US Army tests a low-cost acoustic sensor network in South Korea to detect FPV drones, reporting more than 90 per cent window-level confidence
TLDR
- The 532nd Military Intelligence Battalion tested Project MANOWAR, a passive acoustic network to detect FPV drones, in South Korea.1
- The Army reports more than 90 per cent window-level confidence; Army Recognition says that is not a detection-success rate.12
- The battalion wants scaled production and a programme of record; no fielding timetable has been announced.12
The US Army said on 13 November that the 532nd Military Intelligence Battalion "Black Horse", part of the 501st Military Intelligence Brigade (Theater), has spent eight months testing counter-drone capabilities in South Korea under Project MANOWAR.1 The work was a series of two-week field experiments at Camp Humphreys and Baegiri Airfield, home of the Republic of Korea Army Drone Academy, with the Army Research Laboratory, the 2nd Infantry Division, industry partners and Republic of Korea Army forces.1 MANOWAR stands for Mobile Acoustic Network for Operational Warning and Aerial Reconnaissance, and the Army describes it as a low-cost, attritable capability to detect, classify and locate low-signature small drones, including lethal first-person-view (FPV) types.1
The system pairs a 3D-printed chassis with a machine-learning algorithm developed in-house by military intelligence soldiers and scientists from the Army Research Laboratory's Mechanical Sensing Branch.1 Individual sensors run on standard military batteries and act as relay stations in a passive, self-healing mesh network, which the Army says can cover locations from patrol bases to critical theatre infrastructure.1
The release says the system's confidence rating against low-altitude threats rose significantly between the May and September events, and that by September it reached more than 90 per cent window-level confidence. It also says the system validated high-fidelity classification and localisation.1 It credits "Super Intelligence-enabled detection", which Unmanned Airspace reads as the US government's new term for artificial intelligence.13 Chief Warrant Officer 3 Sidney Beck said the team built a smaller onboard model from sensor recordings labelled by its 35T soldiers, that operators saw steadier detections, and that sensor reports reached a separate local display over the mesh network.1 Army Recognition notes that the release does not define the length of a window or publish detection ranges, false-alarm rates, localisation accuracy or unit costs, and says the figure describes classification confidence within analysis windows, not that more than 90 per cent of approaching drones were detected.2 It adds that MANOWAR provides warning and does not itself destroy drones.2
Lt. Col. William Campbell, the battalion commander, said that on the Korean Peninsula complex urban terrain and dense foliage restrict line of sight, so FPV threats "shorten our decision space and could place tactical units at risk".1 The release says the battalion is now focused on transitioning MANOWAR to scaled production and a formal programme of record; Campbell called that transition critical to achieving scale and said the sensor could also be exported to other geographic combatant commands.1 Those are the battalion's aims; Army Recognition says no fielding timetable has been announced.2 Unmanned Airspace carried the release's account and adds that Camp Humphreys is about 40 miles south of Seoul.3
Why it matters: the Army is testing cheap, passive acoustic sensing as a warning layer against FPV drones, but the headline confidence figure is undefined in public and no production or fielding timetable has been announced.12
Sources
- 501st Military Intelligence Brigade Experiments to Counter Emerging Threats (The United States Army)
- U.S. Army Tests in South Korea AI Acoustic Sensor Network to Detect FPV Drones (Army Recognition)
- US Army tests AI-enabled 3D-printed C-UAS with Republic of Korea (Unmanned Airspace)