A quick look at unmanned air vehicles (UAVs) in operation or development reveals their amazing variety. In wingspan, they range from six inches (15 cm) to 247 feet (75 meters). In flight mode, they embrace fixed-wing, rotary-wing, MV-22 Osprey-like vertical takeoff/landing (VTOL), and birdlike, flapping-wing designs. In altitude, they stretch from 50 feet to 100,000 feet, and in mission, range from intelligence, surveillance and reconnaissance (ISR) to weapons launch. Many UAVs are in development.
Vehicles vary in autonomy, as well. The Predator, an ISR UAV, is flown from the ground by a pilot with rudder pedals. Global Hawk, on the other hand, is flown with a mouse. On Global Hawk’s recent 23-hour trip to Australia, "the only [human] intervention was when the pilot did a ‘turn and hold’ so we could make sure all our communications on the ground were good," says Col. Wayne Johnson, program manager. The ground pilot monitors and adjusts the profile, as necessary.
Equipped with dual-redundant avionics systems–and dual, dissimilar navigation systems–Global Hawk "flies its mission and comes in and lands itself one to two feet [0.3 to 0.6 meter] off the centerline, at max," adds George Guerra, deputy program manager for prime contractor Northrop Grumman.
Vietnam to Macedonia
Unmanned air vehicles operated in the Vietnam, Yom Kippur and Gulf conflicts. More recently the Predator flew in Kosovo and Macedonia. If all goes well, UAVs may fly in large numbers on future battlefields, improving situational awareness, reducing fratricide and saving pilots’ lives.
"The UAV concept is growing exponentially," says Tom Schamberger, business development manager for advanced tactical data link with Rockwell Collins. He predicts the worldwide UAV market will grow from $2.1 billion in 1998 to $42.2 billion by 2008. Collins already provides ARC210 radios to the Air Force Global Hawk and Predator programs, the Navy Fire Scout and Army Hunter programs, and to Boeing on the Unmanned Combat Air Vehicle (UCAV).
Multiple UAVs mean different avionics solutions. Large vehicles are more able to use avionics developed for manned aircraft. The challenge is to integrate individual threads into a single system and adapt them to UAV platforms. The Fire Scout unmanned helicopter and Global Hawk, for example, stress commercial off-the-shelf (COTS) and government off-the-shelf (GOTS) equipment. Miniature UAVs, however, face more challenging avionics size/performance constraints. Programs also are eyeing collision avoidance systems to meet Federal Aviation Administration (FAA) requirements for airspace safety.
The Defense Advanced Research Projects Agency (DARPA) wants to make available for its Organic Air Vehicle (OAV) program uncooled infrared (IR), "panospheric" electro-optical (E/O), laser radar (ladar), and ultra wide-band (UWB) sensors. The OAV program–involving the development of cylindrical, VTOL vehicles from six to 36 inches (15 to 91 cm) in diameter–is a part of the Future Combat Systems program, established to develop technologies for the Army’s futuristic "objective force" concept. OAV could one day serve as the eyes and ears of small, platoon-like units. Honeywell Engines and Systems, Torrance, Calif., and Micro Craft Inc., of San Diego, Calif., are developing vehicles under the program’s first phase.
"Avionics becomes a key challenge," says Siamak Nikbin, a Honeywell senior systems engineer. Mission avionics technologies are being developed under concurrent contracts, so final data on size, interfaces and performance is not yet available.
Flight avionics isn’t easy either. If designers, for example, use GPS (Global Positioning System) to help maintain as well as to establish aircraft position in adverse weather conditions, the 1-Hz sampling rate offered by commercial micro receivers "may not be sufficient," Nikbin says.
For gyros, Honeywell is looking at micro electromechanical systems (MEMS). The company may also use differential GPS, he adds.
DARPA expects to demonstrate several vehicle sizes. "But we’re not selling a particular size–we’re selling scalability," says Sam Wilson, OAV program manager. The vehicles are intended for ISR missions, and are capable of flying under the trees at 50 to 100 feet above the ground, with about one hour of flight time. Other proposed missions include map and terrain generation for unmanned ground vehicles, response to operator requests for information, response to unattended ground sensor network commands, and target identification.
The OAV program plans a 6.2-mile (10-km) communications range. Although the idea is to launch the craft from a small-unit mobile command post or robotic ground vehicle, the vehicles could be dropped from larger UAVs or airplanes, increasing their range. DARPA plans an HF, directional, line-of-sight antenna for the OAV. It provides low probability of intercept/detection.
The coffee can-shaped OAV is essentially a ducted fan engine with fixed-pitch props. Photos of demonstration vehicles show them perched on vertical, X-shaped or circular landing gear. After taking off vertically, the vehicle dips over and flies horizontally. Vanes are used to provide roll, pitch and yaw. GPS is used for positioning and for correcting drift, and inertials for flight control. Avionics pods are located at the center or the perimeter of the cylinder.
A key OAV sensor under development by Honeywell Labs in Minneapolis, Minn., and Infrared Solutions, Plymouth, Minn., will be an uncooled IR "feather camera," based on vanadium oxide (VOx) material. Such detectors have commercial applications but are not produced to the thinness, size and complexity needed by micro air vehicles, says Ray Balcerak, a DARPA program manager. Detector elements need to be about 0.1 micron (1 ten-millionth of a meter) thick.
Commercial providers supply sensors whose individual detector elements, or pixels, measure about 50 microns (millionths of a meter) on a side, discriminate betweentemperature differences of 100 and 200 milliKelvin (thousandths of a degree Kelvin), and weigh from 3.5 to 7 ounces (100 to 200 grams). But DARPA wants to get temperature discrimination down to 50 to 70 milliKelvin, weight to 1.2 ounces (35 grams) "without the battery," and pixel size to 35 microns on a side–by this November, Balcerak says. The agency aims to reach 0.35 ounce (10 grams), 5 to 10 milliKelvin, and 15 microns next year.
The sensor also must minimize heat loss. This calls for the design of "long, skinny, winding legs" that position the detector above the electronics, make an electrical connection and support weight, Balcerak says. Although the exact material and structure have not yet been determined, the substance must be a good electrical conductor but a poor heat conductor, and this poses a lithographic challenge.
Developers also are analyzing the detector to measure its responses–across the full military temperature range–when heated or cooled by the outside environment. Electronics must be developed to compensate for these effects and maintain the sensor’s ability to make fine temperature distinctions, as the unit must be able to function wherever U.S. forces deploy. Otherwise, the variations in detector temperature "are going to mask the temperature changes in the scene," Balcerak explains.
A second project–by Athena and Carnegie-Mellon University–is a 4-million-pixel, panospheric, 360-degree view, electro-optical camera. Part of the processing challenge is deciding what information to send back (640 x 480 out of 4 million pixels) and correcting at an acceptable rate the distorted image created by the sensor’s spherical stainless steel mirror. The camera would see 10 degrees above the horizon and 80 degrees down.
A third sensor that would feed into the OAV program is a laser radar (ladar) called Jigsaw. The ladar would take snapshots of a possible target from different angles through, for example, an opening in a wooded area. Multiple glances then capture different parts of a vehicle, which the processing system reassembles into a single image for human interpretation. Jigsaw is expected to register the geolocation of target segments to "within a few centimeters," Wilson says.
Ladar sensing is ideally suited for target identification because it can "poke through" holes and provide three-dimensional (3-D) information, a DARPA document says. DARPA hopes that ladar sensing can combine 3-D information "from multiple views, segmenting away irrelevant information."
All UAV programs are concerned with collision avoidance—avoiding ground and air obstacles. Low-flyers must see and avoid tree branches, power lines and buildings. Higher-altitude vehicles must avoid other aircraft.
Multispectral Solutions Inc. (MSSI), Gaithersburg, Md., has engineered an ultra-wideband (UWB) radar prototype for the OAV program. It weighs 1.4 ounces (40 grams) and employs standard, printed circuit board packaging, says Don Perino, MSSI’s engineering director. The device radiates about 0.25 watt instantaneous peak power and about 750 microwatts (millionths of a watt) average power. The radar characteristics provide low probability of intercept/detection and high antijam immunity. MSSI’s goal is a weight of 0.35 ounce (10 grams or two nickels), which Perino believes can be achieved with chip-on-board or similar packaging. The sensor operates in the 6-to-6.5-GHz range, in non-restricted spectrum, says Bob Fontana, MSSI president.
The radar will detect an object, such as a tree branch, out to a required 50-foot (15-meter) range. MSSI also wants "to minimize the blind zone up close, to get it to a few feet away" from an object, Perino says. The craft will need to hover right next to a wall and land on a rooftop or building ledge without crashing.
MSSI plans to deliver the radar electronics and antenna in February 2002. The current version measures about 3 by 3 by 0.3 inches–the size of a Post-it, Perino says. The company claims it could develop a multipurpose radio frequency (RF) sensor, capable of communications, positioning, altitude determination and automated landing functions, as well as collision avoidance. But such research is outside the scope of the current contract.
"In the UAV community, a No. 1 concern is to get blessed by the FAA," says Collins’ Schamberger. "FAA’s No. 1 concern with the UAV community is flying through airspace safely."
Global Hawk, the Air Force’s 25,600-pound (11,612-kg), 116-foot (35-meter)-wingspan behemoth is eyeing traffic alert and collision avoidance system (TCAS) technology, Mode-S transponders, and other see-and-avoid solutions, says program manager Johnson. To relay voice messages from FAA controllers, Global Hawk can employ wideband Ku satcom, line-of-sight common data link, or UHF.
The program also is considering a look-ahead electro-optical or IR camera for non-cooperative collision avoidance. "We want to take the visual range that a pilot can see and…do as well as that, if not better," Johnson says. The program also is interested in "a cueing system that will allow you to highlight [on a ground display] objects that could be on a collision course."
Collision avoidance and see-and-avoid technologies are very big on the list, agrees Richard Lee, chief engineer for the Air Force Predator program. But he raises questions concerning the predictability of highly automated collision avoidance systems in unmanned vehicles.
"With TCAS, you have a computer deciding what to do, and you wonder what it’s going to do. If there’s no person in [the UAV], how predictable is it?" Lee asks. While collision avoidance is important, "the FAA and the Air Force have not…figured out how they want to go about it."
"How to take an unmanned airplane and make it GANS/GATM-compliant is a challenge," Lee adds. (GANS/GATM is the Air Force Global Access, Navigation and Safety/Global Air Traffic Management program to ensure military airplanes are equipped with the communications, navigation and surveillance equipment necessary to operate in civil airspace.) The Predator program currently uses Collins ARC210 radios for civil controller voice relay. Pilots see through a nose-mounted, color camera for flight control. Mission equipment includes E/O, IR and synthetic aperture radar sensors.
NASA and General Atomics, meanwhile, are developing larger, more powerful "Predator-B" vehicles, says John Hicks, manager of NASA’s Environmental Research Aircraft and Sensor Technology (ERAST) program. San Diego-based General Atomics hopes to interest the military in the idea, to fill the gap between short-range, tactical, and long-range, strategic UAVs. NASA’s version, the Altair, is a turboprop-powered, 84-foot (25.6-meter)-wingspan vehicle intended to fly to 52,000 feet, with 32-hour endurance and 400-nm range. It can perform Earth science, remote sensing and disaster management missions.
NASA also is developing a compact Ku-band, over-the-horizon satcom system, because the COTS version is too big. L-3 Communications Corp. is designing a system with a 12-inch (30.5-cm)–as opposed to 30-inch (76-cm)–dish. General Atomics is looking at TCAS, Hicks says. But NASA has asked Navy researchers in China Lake, Calif., to develop a supplementary, non-cooperative, IR-based system for the agency with approximately +/- 35-degree elevation and 210-degree azimuth performance.
Navy developers of the Fire Scout, an unmanned helicopter, plan to add collision avoidance in the UAV’s first block upgrade if funding is available, says Cmdr. Osa Fitch, Naval Air Systems Command’s (NAVAIR’s) integrated product team lead for UAV advanced development. In the near term, the program is looking at cooperative collision avoidance, using Goodrich Avionics Systems’ Skywatch HP traffic advisory system or a future derivative. Skywatch HP currently tracks 35 aircraft simultaneously and generates both audio and visual traffic advisories.
Further out, Fire Scout developers are eyeing an active/passive non-cooperative system combining E/O, IR, a light detection sensor, and a laser rangefinder, Fitch says. However, the project involves significant technical risk in integration, algorithm development, cost and schedule.
Fire Scout’s mission payload includes E/O and IR cameras and a laser–all boresighted, or aligned, with each other–integrated into the same turret or ball. The diode-pumped laser will both designate targets and determine the exact range to the target. The E/O camera will identify targets as small as 7.6 by 7.6 feet (2.3 by 2.3 meters) from 6.8 miles (11 km) away.
Fire Scout is one of the first Navy programs to use the 10.71-Mbit/sec tactical common data link (TCDL) to pass real-time imagery, providing range out to 150 nm, line-of-sight permitting, says Capt. Lyn Whitmer, NAVAIR’s UAV program manager. The underlying common data link architecture is capable of supporting up to 45 Mbits/sec.
Creative Use of COTS
"The military is working hard to take a COTS approach," Whitmer says. Fire Scout is tweaking the MPEG (Motion Picture Experts Group) commercial data compression standard for data encoding and decoding. "This may be the first time anyone is trying to use MPEG in near real time," he says. "You can’t have a lot of delay when you maneuver a payload via a hand grip [on the ground] to keep the cross-hairs on the target."
The Fire Scout program also plans to use the Controller Area Network (CAN) bus, a technology common in the auto industry. The contractor, Northrop Grumman, intends to use CANbus to connect the flight computer to the actuators, says Tom Riley, the company’s chief engineer for the program. The actuators change the angle of attack of the main and tail rotors and control fuel flow to the engine.
Using CANbus saves weight because a wire can be run out to a CANbus node, which distributes the signals, avoiding the need for separate lines from the computer to each destination point. CANbus also allows developers to add built-in test (BIT) capability to the actuators, which wouldn’t be available in an analog system, Riley says. The program will employ the Mil-Std-1553 bus to interface with the vehicle’s three ARC210 radios. A commercial-standard RS-422 serial bus is used to connect the dual-redundant flight computers.
In our October issue, we look at the non-U.S. markets for unmanned air vehicles.
Micro Air Vehicles
The Defense Advanced Research Project Agency’s (DARPA’s) Micro Air Vehicle (MAV) program, a pathfinder for today’s Organic Air Vehicle (OAV) effort, proved the feasibility of building miniature unmanned vehicles.
A key player was AeroVironment Inc., Monrovia, Calif., which developed the Black Widow aircraft, a 6-inch (15-cm)-wide, propeller-driven flying wing. The craft features a 0.35-ounce (10-gram) avionics package, representing "the most mature, sophisticated avionics of any of the small planes," says Matt Keennon, the company’s MAV program manager. Two microprocessors serve as its main flight computer and two microactuators control tail surfaces (elevator and rudder). The UAV has a two-axis magnetometer for compass heading, a rate gyro for wing leveling, and a pitot-static probe for air speed and pressure altitude.
Black Widow has flown 1.1 miles (1.7 km) away from the ground station under 100% solid control. It can fly from 8 to 800 feet above the launch point. The UAV receives digital commands through a UHF/FM uplink. Bomb/chemical detection and law enforcement, as well as military applications, are possible.
Although the MAV program no longer exists, AeroVironment continues small vehicle R&D under other DARPA programs, Keennon says. Areas of interest include longer-endurance fuel cells and lithium polymer batteries.
AeroVironment also worked on the MAV program’s flapping-wing Micro Bat vehicle and built a 9-inch (23-cm)-wingspan flying prototype as a subcontractor to the California Institute of Technology, Keennon says. (The vehicle later moved to DARPA’s Micro Adaptive Flow Control program.)
This sparrow-sized UAV "looks and sounds like a bird when flying," he says. Flight time so far has been about 40 seconds with remote piloting through radio control. The 0.4-ounce (12-gram) prototype uses an 8-bit processor for its main computer and has a range of about 500 feet (152 meters).
Although this type of aircraft is still in the experimental stage, "in the very long term, micro-size UAVs will most likely have flapping wings," Keennon predicts. "As you get smaller and smaller, the operation of a propeller appears to be not the most efficient way of putting energy into the air."