The Army has long assumed that clearing enemy wire obstacles can cost roughly half the engineers assigned to the job. At an Idaho training range, a 45 pound commercial drone just delivered the explosive that opens that lane, and no soldier went in first.
The U.S. Army's planning math has long treated a deliberate breach as one of the most expensive things an engineer unit can be asked to do. To clear a path through enemy wire obstacles for the rest of the force, the service has historically assumed the mission could cost roughly half the engineers assigned to it. On a windy June day at Orchard Combat Training Center in Idaho, an Oregon Army National Guard unit flew a 45-pound commercial heavy-lift drone called the Mule 28 through a simulated breach lane, lowered a live Bangalore torpedo (a tube packed with explosive and pushed or pulled into wire and detonated to clear a path), and triggered it without putting a single soldier in front of the obstacle.
The test grew out of a question the unit's commander could not stop asking. As one of the battalion's platoon leaders told Military Times, "the most casualty-producing thing that Army engineers do is the breach." Lt. Col. Eric Zimmerman, the battalion commander, stood up a small drone working group inside Bravo Company, 741st Brigade Engineer Battalion, 41st Infantry Brigade Combat Team, with a narrow brief: figure out how to clear a wire obstacle with a commercial drone.
The hardware is not what most readers picture when they hear the word drone. The Mule 28 was built by Lorica Technologies for this mission. It is an eight-motor heavy-lift with 28-inch bi-blade propellers and a roughly 200-pound payload rating, with airframe work finished in about six weeks. The commercial drones the unit surveyed as it designed the test ranged from roughly $2,000 to $40,000, a band that puts the off-the-shelf options in a different cost neighborhood from a purpose-built heavy-lift airframe. Pilots flew the Mule 28 manually through the breach lane; a separate arming system handled the explosive. The aircraft is not autonomous in the sense of picking its own path through the obstacle.
Zimmerman's working group framed the project as a direct response to what the war in Ukraine has done to the calculus of getting engineers close to enemy wire. Ukraine's drone innovation has been a steady backdrop for U.S. tactical experimenters since 2022, and watching small commercial systems take on tasks that once required a soldier to be physically present was the prompt for asking whether a heavier drone could carry a heavier charge to a target a sapper would otherwise have to approach on foot. The Orchard test is the first documented U.S. Army attempt to fly a live Bangalore through a wire obstacle on a commercial heavy-lift.
It is also, by the working group's own admission, a single data point. The unit flagged that it had not found another U.S. Army example of a drone-delivered Bangalore torpedo, a caveat that frames the test as an experiment rather than a doctrinal shift. The Army's own press release, mirrored on GlobalSecurity, confirms the Orchard trial and the unit's intent to expand drone use across engineer missions, and the local KDRV report placed the test in 25 mph gusts that would have complicated any human approach on foot.
The breaching experiment is one piece of a broader pattern in the Army's 2026 tactical-innovation cycle. Earlier in June, soldiers from the 101st Airborne Division used small commercial drones to drop grappling hooks and 3D-printed munitions against razor wire, a separate attempt to push the soldier out of the front of the obstacle, according to Military Times. In the Pacific, the service has been running autonomous and AI-assisted surface vessels in sustainment and logistics roles, work covered by Defense One and by Stars and Stripes. The breaching trial is the most casualty-oriented of those threads and the first to put a live demolition charge on a heavy-lift airframe.
What to watch next is whether the Army treats the Orchard test as the start of a program of record or as a one-off the working group dissects and shelves. The honest read sits between those poles: one unit, one range, one munition type, in winds that would have made a manual approach punishing, executed against simulated wire rather than live defenders. The mechanism the test demonstrated is not that drones replace combat engineers. It is that unmanned delivery can decouple the casualty cost of the breach from the breaching mission, and a single demonstration on a training range cannot settle whether that decoupling survives contact with a real obstacle under fire.