U.S. Army Tests in South Korea AI Acoustic Sensor Network to Detect FPV Drones.

The U.S. Army has tested an AI acoustic sensor network to detect FPV drones in South Korea under Project MANOWAR. The experiment pairs soldier-developed machine learning with distributed sound sensors to add a warning layer against small aircraft that threaten troops and military sites in terrain with limited visibility.

A U.S. Army soldier assigned to the 501st Military Intelligence Brigade observes unmanned aircraft at Baegiri Airfield, South Korea, on November 9, 2026, as part of the unit’s work on Project MANOWAR, an acoustic counter-drone detection network. (Picture source: U.S. Army)

Key Tactical Insights:

• MANOWAR distributed detection: MANOWAR combines acoustic sensors and machine learning to identify and locate small drones, potentially extending warning coverage around dispersed positions and supporting a layered counter-drone defense.

• U.S. Army-wide adaptation: The U.S. Army is expanding unmanned aircraft employment and updating doctrine around persistent aerial observation, increasing the importance of integrating drone operations with protection against hostile aircraft.

• Experimental status: The reported confidence figure does not establish a verified detection-success rate. MANOWAR remains an experimental sensing capability, with production, formal adoption, and operational fielding still unconfirmed.

What is Project MANOWAR?

The U.S. Army disclosed Project MANOWAR’s results on November 13, 2026, following experiments conducted in May and September in South Korea. The 532nd Military Intelligence Battalion, part of the 501st Military Intelligence Brigade, worked with South Korean forces, Army Research Laboratory scientists, the 2nd Infantry Division, and industry partners at Camp Humphreys and Baegiri Airfield.

MANOWAR stands for Mobile Acoustic Network for Operational Warning and Aerial Reconnaissance. The system combines 3D-printed sensor housings with machine-learning software developed by military intelligence soldiers and Army researchers. It is designed to detect, classify, and locate small unmanned aircraft, including first-person-view attack drones.

Related Topic: U.S. Army Funds New Drone-Based Protection for M1A2 Abrams Against Mines and Aerial Drone Threats

Individual sensors operate on standard military batteries and relay reports through a distributed mesh network to a separate local display. The U.S. Army describes the network as self-healing, allowing reporting connections to recover as the network changes. Its intended role is to add a sensing layer around locations ranging from patrol bases to theater infrastructure.

Between the May baseline experiment and September testing, developers produced a smaller onboard model using sensor recordings labeled by U.S. soldiers. Operators reported steadier detections, while the Army announced more than 90% window-level confidence against low-altitude threats.

That figure requires careful interpretation. It describes classification confidence within analysis windows, not that more than 90% of approaching drones were detected. The disclosure does not define the window duration or publish detection ranges, false-alarm rates, localization accuracy, or unit costs. MANOWAR provides detection and warning; it does not destroy drones itself.

How the U.S. Army is expanding drone use

The experiment accompanies a broader U.S. Army effort to integrate unmanned aircraft into ground formations. Small drones support reconnaissance, target acquisition, strike and electronic warfare, giving soldiers additional ways to observe terrain and engage threats while reducing exposure during some tasks.

A Congressional Research Service report updated May 6, 2026, describes an approach that aligns small-drone capabilities with squad, platoon, company and battalion requirements. It identifies a platoon-level program intended to provide FPV drone capability and records a Pentagon directive to field unmanned systems in every U.S. Army division by the end of 2026. That deadline represents a stated objective, rather than confirmation that the U.S. Army has completed the effort.

The service is also changing how it incorporates operational experience into doctrine. In March 2026, the Army described a force-wide effort to update its guidance as units gain experience with unmanned systems. Its operations doctrine emphasizes protection against constant observation and making initial contact through sensors, unmanned systems, or the smallest suitable element.

This expansion creates an integration requirement alongside the equipment requirement. Units need to coordinate friendly drone flights, manage communications, and distinguish hostile aircraft from their own systems. A distributed warning network could support that process, provided its reports reach operators quickly and reliably enough to inform action.

Lessons from Ukraine and recent Middle East conflicts

The war in Ukraine has directly influenced U.S. Army counter-drone training. In May 2026, soldiers at Fort Bragg tested existing weapons and ammunition against small drones during an exercise explicitly framed around lessons from the Ukrainian battlefield. The U.S. Army presented the effort as a search for affordable defenses that soldiers could employ without waiting for lengthy development and procurement cycles.

Ukraine’s experience has reinforced the importance of adapting equipment, training and tactics together. Persistent aerial observation complicates movement and concealment, while FPV attack drones create threats that reach beyond traditional frontline positions. For the U.S. Army, this means both employing unmanned aircraft and maintaining the ability to maneuver under drone surveillance.

Middle East operations have also demonstrated the vulnerability of support installations. On January 28, 2024, a one-way attack drone struck Tower 22, a logistics support base in northeastern Jordan, killing three U.S. service members. The attack illustrates why drone warning and protection requirements extend to the locations sustaining deployed forces.

The U.S. Army has documented a connection between experience from both theaters and its development work. An October 2026 account of Project Flytrap states that the 2nd Cavalry Regiment’s evaluations drew on combat data from Ukraine and the U.S. Central Command area. Testing highlighted the need for layered defenses and exposed difficulties combining separate sensor feeds into one operational picture.

MANOWAR fits this wider emphasis on distributed sensing, rapid experimentation and integrated warning. However, the U.S. Army’s disclosure does not establish that the project directly incorporates Ukrainian acoustic technology or a particular Middle East combat lesson. The battalion is now seeking scaled production and establishment of a formal Army program of record, with no approved fielding timetable announced.

Explore More Defense News

• Land Defense News
• Naval Defense News
• Defense Aerospace News

Written by Alain Servaes – Chief Editor, Army Recognition Group

Alain Servaes is a former infantry non-commissioned officer and the founder of Army Recognition. With over 20 years in defense journalism, he provides expert analysis on military equipment, NATO operations, and the global defense industry.