion, or decision support.

In this exercise, STAR.SDK linked the contingency-management AI directly to a user interface that included a chat-based assistant. This arrangement enabled operators to engage naturally with the AI assistant, allowing them to see re-tasking options and approve mission-plan changes without having to navigate through complicated system menus. By facilitating seamless communication and collaboration among various AI services, STAR.OS lays the groundwork for future autonomous operations across multiple domains. This encompasses the capability to integrate third-party unmanned systems or new technologies without significant software overhauls, a crucial requirement as armed forces increasingly embrace open-architecture systems.

The successful AI/MCM demonstration strengthens Lockheed Martin's continuous efforts toward advanced autonomy, distributed teaming and resilient mission systems. For military operators, the capability of unmanned platforms to identify, diagnose and adjust to mission disruptions independently can significantly lessen workload during complex operations and improve survivability in today's battlespace.

It also frames AI as a force multiplier: a resource that enables crews to oversee larger groups of unmanned aircraft and ground vehicles, react more swiftly to new threats, and sustain operational pace even in challenging conditions. With this recent achievement, Skunk Works upholds its longstanding tradition of developing technologies that transform air power. As autonomous capabilities advance and AI-supported decision-making gains more trust, the line between manned and unmanned operations will increasingly blur, leading to a new era of intelligent, integrated mission systems for the U.S. and its allies.