U.S. Army Tests ‘Golden Shield’ Sensor-to-Shooter Network to Destroy Drone Swarms at Machine Speed.

The U.S. Army’s 1st Cavalry Division proved it can autonomously detect, track, and destroy drones during a live-fire test of its Golden Shield air defense network at Fort Hood. By linking distributed sensors directly to weapons, the unit compressed the sensor-to-shooter chain to machine speed, demonstrating mobile, formation-level protection against the accelerating small UAS threat to armored forces.

Conducted April 7 to 9 under the Pegasus Charge initiative, the event delivered the Army’s first end-to-end engagement in which an autonomous sensor identified and classified a drone and immediately cued a separate weapon system to kill it. The integrated mix of sensors, robotic platforms, and interceptors shows how armored units can keep maneuvering under constant drone surveillance and attack without overloading crews, a critical step toward scalable, autonomous air defense in contested environments.

Read also: Golden Dome four-layer missile shield reflects the complexity of US aerial defense.

Golden Shield is a new U.S. Army layered air-defense concept tested by the 1st Cavalry Division at Ford Hood, linking autonomous sensors, robotic platforms, and kinetic effectors to detect and destroy small drones faster and with less crew workload (Picture source: U.S. DoW).

Led by the 1st Cavalry Division with Army DEVCOM and industry partners under the division’s Pegasus Charge transformation effort, the event produced the first live demonstration in which an autonomous sensor on one platform detected and classified a hostile drone, then passed engagement data to an autonomous weapon system on another platform for the kill. That is the kind of compressed sensor-to-shooter chain armored formations need if they are to keep moving under constant drone observation and attack.

Golden Shield should not be understood as a single launcher or radar, but as a formation-level system-of-systems. The Army says the architecture combines a next-generation command-and-control layer, multiple sensors, kinetic and non-kinetic effectors, and the Vehicle Protection System Base Kit, while Army budget documents describe that base kit as a modular open-systems survivability controller and the foundation for future hard-kill and soft-kill active protection integration. In other words, Golden Shield is designed to be a digital backbone onto which new defeat mechanisms can be added without rebuilding the entire vehicle protection stack.

That design choice matters because it reflects a deeper Army trend. Army research has already focused on distributed autonomous countermeasures for ground formations, including vehicle-to-vehicle threat sensing and response tied to the service’s modular active protection architecture. Golden Shield is important because it appears to bring that research out of the laboratory and into maneuver experimentation, where one vehicle no longer needs to carry every sensor and every effector as long as the formation can share tracks, cues, and engagement authority across the network.

Although the Army has not publicly identified the exact software stack used in the experiment, its description closely aligns with the service’s broader next-generation command-and-control direction, which emphasizes data-centric architecture, open standards, rapid software iteration, and the use of AI and machine learning to accelerate decisions at the division level and below. In a counter-UAS fight, that matters because machine-speed track management can prioritize shooters, reduce duplicate engagements, and preserve a common air picture even while an armored column is moving, dispersing, or masking behind terrain.

The armament revealed in Army imagery is especially notable. A DVIDS photo caption identified a Perseus Defense HARPE missile system rocket being test-fired during the exercise, while another image showed a Swarmbotics FireAnt V4 autonomous ground drone scanning terrain during Project Golden Shield. Perseus describes its offering as a mass-manufactured counter-UAS solution built around guided micro-missiles, while Swarmbotics presents FireAnt as an attritable, man-portable unmanned ground vehicle for counter-UAS, ISR, and electronic warfare with modular payloads and swarm autonomy. That combination suggests Golden Shield is already exploring both low-cost kinetic interceptors and robotic forward sensing inside the same engagement web.

Technically, that points to a layered defeat philosophy rather than a single exquisite interceptor. A guided micro-missile is relevant because it offers a lower-collateral and potentially lower-cost answer to small drones than firing scarce high-end SHORAD missiles, while robotic ground nodes can extend line-of-sight sensing, carry payloads away from manned crews, and create off-axis detection angles that improve classification in cluttered terrain. That is the logic behind formation-based protection: disaggregate the kill chain, distribute the risk, and make the cheapest effective effector available at the right moment rather than forcing every target into the same expensive engagement model.

Golden Shield also sits naturally alongside the Army’s existing counter-UAS inventory rather than replacing it. The service is fielding multiple defensive layers: Sgt. Stout on Stryker for maneuver SHORAD, M-LIDS for Group 1-3 drone defeat with radar, EW, 30 mm firepower, and Coyote interceptors, plus soldier-portable RF detection and jamming tools such as Wingman and Pitbull. Golden Shield’s real value is that it may allow those kinds of sensors and effectors to be federated at the formation level instead of operating as isolated point solutions.

One reason this is operationally important for armored brigades is that future effectors do not all need to be separate launchers. The Army is already developing the XM1228 BADGER 25 mm round for the Bradley, a proximity-fuzed aerial defeat munition that uses a miniaturized radar to sense a target in flight and detonate within lethal distance, without requiring a new weapon system or major vehicle modification. If a Golden Shield-type network can provide better detection and cueing, munitions like BADGER could allow ordinary Bradleys to contribute to the anti-drone kill web while preserving specialized missiles for more stressing threats.

The payoff is in seconds and workload. Recent Army analysis has stressed that armored formations cannot rely on exquisite radars on every platform or on dedicated air-defense assets at every lower echelon, while the drone threat is now cheap enough to arrive in swarms and persistent enough to paralyze movement. Golden Shield addresses that by reducing cognitive load, automating detect-track-cue functions, and potentially assigning the right shooter to the right target across multiple vehicles. For tank and Bradley crews already juggling navigation, fires, reporting, and survivability, that is not a convenience feature; it is what makes mobile air defense practical during combat maneuver.

The April live-fire event was explicitly a counter-small-UAS experiment, not proof that Golden Shield has already solved cruise-missile or broader integrated air-and-missile defense for armored formations. Yet the underlying architecture clearly has growth potential beyond quadcopters. The Army’s Vehicle Protection Suite effort is intended to integrate detection, intercept, and mitigation technologies against evolving direct and indirect threats, so Golden Shield’s significance lies partly in its ability to become a bridge between counter-drone defense, active protection, and future defeat mechanisms for missile-like loitering threats as the effectors mature.

Strategically, Golden Shield fits the Army’s emerging formation-based layered protection model, which senior leaders now describe as a continuum of avoiding detection, avoiding understanding, avoiding being hit, and surviving if hit. It also fits Transforming in Contact, whose purpose is to push commercial and non-developmental technology into soldiers' hands quickly enough for units to experiment under realistic conditions and reshape how they fight. That is why this test matters beyond Fort Hood: it is not simply a range event, but an early field argument for a new survivability architecture in which armored brigades carry their own adaptive anti-drone shield as an organic combat function.

The data now coming out of Fort Hood will determine whether Golden Shield remains a promising experiment or becomes a repeatable capability for armored formations. If the Army can harden the autonomy, validate the control logic, and prove reliable mixed kill chains using missiles, EW, robotic sensors, and organic vehicle weapons, Golden Shield could become one of the most consequential near-term survivability upgrades for U.S. ground combat units. In that sense, it is best seen not as a single air-defense weapon, but as a digital protective layer wrapped around the brigade.