Defense Technology International - September 2007 - (Page 32) DISPATCHES GLOBAL SWARMING HOLE U.S. Navy nixes swarming for UCAS-D, but industry work continues DAVID AXE•WASHINGTON I ndustry has been refining the hardware, software and concepts for getting military robots to swarm like insects, but one of the most promising applications for swarming, the U.S. Navy’s Unmanned Combat Air System Demonstrator program, has been gutted by budget cuts. As a result, it might be years before swarming becomes a reality. There are thousands of unmanned aerial vehicles in service with U.S. and allied forces. But most operate singly and with a human operator in the loop. Autonomous swarms offer advantages in that they can Force as a partner and so had potential for more robust funding. But USAF dropped out last year to devote resources to its new manned bomber. Now there’s just enough money for a bare-bones UCAS demonstration. Even so, industry is banking on the Navy eventually issuing requirements for swarming. Several companies are working on methods of getting large groups of drones to navigate without colliding, collaborate on tasks and share data so information isn’t lost if one drone is disabled or destroyed. High-fidelity sensors on board each drone, combined NORTHROP GRUMMAN Northrop Grumman’s X-47 drone will be performing carrier operations, but no swarming, in the Navy’s UCAS-D trials. carry a variety of sensors and munitions and are less vulnerable than a single robot. The Navy’s $600-million UCAS-D program, which aims to test a fast, fighter-sized UAV, was an early opportunity to experiment with a swarm. A week after it tapped Northrop Grumman’s X-47 drone for the contract, though, the Navy announced that the six-year program would demonstrate only carrier operations. No weapons. No sensors. No refueling and no swarms. “The Navy told us they are going to fund the UCAS-D demonstration . . . and future activities are undefined,” says Bruce Parmenter, Northrop Grumman’s X-47 program manager. Economics are a factor. An earlier incarnation of the UCAS program included the Air 32 with software algorithms that tell the drone to sense and avoid objects, are one strategy for keeping a swarm from destroying itself. But experts say sense-and-avoid is expensive. Instead, several companies are working on networks that maintain distance between their nodes. These can also move imagery and targeting data from node to node. A team at New Jersey-based DRS Technologies is developing what team leader Pat Johnson calls an “autonomous collaboration network” for drones that is independent of GPS and of the Pentagon’s Global Information Grid. “We’re getting drones to talk to each other so they can swarm, work in teams, exchange target information and record strikes for battle-damage assessment.” Johnson says the network has been tested on a company aircraft, and undisclosed customers are interested. UCAS is an obvious application. The network is based on a station-keeping data-link system used by USAF cargo planes in the pre-GPS era. It relies on radio frequency transmissions using Ethernet standards and protocols, combined with proprietary software, to form a high-capacity data network where every node is aware of the distance to every other node. “We can also do autonomous fueling,” Johnson says. The ranging precision is such that a UAV with the system should be able to get close enough to a similarly equipped tanker that the drone’s short-range electrooptical and infrared sensors take over for making contact between the receiver’s probe and the tanker’s basket. Similarly, U.K. company Cobham and its partner StarVision Technologies are developing an autonomous refueling system that might have swarming applications. StarVision’s Zach Langenkamp says that the “intelligent supervisory controller” software, which coordinates UAVs refueling from manned and unmanned tankers, could manage swarms, since the basic functions—navigating and swapping range data—are essentially the same. Cobham and StarVision are aiming for a flight test by the end of the year using company-owned UAVs with inert tanking gear, but tests await FAA approval. FAA limitations on drone flights over the U.S. have both delayed testing of swarming technologies, and spurred them, since the FAA’s concerns are the same ones that keep engineers awake at night. Both the FAA and swarm architects worry that numerous semi-autonomous robots flying in the same airspace will collide with each other or with manned aircraft. Johnson says that the DRS network has safety-of-flight applications, and for that reason his team has built everything to commercial standards. Virginia-based Athena Technologies is working on safety-of-flight technology that might boost swarming efforts, says CEO David Vos. Athena is offering to outfit UAVs with the Automatic Dependent Surveillance-Broadcast system that will be an FAA safety-of-flight tool by 2013. Vos says it’s easy and cheap ($10,000) to equip a drone with modified transceivers that keep the aircraft up-to-date on locations of other aircraft. With ADS-B or a derivative, turning air traffic management into swarm coordination is simply a matter of fidelity and scale. Whatever program follows UCAS-D might have a swarming requirement, but not until after 2012, says Parmenter. “The next four or five years are about defining what the Navy wants to do.” I www.aviationweek.com/dti DEFENSE TECHNOLOGY INTERNATIONAL SEPTEMBER 2007 http://www.aviationweek.com/dti
For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.