UAMS Experiment Brings Deconfliction Closer for Smaller UAVs
Limitations on UAV use are imposed by the threat of collisions between UAVs and manned aircraft. An RQ-7 Shadow UAV is definitely large enough to create real problems if it hits a helicopter or other aircraft, and a UAV’s extremely narrow field of view is a lot less safe than the awareness available to a human in a cockpit. Worse, many UAVs are small enough that a potential collision may not be noticed by other aircraft until it’s too late. There have already been accidents.
This isn’t just a military problem. It also represents the largest barrier to widespread civil UAV use. Europe’s EDA has a program underway to address deconfliction, the Israelis are looking into it, the US military is funding research from multiple UAV controllers to SWARMs, and even private contractors are busy searching for the key that will unlock a vast UAV market. The ultimate goal is a system that’s small enough to equip smaller and more affordable tactical and civil UAVs, as well as larger and more expensive military UAVs like the MQ-9 Reaper and RQ-4 Global Hawk.
A recent project sponsored by the US Army, and led by Lockheed Martin, is bringing that goal closer – and may have ramifications for the inter-service balance of power.

The US Army’s Aviation Applied Technology Directorate sponsored the research that led to Lockheed Martin’s Unmanned Aerial Vehicle Airspace Management System (UAMS), after 3 years of research & development.
UAMS separates airspace deconfliction for UAVs into 3 activities: (1) maintaining situational awareness and a sufficiently up-to-date common picture of aircraft and UAVs in an airspace; (2) detecting conflict; and (3) modifying flight paths. The key problem with smaller UAVs, however, is their lack of carrying capacity. The logical solution is to distribute any deconfliction approach, in order to place heavier items on the ground, other aerial platforms, et. al.
UAMS uses this distributed approach, with software agents on the UAVs, and a centralized server on the ground that collates the information and distributes a common picture of what’s going on in the air near participating UAVs. UAMS can also use a combination of techniques, dynamically shifting among the 3 core activities based on the situation, user-defined policies, bandwidth and server load, and other factors.
- Lockheed Martin Advanced Technology Laboratories developed the distributed, vehicle-information-management technology, concept of operations, and systems engineering.
- SRI provided avoidance-planning algorithms.
- SkEyes added key sensors, including forward-looking, conic, laser radar and acoustic sensors, in order to provide UAVs with the “sense and avoid” capabilities that are naturally resident in human pilots.
An August 2008 test near Pittsburgh, PA capped the initial effort, by demonstrating its ability to deconflict groups of in-flight UAVs. Lockheed Martin explicitly describes UAMS as a “battalion echelon system,” which means it’s designed to work with the tier just above mini-UAV size: Elbit’s Skylark II, IAI’s I-View 250, Textron’s RQ-7, et. al.
Implications
Further tests, and further development, will be necessary before UAMS becomes thinkable as an operational system. Should it succeed, however, the implications could be profound.
“CENTCOM Looks to Boost ISR Capabilities in 2008-2009” discussed the profound effect on close combat support being created by the combination of precision artillery, and small surveillance and targeting turrets on aircraft and UAVs. Army efforts like Project ODIN have demonstrated impressive battlefield results, and are being rewarded with support from the current Secretary of Defense.
The US Air Force has made several moves in recent years to bring all UAVs under its jurisdiction, at both procurement and operational levels. It has few successes to show for these efforts thus far, but the service has seen the relative flying hours chalked up by UAVs and manned aircraft in theater, and remains persistent.
One of its strongest arguments has always been the need for a common operator, in order to address airspace deconfliction issues and prevent lethal tragedies.
If that problem goes away, and does so in a manner that allows smaller Army UAVs like Textron’s RQ-7s to fly with fewer restrictions, a couple of things happen. One is that the overall argument for a single service coordinating all UAVs becomes much, much harder to make successfully. As a related consequence, control of multiple UAVs from new platforms like the Army’s AH-64D Block III Apache Longbow fleet becomes much easier. In effect, each AH-64D Block III could become its own self-contained hunter-killer flight, with associated battalion-level UAVs cueing their attached helicopters in a manner similar to the current Project ODIN – but with less need for backhaul bandwidth. A development that would only cement a strong argument for federated, rather than centralized, control of battlespace UAVs.
Additional Readings & Sources
- Defense Update (Aug 25/08) – Army Sponsored Team Demonstrates Airspace Deconfliction of Multiple UAVs. See also Avionics magazine summary.
- DID DOCUS – Warrior ER/MP: An Enhanced Predator for the Army. See esp. Appendix A: US Army et. al. vs. USAF Over UAVs.
- NY Times (June 22/08) – At Odds With Air Force, Army Adds Its Own Aviation Unit. Project ODIN includes both manned C-12s and UAVs like SkyWarrior. “The work of the new aviation battalion was initially kept secret, but Army officials involved in its planning say it has been exceptionally active, using remotely piloted surveillance aircraft to call in Apache helicopter strikes with missiles and heavy machine gun fire that have killed more than 3,000 adversaries in the last year and led to the capture of almost 150 insurgent leaders.” See also…
- CASR (June 22/08) – Counterinsurgency Legacy – US Army Aviation Supports its Own: US Air Force turns out to be too Tardy to be Tactically Useful. Excerpts and background in the NYT article. The US Army is reportedly seeking money to raise a similar unit in Afghanistan by late 2008/early 2009.