Nearly all the UAVs described in the previous chapter had been avgas fueled. As we have already shown in Chapter Four, the radio-controlled electric model airplanes designed and flown by Fred Militky and the Boucher brothers in the early 1960s could be considered as hobby drones, whose targets were only altitude, speed and range. Then, in the early 1970s, development of electric and solar drones began in earnest
In the spring of 1973, Gerry Sayer, a founding partner in Developmental Sciences Corporation, developed the SkyEye, one of the first battlefield UAVs. He flew the first prototype R-4E-10 with a 2 hp pusher prop with a tricycle gear at El Mirage Lake, a dry lake bed in the northwestern Victor Valley of the central Mojave Desert, in California. They then tested the 10 hp electrically-propelled R-E-20.
Bob Boucher of Astro Flight has insisted:
In 1974, our patented model 7212 built for the military demonstrated the ability to carry a six pound payload (camera) for 1 hour and 20 minutes at speeds of between 50 and 75 mph (over 100kmhr). Model 7404 (Sunrise) proposal to the military was for a solar UAV to carry a 50 lb (23 kg) payload up to 90,000 feet and remain aloft indefinitely. These could not be considered model airplanes in that they were too large and too heavy to qualify (over 10kg gross weight). Flight tests took place at Camp Irwin, Bicycle Lake, California, in October 1974, and four with the 7404–02 at Nellis AFB in September 1975. As far as I know we were first in the world to build such vehicles. When we started Astro Flight we dreamed of building thousands of UAV for the military and competing with the giants of the industry like Boeing, Lockheed, Northrup [sic], etc. Alas this was not to be. But the motors I designed and built were incorporated in many surveillance UAV and in underwater vehicles.1
Between 1980 and 1982, the Flight Dynamics Laboratory (FDL) at Wright-Patterson AFB tested out electric propulsion for RPVs (remotely piloted vehicles) by adapting an XBQM-106, a tail-stabilized, pusher-propeller teleplane or mini-drone, normally powered by a Herbrandson DH220, 18 hp, 2-cycle gas engine. This was replaced by a samarium cobalt brushless DC unit developed by Timothy F. Glennon of Sundstrand Corporation in Rockford, Illinois. Two batteries were to be evaluated: a Honeywell lithium-thionyl-chloride battery (Li/SOCL2) and a Yardney silver-zinc alkaline battery, both of which differed considerably from the more familiar lead-acid battery and to a certain extent from other alkaline batteries such as nickel-cadmium and nickel-iron. Tests were conducted both at the Clinton County Airport, Wilmington, Ohio, and at Eglin AFB, southwest of Valparaiso, Florida.
On September 14–15, 1982, a workshop on aircraft electric secondary power was held at NASA’s Lewis Research Center in Cleveland, Ohio, to discuss the technologies related to aircraft power systems with a view toward aircraft in which all secondary power would be supplied electrically. This would later become known as an “all electric aircraft.” Among the speakers were Timothy F. Glennon of Sundstrand Corporation, who spoke about 400-Hz aircraft power generation systems, and Robert C. Webb of General Electric, who described the application of a cycloconverter to a permanent-magnet generator which could either be integrated with the engine or mounted on the accessory gearbox.
It was inevitable that when Paul MacCready built solar-powered man-carrying airplanes (see Chapter Five), he too would focus on UAVs or drones. A solar-powered UAV could in principle stay aloft indefinitely, as long as it had a power-storage system to keep it flying at night. The aerodynamics of such an aircraft were challenging, since to reach high altitudes it had to be much lighter per unit area of wing surface than the Solar Challenger, and finding an energy storage system with the necessary high capacity and light weight was troublesome as well.
Ray Morgan recalls: “After Challenger’s flights in Phoenix, a US Government Agency, whose name I cannot reveal, got interested in what we were doing and asked us at AeroVironment for what they called an atmospheric satellite to stay in orbit 12 miles above a certain spot. We did a study on paper and calculated that if we had a certain kind of airplane it could do it. It would mean a lot of wing area, low drag. They called it High Altitude Solar (HALSOL).” On paper, and based on projected technology, the mission could be achieved with an aircraft that had a span of 300 feet, and a wing-loading (weight divided by wing area) of approximately 1 pound per square foot. An approach for the design was laid out by Ray Morgan, Peter Lissaman, Bob Radkey and Bart Hibbs, which used a very flexible, aerodynamically pitch stable wing with distributed mass along the wing span as well as distributed overhung moments (the engine pylons) that matched aerodynamic moments along the span, thus minimizing structural weight. Analytically, it showed itself to be feasible with solar cells over 30 percent efficient and an energy storage system that weighed less than 1 kg for 600 watt-hours of energy stored. However, to prove out such a radical conceptual design, a series of progressively larger models were planned to be built and test flown.
A series of very small models were built and flight tested, some simply gliders and some powered. A total of six were built and tested before building the first that incorporated a sub-segment of the full-sized airplane. This first partial full-scale aircraft was a single 100-ft (30 m) wing (slightly longer than a Boeing 737) intended to be covered with 8 kW maximum of solar panels and a power source of six 16 1.50.6 kW electric motors, paired in parallel into 8 pylons and propellers, composing propeller speed pitch variations for directional thrust control, allowing differential thrust and drag to provide directional control. Its propellers and wing chord were the same size as the ultimate full-span aircraft would be. It was planned to fly larger span models, adding motor pylons as the span was increased.
It flew using silver-zinc batteries in tests, and a nominal solar array for evaluation. However, the intended lithium-based, regenerative fuel cell (for storing energy to fly overnight) and the so-called “ribbon” solar cells failed to meet the claimed performance, and the HALSOL program was retired after nine test flights—up to 3,000 feet (914 m) above ground and accumulating about 25 hours flight time in 1983 at a remote test facility in Nevada. The aircraft showed good aerodynamic performance and very good stability and control, but the power source could not meet the requirements. Consequently, the program was canceled after a corroborative study conducted by the Jet Propulsion Laboratory. The aircraft was mothballed and put in storage in 1984 in the Nevada desert.
Interest by Col. Dale Tietz in the Ballistic Missile Defense Office of the Pentagon and Doctors Lowell Wood and Nicholas Collela of the Lawrence Livermore National Laboratories (LLNL) in developing a Boost Phase Intercept (BPI) platform reawakened the concept about a decade later, about the same time as the HALSOL program was declassified, and the aircraft was offered to AeroVironment. Project Pathfinder was resurrected in 1992. This led to the de-mothballing, refurbishment and upgrading of the HALSOL airframe with digitally controlled, electronically commutated motors (derived from SunRaycer and Impact experience), improved propellers (no moving parts—with electronic motor commutation, props were changed to fixed pitch, and no gear reduction was required), and eliminated gear drives. To this were added improved and more complete solar arrays (although still terrestrial grade to save costs, initially), as well as a more modern, digital flight control (with autonomous navigation capability), stability augmentation, and datalink system with redundancy. LLNL invested in regenerative, hydrogen/oxygen fuel cell technology as an improved energy storage system as well as developing potential payloads for the mission. Pathfinder was successfully test flown at low altitude (below 500 feet) with terrestrial grade solar cells at Rogers Dry Lake Bed under NASA DFRC’s oversight in 1994. Unfortunately, funding priorities for BMDO changed in 1994, ending their support for Pathfinder. However, through the efforts of Dr. Nicholas Collela of LLNL, AeroVironment was able to move the airframe into a cooperative development program between NASA Aeronautics and industry, some twelve years later, for NASA’s Environmental Research Aircraft and Sensor Technology (ERAST) program.
This program was initiated by the acting associate administrator for aeronautics, Rich Christiansen, and the program manager was Jenny Baer-Riedhart at the Dryden Flight Research Center, co-located with Edwards AFB in California. The purpose of this program was to tap into existing efforts to develop unmanned, high-altitude aircraft which could provide atmospheric science measurements up into the stratosphere, with a target of reaching 100,000 feet, at which point the density is very similar to that of the atmosphere near the ground on Mars. A secondary goal was to demonstrate multi-day flight in the stratosphere. Three other UAV developers joined AeroVironment as “Alliance B Members,” but only AeroVironment was pursuing solar-powered flight to meet the mission. (It was intended that data gathered during this program would be shared with all participating members. This worked to varying degrees, depending on the individual company cultures.) On paper, based on projected available technology for solar cells and energy storage, a larger version (200 feet/60 m in span) with little or no energy storage, but 22 percent efficient solar cells, could reach the 100,000 feet altitude. This model was named Centurion. A 250- to 300-foot (70 to 90 m) version with high specific energy storage should meet the multi-day operation above 65,000 feet (20,000 m) goal. This model would be called Helios (Greek for the sun).
On September 11, 1995, Pathfinder set an unofficial altitude record for solar-powered aircraft of 50,000 feet (15,000 m) during a 12-hour flight from NASA Dryden. Due to its success, the Pathfinder was put on display at the Edwards AFB open house and air show the following month. It shared the hangar with two (then) classified aircraft, the B-2 Bomber and the F-117. Unfortunately, after the air show, when the Air Force decided to move out these two aircraft at night (for security reasons), there happened to be a windstorm sufficiently strong to blow over portable restrooms. When three of the hangar sides were opened for the entry of the aircraft tugs and towing of the two classified planes, the Pathfinder was caught in a wind gust of about 30 knots and blown into the F-117. Damage was severe. Much of the array and about half of the wing structure were destroyed, as well as several pylons. Fortunately, Dryden was able to come up with sufficient funds to repair the damage, and AeroVironment was able to upgrade the arrays replaced with more efficient SunPower terrestrial grade solar cells producing about 20 percent efficiency.
After further modifications, the aircraft was moved to the U.S. Navy’s Pacific Missile Range Facility (PMRF) at Barking Sands on the Hawaiian island of Kauai, to take advantage of the lower latitude, a longer flight season, and a much larger range, that would allow the team to maximize the altitude envelope by optimizing the sun angles on the array during the climb.
On June 9, 1997, Pathfinder made its first flight from Kauai and reached an altitude of over 67,000 feet (20,400 m). Then, on its second flight, July 7, 1997, Pathfinder raised the altitude record for solar-powered aircraft to 71,530 feet (21,800 m), which was also the record for all propeller-driven aircraft. It also demonstrated the ability to re-fly to the stratosphere after a two-day turnaround, and did two more flights over the Hawaiian Islands flying multi-spectral imaging and other environmental payloads over the littoral waters, inland forests, and agricultural fields of sugar and coffee.
Pathfinder was followed by Pathfinder Plus, with the wingspan increased another 20 feet by replacing the middle section of the wing with a longer span that had an airfoil designed for 100,000-foot altitudes (30,000 m), new, higher efficiency solar arrays, new propellers, and various avionics improvements, as well as more solar cells. The ERAST goal for Pathfinder Plus was to reach 80,000 feet and validate the new wing airfoil, new motors, and new propellers intended for the 100,000-ft goal. On 6 August 1998, Pathfinder Plus raised the national altitude record to 80,201 feet (24,445 m) for solar-powered and propeller-driven aircraft.
As part of NASA’s Environmental Research Aircraft and Sensor Technology (ERAST) program, on August 6, 1998, AeroVironment’s Pathfinder Plus raised the national altitude record to 80,201 feet (24,445 m) for solar-powered and propeller-driven aircraft (NASA).
This was followed by Helios, which, on August 14, 2001, remotely piloted by structural engineer Greg Kendall, set an altitude record of 96,863 feet (29,524 m)—the record for FAI class U (Experimental / New Technologies), and FAI class U-1.d (remotely controlled UAV: mass 500 kg to less than 2,500 kg). In fact, this altitude is higher than any non–rocket powered aircraft has flown in level flight to date. Ray Morgan retired from his position as vice-president of AeroVironment and director of the Design Development Center in 2000, but subsequently was hired as a consultant to NASA Dryden to serve as their technical representative for ongoing flight tests and demonstrations of the solar-powered aircraft developed for ERAST. At this point, to save money due to budget constraints, it was realized that both the high-altitude flight and the long-endurance flight could be achieved, and Centurion was changed to Helios (now 250' in span), after its initial, battery-powered flight tests at Dryden.
It had been initially planned that this long-endurance version of Helios would be modified with two to three regenerative fuel cells, comprising a hydrogen-oxygen fuel cell and an electrolyzer with gas storage tanks integrated into the tubular composite spars—resulting in a mostly span-loaded distribution of mass. However, funding limitations for the regen fuel cell system prevented reaching the goals before the program was slated to end in 2003. Consequently, AeroVironment, with NASA concurrence, compromised the design, and instead integrated an available automotive hydrogen-air fuel cell and two external hydrogen gas storage units, estimated to be sufficient to demonstrate nighttime flight, with a two-day total flight endurance in the stratosphere. This required placing a single, 500-lb (220-kg) fuel cell system in the center of the wing, with composite hydrogen gas tanks mounted near each tip to counterbalance this point load, which was beyond all experience to date with the span-loaded flying wing design. Further, it increased the gross takeoff weight about 40 percent. It was hoped that modifications to the flight control system could compensate for this change in load distribution and relative stiffness.
On the first flight of the Helios in this configuration, the aircraft seemed to be reasonably stable and controllable in benign conditions, but a leak in the gas handling system forced an early landing. On its second flight (June 26, 2003), the winds were much stronger than prior experience. Because of its slow climb rate at this higher gross weight, it encountered a much stronger shear line as it exited the wind “shadow” of the mountains of Kauai about 2,000 feet (600 m) lower than previous flights, which, coupled with the highly center-loaded wing, caused the tips to bend up much higher than the center wing, resulting in an oscillation in airspeed and pitch. A relatively inexperienced pilot erroneously turned off the airspeed hold system (thinking it had failed), rather than increasing the gain of that system (as planned in the emergency procedures), resulting in an unmitigated dive, over-speeding the aircraft and ripping off the leading edge, causing the Helios to crash into the waters between Kauai and Niihau, ending the program.
A follow-on flight research program funded by NASA, outside the ERAST goals, evaluated the effects of turbulence on the Pathfinder Plus, still in flight status, during low altitude flights at Rogers Dry Lakebed (part of the Edwards/Dryden complex). The Pathfinder Plus was outfitted with very sensitive turbulence-measuring devices and also load-measuring devices for the airframe for these tests, which were conducted in 2005. Morgan was able to invite Captain Steve Ptacek (previously one of Solar Challenger’s pilots, now flying 777s) back to ride in the chase car during these low-altitude flight tests. Subsequent to these last flights, the Pathfinder Plus was hung in the Udvar-Hazy Air and Space Museum at Dulles Airport outside Washington, D.C.
In 2005–6, Ray Morgan participated as a subject matter expert in a NASA Langley-led study considering options available with near-term technology for flying above hurricanes to take measurements, track the eye, and predict the track more accurately. The conclusion was that current technology readiness levels for solar arrays and energy storage systems were inadequate to allow an airplane to maintain position above hurricanes through the required months, even to just 30 degrees of latitude, although a solar-powered blimp may be able to do so (provided the ground speed was sufficiently low).
In early 2008, Morgan teamed with Dr. Vince Castelli to convince Defense Advanced Research Projects Agency to create a program to study potentially available (3- to 5-year time frame) technology advances that would permit an essentially geostationary stratospheric satellite to perform militarily useful missions in the future. The program was accepted by the head of DARPA, Dr. Tony Tether, and Morgan was tasked to lead an internal feasibility study before finalizing requirements for a Broad Area Announcement (a solicitation for the aerospace community to propose concepts). The result of this feasibility study showed that it may be possible to hold station against probable stratospheric winds year round approximately 95 percent of the time up to about 35 degrees of latitude. In order to make it “DARPA-hard,” however, Dr. Tether required that the announcement ask for a target of 99 percent station keeping up to 45 degrees of latitude year-round for a total flight time of 5 years on a single aircraft. Multiple companies bid, with three participating in the phase I conceptual design study—Aurora Flight Sciences, Lockheed-Martin, and Boeing. This pushed the design flight speed requirement to around 80 knots, the solar cell efficiency requirements to around 40 percent, and the specific energy and round-trip efficiency of the batteries (energy storage system) beyond state of the art. In Phase 2 only one performer was still funded, Boeing, which had the highest probability of meeting the energy storage requirement, but funding was reduced dramatically, turning the project from a system design to a technology advancement study. In the fall of 2017, Boeing purchased Aurora and its autonomous LightningStrike XV-24A technology in the race to develop air taxis.
Alongside this, AeroVironment had been working on the development of a small hand-launched remote-controlled unmanned aerial vehicle (or SUAV). Named the RQ-11 or Raven, this was a half-scale derivative of the Pointer, which had been developed under Ray Morgan’s direction in the 1980s:
I’d first proposed its development internally to Paul McCready in 1984, based on then available technology and a design derivative of one of Martyn Cowley’s world-class, competitive free-flight glider designs that was both very efficient, inherently stable, and capable of “deep stall” landings. In 1986, we got approval from Peter Lissaman (who was technically my boss then, before I was promoted to VP) to develop a prototype with a black and white video camera on board, Ni-Cd batteries driving one of Boucher’s electric model motors, and about 9 feet in span. The airplane came apart in six, plug-together pieces, that, with its ground control and video display fit into a single plastic backpack that I then took to Washington, D.C., and flew demos for various government organizations and military folks, including walking through the halls of the Pentagon carrying the fuselage in my arms while a lt. col. watched the video at his desk. After about a year of doing this, we got an order for four aircraft and a ground station from an organization that wished to stay unknown, as well as funding to build 6 airplanes and two ground stations for a demo system to be evaluated by the Marines over about a year. In 1988, we let a couple of enlisted Marines demo the system to the Marine Commandant, including letting him fly and land it with one minute’s instruction, which resulted in a million-dollar contract to build forty aircraft and 8 ground stations (with militarized backpacks) for several years of test and evaluations by Marines, Army, DEA, and other governmental organizations. In the mid ’90s, we knew the state of the art for batteries, motors, but, especially the payloads (IR, EO, night-vision cameras) and navigation systems would allow reducing the size of the Pointer by about a factor of 2, but lacked internal funds to develop it on our own, and couldn’t get any more from outside. Some of these were taken to the middle east for the first Gulf War.2
The Raven made its first flight in October 2001. It can fly up to 6.2 miles (10.0 km) at altitudes of approximately 500 feet (150 m) above ground level (AGL), and over 15,000 feet (4,600 m) above mean sea level (MSL), at flying speeds of 28 to 60 mph (45 to 97 kph). The Raven was the winner of the U.S. Army’s SUAV program in 2005, and went into Full-Rate Production (FRP) in 2006. Shortly afterwards, it was also adopted by the U.S. Marines, and the U.S. Air Force for their ongoing FPASS Program. After 9/11, they started buying Pointers, and then AV got the internal funding for developing the Raven, with improved performance and capability and reduced size and weight, in extreme quantities. It has since been adopted by the military forces of many other countries. More than 25,000 Raven airframes have been delivered to customers worldwide to date. Alongside the Raven, AeroVironment of Monravia, led by Kirk J. Flittie, developed a family of military drones: RQ-12 Wasp, RQ-20A Puma, and Shrike VTOL.
Perhaps one of the most remarkable UAV programs began in 2001 when the British Ministry of Defense (MOD) split its Defense Evaluation and Research Agency (DERA) in two. DERA was privatized and renamed QinetiQ (as in Kinetic). One of QinetiQ’s first projects was to find a way of filming an attempt to break the world altitude record in a balloon. Chris Kelleher was chosen as designer, technical director and flight operations manager to lead the Farnborough-based team.
Born in 1958, Christopher Charles Kelleher was as keen on aeronautic research as on flying aircraft. The son of a wartime RAF pilot, he studied at Queen Mary University of London, and also attended at the RAF College in Cranwell. A member of the RAE Aero Club, he flew light aircraft as a hobby and in occasional air displays, being a three-time winner of the British Aerobatics Association Advanced Level. He was also an Approved Person for airworthiness flight tests (including prototype). At the same time he had also worked on the UK’s 4 Skynet military communications satellites, and also on two spacecraft for NATO, developing himself an international reputation for orbital dynamics and satellite operations used by both the UK and U.S. governments. So he understood the potential for a loitering stratospheric vehicle that could be a low-cost complement or alternative or “an eternal solar platform.”
Bringing together a 20-strong team, Kelleher now began to work on a series of UAVs named Zephyr. For Zephyr 2, a proof-of-concept aircraft weighing less than 7 kg flew in both free and tethered modes off the Clifton suspension bridge in Bristol. By 2002, a 12 m (40-ft.)-long, 15 kg (33 lb.) UAV called Zephyr 3 had been created. It was supposed to fly tethered to a manned balloon attempting a world record altitude of 132,000 ft. (4,023 m), but unfortunately the balloon had a technical problem, and thus neither the balloon nor the aircraft ever flew. Following on from this, the Zephyr 4 project was created in order to carry out critical technology risk reduction and concept development work. Zephyr 4 also had a 12 m wingspan, but weighed approximately 17 kg, and was designed to be launched by the use of a helium balloon. In February 2005, Zephyr 4 underwent a test flight in Woomera, South Australia, where it flew for one hour after being launched at 30,000 ft. (9,000 m) by the balloon. In December 2005, Zephyr 5 flew in New Mexico for four hours, then for six hours, demonstrating successful ground launch, ascent, cruise and descent. In July 2006 the two aircraft were flown again in the USA with Zephyr 5–1 reaching 36,000 ft. (11,000 m) on an 18-hour flight, including 7 hours overnight. Collectively the two aircraft had a flying time of 35 hours. Next came Zephyr 6 with an 18-meter wingspan but weighing less than 30 kg (66 lb.) as it was constructed out of ultra-light carbon fiber. In July 2007 Zephyr 6 flew for 54 hours in New Mexico at a peak altitude of over 58,000 ft. (17,000 m), thus flying in the targeted 50,000–60,000-ft. range and validating the thermal and performance models. In August 2008, Zephyr 6 flew for over 87 hours above Yuma, Arizona, reaching an altitude of 61,000 ft. (18,600 m) and surpassing the previous world record of 30 hours for the longest unmanned flight, set by Global Hawk in 2001. This was homologated by the Fédération Aéronautique Internationale.
On July 9, 2010, monitored as usual by Chris Kelleher and his team, the Zephyr 7 took off from the Yuma Proving Ground and stayed airborne for over 2 weeks (336 hours), at an altitude of 70,742 feet (21,562 m), setting new world records, and was dubbed the first “eternal plane.” Of this flight, Kelleher observed, “Zephyr will transform the delivery of current services such as communications and lead to many new applications that are not possible or affordable by other means.”
Since 2008, Airbus had been working with its subsidiary Astrium on High Altitude Pseudo Satellites (HAPS). In 2013, Astrium acquired the Zephyr assets from QinetiQ, integrating it into the Airbus High Altitude Pseudo-Satellite program. As further proof, in 2014, an 11-day flight was launched from British-controlled Ascension Island in the South Atlantic during equal day–equal night conditions just south of the equator. This explored the ability of the Licerian battery packs provided by Sion Power to store enough solar energy to power the Zephyr through the hours of darkness, without significant loss of altitude. This was also the first flight controlled by a satcom datalink. There was also a flight in 2014 from Dubai that explored the Zephyr’s integration into civilian airspace. The UAV was equipped with Mode S ADS-B so that it could report ascending through a “bubble” of airspace that other traffic could avoid. The Zephyr 7 takes up to 12 hours to reach its operating altitude above the jet stream. It has almost no ground speed when flying into wind. When on station at 65,000 feet (20,000 m) or above, it loiters by GPS control. It is hand-launched by a five-man ground crew, and recovered by belly-landing. Zephyr 7 has flown for over two weeks continuously—eight times longer than any other UAS.
Tragically Chris Kelleher, the man behind the Zephyr program, died on August 22, 2015, of natural causes while cycling near his home in Church Crookham. Three months later, in November 2015, in the House of Commons, British Prime Minister David Cameron laid out plans during the 2015 Strategic Defence Review (SDR) to further enhance Great Britain’s intelligence, surveillance and reconnaissance (ISTAR) capacity. In the speech, he stated that the UK was to field a “British-designed unmanned aircraft that will fly at the edge of the earth’s atmosphere and allow us to monitor our adversaries for weeks on end, providing critical intelligence for our armed forces.” The British Ministry of Defence later confirmed the purchase of three Zephyr 8 platforms. The Zephyr 8 aka Zephyr S pseudo-satellite would have roughly 24 kg (53 lb.) of lithium batteries and a 5 kg (11 lb.) payload, and be 30 percent lighter. At an altitude of 20 km the aircraft has a visual range of over 400 km, offering a data collection and high bandwidth communications relay capacity to areas in excess of 1,000 km2. The UK MOD gave Airbus D&S a $14 million (£10.6 million) contract to provide two larger UAVs for an operational concept demonstration in 2015. Further still, the Airbus Zephyr T, with its twin tails to support the additional weight, would weigh up to 140 kg (309 pounds) and carry a payload of 20 kg (44 pounds) on a 33-meter (108-foot) wingspan. Full-scale flight testing is scheduled for the Zephyr T in 2018 aimed at providing a maritime surveillance and communications capability. The lithium sulphur (Li-S) batteries being developed by OXIS of Abingdon, Oxfordshire, for the Zephyr program have a theoretical energy density five times greater than lithium-ion, with lighter weight and longer life-cycle characteristics. If OXIS arrives at the specific energy target of 425Wh/kg, the Zephyr will be able to fly above any weather in the troposphere and remain aloft for three months without needing to land. All the various Zephyr demonstrations to date have logged more than 1,000 hours.
In 2015, Bill Fredericks and a team at NASA’s Langley Research Center in Hampton, Virginia, developed the GL-10 Greased Lightning, an unmanned hybrid-electric aircraft with a wingspan of 10 feet (3 meters), that can swivel its wings and engines—into the vertical position for vertical takeoff and landing (VTOL), and then horizontal for conventional forward “wing-borne” flight. GL-10 made its first tethered flight on August 19, National Aviation Day, performing transitions between vertical and horizontal flight. Its VTOL could be used for small package delivery, or long-endurance surveillance for agriculture, mapping and other applications. The plan is to demonstrate that the GL-10 is four times more aerodynamically efficient in cruise than a helicopter.3 A full-scale version would involve 26 hybrid-electric propulsion fans distributed on the aircraft. In July 2017 the Greased Lightning concept was licensed to Fredericks’s Virginia-based startup, Advanced Aircraft Co. (AAC); its first product, the hybrid-electric 65-inch (165-cm) octocopter Hercules is currently in its flight-testing phase, but with the ability to fly for 19 hours with a 5-pound payload. A two-stroke gasoline engine balanced by battery power provides fail-safe power supply to ensure safe flight operations. Deliveries of the Hercules were planned for December 2017.
A team led by Gregory W. Walker at Silent Falcon UAS Technologies in Albuquerque, New Mexico, has developed a thin film photovoltaic (TFPV), carbon-fiber fixed-wing, long-range unmanned aerial vehicle (UAV) with the ability to stay in the air for extended periods of time—up to five hours depending on flying conditions. Once the Silent Falcon reaches 100 meters, it’s effectively undetectable.
On September 14, 2017, the Tailwind hybrid UAV developed by Troy Mestler, Robert Karol and Ivan Qiu of Skyfront, Menlo Park, made a flight of 4 hours and 34 minutes in winds that were between 8 and 10 mph, with gusts up to 15 mph. Menlo Park, of course, was once the headquarters of Thomas A. Edison, who had envisaged electric flight some 130 years before.
Alongside their primary role in high-altitude observation, UAVs have recently entered the realm of delivering goods.
In December 2013, German postal and logistics group Deutsche Post DHL used their Parcelcopter, a Microdrone md4-1000 drone, to deliver medical products from a pharmacy across the Rhine River. It was the first civilian package delivery via drone. A more challenging location was next selected for a 2014 trial: the North Sea island of Juist, again to be supplied with time-sensitive goods and urgent medicines. During early 2016, DHL directly integrated a parcelcopter logistically into its delivery chain for the Bavarian community of Reitim Winkl.
From October 2016, organized by Arthur Draber of Zipline, a Californian-based start-up, drones began regularly delivering batches of different blood groups to an increasing number of village hospitals in southern and western Rwanda, up to a range of 150 km (90 mi). Following a command by SMS to Zipline’s Gitarama base, each trip was followed on electronic charts with deliveries arriving within a predictable radius of 5 meters. In early 2018, Zipline’s Nest 2 would go into service for 20 hospitals located in eastern Rwanda, while negotiations were in hand for deliveries in Costa Rica and Tanzania.
As for fast-food deliveries: on May 11, 2014, Francesco’s Pizzeria used a drone to deliver a pizza to a friend of the restaurant CEO on an apartment rooftop in Mumbai, India. A Russian pizza chain called Dodo Pizza conducted six drone pizza deliveries on June 21, 2014, but was later fined by Russian authorities for the illegal flights. Matt Sweeney of Flirtey in Nevada has organized Smartphone-related drone delivery service. The drone carries items weighing up to 5.5 lbs. (2.5 kg) on round-trip journeys of up to 10 miles (16 km). To avoid hitting trees or power lines, they hover about 40 to 50 feet (12 to 15 m) off the ground and lower their deliveries by cable to the waiting customer below. The cable-lowering method also keeps customers at a safe distance from the drone’s whirling rotors. In November 2016, Domino’s Whangaparaoa store in New Zealand used a drone to deliver a piri-piri chicken pizza and a cranberry chicken pizza to a local customer. And in the United States, 7-Eleven said in July 2016 that it had delivered Slurpees, doughnuts and other food to a customer in Reno, Nevada. Google has tested drone delivery of chipotle burritos on the Virginia Tech campus. Among those bigger giants seeing the advantages of drone delivery are companies like Amazon, Google, FedEx, UPS, and DHL. In September 2016, United Parcel Service (UPS) used a CyPhy drone to deliver a medical inhaler to an island near Boston.
From November 2016, one of China’s biggest e-commerce companies, JD.com, founded by Liu Qiangdong as Jingdong Mail and headquartered in Beijing, had a fleet of drones flying autonomously on round trips of a maximum of 15 miles to reach rural communities (though a person still takes the package on the last leg of its journey to the recipient). Developed via its in-house drone team called JDX, the drone delivery program has so far licensed more than 30 flight routes in four different bases: Beijing, Jiangsu Province near Shanghai, Shaanxi Province in northern China, and Sichuan Province. Since China’s Air Force controls the air space, JD.com is leasing the drone flight routes from the government on an annual basis and has to notify officials for each flight. When a rural customer orders goods online, the order is flagged when it is suitable for drone delivery. Once the parcel reaches the delivery station, a team fixes it to a drone that will follow a fixed route avoiding populated areas, highways and signal towers. Most can travel 10 km in one trip. During a recent flight test with the media, the model used, called M-TC2, was able to travel at a maximum speed of 100 kph. The drone hovers above a designated area in the destination village, drops the parcel down, and leaves immediately after. Fragile items are clearly not suitable for drone delivery. JD has invested CNY 20.5 billion to implement an in-depth cooperation with Xi’an National Civil Aerospace Industrial Base at Shaanxi over the next five years, to develop and manufacture seven different types of delivery drones. These range from drones that can fly up to 60 mph (100 kph), delivering packages weighing from 5 kg to 30 kg (10 lb. to 65 lb.) to drones which can carry as much as 1,000 kg (2,200 lb.), or one metric ton, with a maximum distance of approximately 100 km (60 mi) before recharging. JD, with its own nationwide network of thousands of delivery stations manned by 65,000 employees and 235 million regular customers, will establish three headquarters, three platforms, and four major industries. The three headquarters include JD’s global logistics headquarters, global unmanned system industry headquarters, and JD Cloud’s Shaanxi big data operating headquarters. The three platforms include a fusion smart logistics platform, a flight transport platform, and a big data operating platform. The four major industries cover smart manufacturing, smart logistics, cloud computing, and “characteristic town.”
On December 7, 2016, Amazon made its first commercial drone delivery of an Amazon Fire streaming device and popcorn to a customer identified only as Richard B., in Cambridgeshire, England. The flight took off from a nearby Amazon warehouse and lasted 13 minutes, covering about two miles.
At the end of September 2017, Matternet, based in Menlo Park, California, began to deliver toothbrushes, deodorant and smartphones 8 to 16 kilometers (5 to 10 miles) above Zurich, Switzerland (population 391,000), to awaiting Mercedes-Benz electric delivery vans in populated areas such as congested urban streets and beyond natural barriers such as Lake Zurich. Customers ordered products using an on-line commerce startup called Siroop. The initiative has been approved by Switzerland’s aviation authority, as has a separate Matternet project to carry medical supplies between Swiss hospitals. At PACK EXPO 2017 in Las Vegas, PMMI (Association for Packaging and Processing Technologies) collaborated with electric truck and HorseFly octocopter maker Workhorse to give hourly demonstrations of how, across a virtual landscape of 15,000 square feet, they would be delivering packages up to 10 lb. from stationary electric vans to the mailboxes and doorsteps of mock houses.
A team led by Przemyslaw Kornatowski at École Polytechnique Fédérale de Lausanne (EPFL) in Switzerland has developed a delivery-focused drone with an origami-inspired foldable carbon-fiber cage that protects the aircraft and its cargo in case of a collision or fall. The origami design means that the frame can be folded and unfolded in a single movement, reducing the volume by 92 percent, and can conveniently fit in a rucksack. Furthermore, the quadcopter can take off and land vertically and carry a package weighing up to 500 grams (1.1 pounds) over a distance of 2 kilometers (1.24 miles). The rounded protective cage is not only safe for regular handling of the cargo drone, but is also useful in emergency situations, where there is no landing spot, where instead the drone can be safely grabbed as it approaches the recipient. The EPFL team completed more than 150 test flights around the institute’s campus and is already planning improvements such as an emergency parachute for breakdowns.
Indeed, in July 2016, Amazon.com filed a patent on “docking stations” for its delivery drones that will be built on tall structures such as lampposts or churches as drone “perches” which will allow the unmanned machines to recharge and pick up packages.4 Another Amazon patent is for delivery whereby the drone would deploy packages at altitude and monitor and adjust (radio-control) package trajectory during descent using either a parachute, a compressed air canister, or a landing flap. The package could carry multicolored markings so that a camera on the drone could tell exactly how it’s oriented on the way down.
In the fall of 2017, Amazon opened a new 60,000 sq. ft. (5,500 m2) site at One The Square in the CB1 business district of Cambridge to research artificial intelligence (A.I.) and drone delivery. Among the 400 mathematical experts, scientists and engineers, teams working on products such as Kindle, Echo and Amazon’s voice assistant Alexa also moved into the new Cambridge location. Amazon partnered with the UK government to test drone deliveries and recently designed a UAV with robotic wingtips and legs that act as landing gear to help them touch down on uneven surfaces.
In October 2017 Amazon patented a system whereby a drone could deliver battery energy to an uncrewed electric automobile running short of energy (U.S. 9,778, 653 B1). Servers relay energy requests to drones, which would then be dispatched to meet vehicles on the road. The drones would authenticate the connection, attach to a vehicle connector on the roof or door, and start fueling—even if the car is in motion. The passenger could even make the help call to Alexa, Amazon’s personal assistant. The battery would be independent of the drone’s own battery supply to fly back to base.
Skysense, originally from Trento, Italy, has developed a drone-charging pad made up of modular 50 cm2 pads, a tessellation of conductive tiles that can be dynamically activated. Whenever a drone lands, two spring-loaded contacts power the on-board Li-Po balance charger to fast recharge the batteries. The pad can be plugged into an electrical socket or powered by solar panels, and works even in rain and extreme weather conditions.
In an approach to resolve the issue of recovering the delivery module, Saul Griffith and an ex–MIT team at Otherlab engineering laboratory, San Francisco, have developed an industrial-grade paper airplane as a low-cost disposable aerial delivery vehicle based on DARPA’s ICARUS program (for Inbound, Controlled, Air-Releasable, Unrecoverable Systems). They call this APSARA (for Aerial Platform Supporting Autonomous Resupply Actions). Once the goods have arrived, the drones biodegrade in a matter of days. And because it is a glider without motors and rotors, it means that all of the onboard electronics, courtesy of DARPA’s VAPR program, go with it. Whilst it has its commercial application, several hundred APSARA gliders loaded with critical medical supplies could be used in emergency zones.
But beware! As with any other form of transport, according to a report by Intel’s McAfee Labs, cybercriminals are threatening to target regular drones used for deliveries, law-enforcement or camera crews, in a crime known as “dronejacking.” Once a package delivery drone is overhead, the drone could be sent to the ground, allowing the criminal to steal the package. The culprit could also steal expensive photographic equipment carried by drones, in order to knock out surveillance cameras used by law enforcement. Dronejacking toolkits are already being traded on “dark web” marketplaces.5
Counter-drone strategy involves detection at up to 200 yards (183 m), interception and neutralization. This can be done by training birds of prey or using mobile jamming systems, stationary jamming systems, drone interception systems, drone capture nets, or anti-drone lasers. Detection methods could use both radar and acoustics to detect and identify the flying drone, displaying the flight speed and direction of the drone. The challenge is to detect the right object, because efficient radar should identify the drone as not a bird. As most drones use the same 2.4 GHz frequency that is used in telecommunications like Wi-Fi, a system of selected jamming has been innovated.
Ever since Capitaine Max Boucher’s experiments with radio-controlled aircraft during World War I, the French have sustained a fine interest in model aircraft. Since childhood, Pascal Zunino and his friend Fabien Paganucci of Les Pennes-Mirabeaux, in the Bouches-du-Rhône Department of southeast France, enjoyed building electro-mechanical gadgets. Both attending the local la Renardière primary school in the 1990s, they read an article in an aeromodeling magazine about a quadcopter, and decided to build one themselves in the Zunino garage. Moving on to the Jacques Monod College, they continued to work together on projects, with the mechanics developed by Fabien and electricity, then electronics, by Pascal. At the Pierre Mendès France secondary school in Vitrolles, Zunino and Paganucci assembled a team to compete in the E=M6 robotics competition. They called their robot “Glouton” (Glutton). The circuit boards built were used to liaise between the robot and the control unit through a PWM modulation, and so command the two propulsion engines and that of the suction turbine. Winning, they were officially presented with the 2000 E=M6 Trophy at the City of Science and Industry in Paris.
In 2001, while Pascal Zunino studied electronics at the University of Aix in Marseilles, Fabien Paganucci studied microtechnology and computer-aided architectural design at the same city. They assembled the red micro-robot Z6 prototype drone, which made its very first flight in the early hours of January 25, 2001. By 2002, they had progressed to their more spindly 70 g (2.5 oz) carbon fiber X4. Its electronics brought together on one card an 8051 microcontroller, a 3-gyro inertial unit, and 4 speeds. They also built an 80 g (2.8 oz) Mini X4 with a 34-cm (13-in.) wingspan and an even smaller 40 g (1.4 oz) Micro X4 with a 16-cm (6-in.) wingspan. This gave them confidence to assemble a team of students at Grenoble INP and to build the foldable-rotor CPX4 for the international miniature drone competition organized by the ONERA and the DGA that September 2005. At the Mourmelon Military Camp near Châlons-en-Champagne, they demonstrated the CPX4 to civil and military professionals in the industry and won 1st prize ex-aequo. It had particularly attracted the attention of the DGA (Délégation Générale de l’Armement, the state organization responsible for armament programs) as being operational and innovative due to its size and previously unseen ease of use: with its hovering capability, the CPX4 could also be very easily fitted with sensors or cameras (day/night, infrared).
In 2006, financially backed by the Michel Caucik Innovative Business Incubator, they created the start-up Novadem Aerial Robotics in Aix-en-Provence with the aim of democratizing aerial robotics. In this they were joined by Pascal’s brother, Eric Zunino, himself specializing in information technology, and who had often helped with previous prototypes. Before long, their enterprise came to the attention of Henri Seydoux of the Parrot Company. In 1994, Seydoux (son of the head of the Pathé-Gaumont Cinema empire, Jérôme Seydoux), together with Christine de Tourvel and Jean-Pierre Talvard, had set up Parrot to create a vocal recognition system for the blind in the burgeoning smartphone and tablet market.
To diversify their business, Seydoux was looking into a Wi-Fi remote-control automobile. The product was ready for production. But it was not fun enough. Seydoux was about to cancel the project when he got the idea to turn the car into a drone, formed a team called Commando, and contacted Zunino and Paganucci at Novadem, who would help Parrot to develop various key aspects of the drone: propeller design, control system and stabilization algorithms, ultrasound telemeter, benchmarks, noise reduction, and testing procedures, while Parrot focused mainly on video. That was the beginning of a time-saving collaboration that would last for three years. The augmented reality AR. Drone was officially unveiled at the 2010 Consumer Electronics Show (CES) in Las Vegas as the first Wi-Fi augmented reality quadcopter drone, a form of flying video game. It could be controlled by mobile or tablet operating systems such as the supported iOS or Android within their respective apps or the unofficial software available for Windows Phone, Samsung BADA and Symbian devices. The drone was loaded with sensors and video cameras. It had debuted just in time to surf the wave of the famous miniature remote-control helicopter PicooZ. In fact, Parrot was not just catching the wave but was extending it. The AR.Drone was launched onto the market just a few months later, with a global launch beginning in August 2010, starting in Hong Kong and France before spreading to the rest of the world. This initial product was a major success: 120,000 AR.Drones would be sold by the end of the year.6
2010, left to right: Fabien Paganucci, Eric Zunino and Pascale Zunino. Pascal Zunino founded Novadem Aerial Robotics in Aix-en-Provence with the aim of democratizing aerial robotics (NOVADEM).
Along with AR.Freeflight, the application designed for free operation of the drone, Parrot also released AR.Race, allowing users to take part in solo games or interact with other drones in combat simulations. Inside the airframe, a range of sensors assist flight, enabling the interface used by pilots to be simpler, and making advanced flight easier. The onboard computer runs a Linux operating system and communicates with the pilot through a self-generated Wi-Fi hotspot. The onboard sensors include an ultrasonic altimeter, which is used to provide vertical stabilization up to 6 m (20 ft). The rotors are turned by 15-watt brushless motors powered by an 11.1-volt lithium-polymer battery. This provides approximately 12 minutes of flight time at a speed of 5 m/s (11 mph). Coupled with software on the piloting device, the forward-facing camera allows the drone to build a 3D environment, track objects and drones, and validate shots in augmented reality games.
In 2012, Parrot acquired 57 percent of the Swiss aerial imaging drone manufacturer of senseFly4 drones and invested more than 2.4 million Swiss Francs in Pix4D, both in Lausanne. In 2015 Parrot invested in drone startups: Airinov, drones for agriculture; Micasense, sensors for agricultural drones; EOS Innovation, robot for inspection; Iconem, drones for archaeology. The same year, Parrot separated its core business into two distinct subsidiaries: Parrot Drone and Parrot Automotive.
2010: Henri Seydoux of the Parrot Company launches the AR.Drone; 120,000 would be sold by the end of the year (PARROT).
In May 2014 at the annual AUVSI conference in Orlando, Parrot announced the AR.Drone 3.0, code-named Bebop, with the option for a Skycontroller, when purchasing the Bebop. The Skycontroller allows the Bebop Drone to fly up to 2 km (1.2 mi). The Parrot Bebop Drone was released in December 2014. In almost a man’s world, Christine Caubel, mechanical design engineer at Parrot Drones near Paris, has led several innovations for her firm since 2008. In 2015 she gave quadcopters protection by providing them with removable bumpers for protecting their propellers. Each bumper is connected to the propulsion units on the same side by connection arms, each comprising a pair of elastically deformable elongated blades with a clamp at their end for mounting on a barrel of the motor.
There were many who preferred building their own drones to buying them off the shelf. In 2006 London-born Chris Anderson, then editor of Wired magazine, built a drone for his kids for fun. In 2007 he turned his new hobby into a Web-based community called DIYdrones.com to trade hardware tricks and software tips in forums; during the next decade it would grow to over 70,000 members. In October 2007, Anderson flew a remote-controlled aircraft allegedly equipped with a camera over Lawrence Berkeley National Laboratory, causing security concerns when the aircraft crashed into a tree. In 2009 Anderson met Mexican Jordi Muñoz online and together they formed 3D Robotics in Berkeley, California, to design and build UAVs.
In Italy in 2008, Paolo Marras and his team founded Aermatica SpA in Gironico, Como, with the goal of developing solutions to capture data and generate enhanced information through the use of small aircraft to pilot remote systems (drones, UAV or UAS). They developed the variable-pitch Anteos quadcopter and obtained official permission from the Italian Civil Aviation Authority (ENAC) to fly in the nonsegregated civil airspace. They patented their “Aerial Robotic System” in 2009 (WO 2010128489 A3).
A rival to Parrot is DJI, directed by Frank Wang. In 2004, Wang’s dream of landing at an elite American university such as MIT or Stanford was thwarted due to his unimpressive academic performance, so he ended up at the Hong Kong University of Science & Technology, where he studied electronic engineering. In 2006, during his senior year, working in his dorm room, Wang developed a helicopter flight-control system. Even though the hovering function for the on-board computer failed just before the presentation, Li Zexiang, his robotics professor, noticed him and brought him into the school’s graduate program.
In 2006, Wang and two classmates moved to a manufacturing hub of Shenzhen, Guangdong, China, and started working out of a three-bedroom apartment. Wang funded the venture Shenzhen Dajiang Innovation Technology, or DJI, with what was left of his university scholarship, and modest sales (to Chinese universities and state-owned power companies) were enough to help him pay for a small staff. In January 2013, DJI produced their white Phantom 1 drone. It was a multirotor that most people could order online and fly without much trouble after reading the manual. It was commonly equipped with a GoPro camera for amateur filmmaking or photography. Its battery life was around 15 minutes with a GoPro.
Eleven months later, DJI released the Phantom 2. Upgrades included auto-return, increased flight speed, increased flight time and controllable range, increased battery capacity, compatibility with smartphones and tablets, and a feature called Smart Guidance Control (or IOC), that made the Phantom 2 ridiculously simple to fly. By the end of 2014, DJI, with a 2,000-strong workforce, had sold an estimated 400,000 units, making Frank Wang the world’s first drone billionaire.
The same year, the American Federal Aviation Administration (FAA) proposed some regulations and guidelines to ensure proper operation of these vehicles in the USA, and the regulations came into effect on December 21, 2015. As CES 2017 was opening, news broke that DJI had bought a majority stake in the German camera manufacturer Hasselblad—the company that made the cameras American astronauts took to the moon. Soon after, DJI departed from its white Phantom range to present its “H” drone, a black hexacopter with ST16 remote control with Android software and 25-minute autonomy.
Lorenz Meier, a Ph.D. student in electronics at the ZTH Swiss Federal Institute of Technology Zurich, was curious about technologies that could allow robots to move around on their own, but in 2008, when he started looking, he was unimpressed—most systems had not yet even adopted the affordable motion sensors found in smartphones. So as part of what he called the “sFly Project: Swarm of Micro Flying Robots,” Meier built his own system instead: the Pixhawk flight stack platform for autonomous drone control. In 2009 he launched, the results inspiring several drone companies such as Parrot, DJI, Yuneec, and Skydio. Importantly, Meier’s system aimed to use cheap cameras and computer logic to let drones fly themselves around obstacles, determine their optimal paths, and control their overall flight with little or no user input. In 2011, he open-sourced it under the PX4 software brand following through with the Pixhawk open hardware autopilot in 2013.
The following year, the Dronecode Opensource UAV Platform, which is a nonprofit organization governed by the Linux Foundation, was formed by Chris Anderson with the goal of using open-source Linux for the benefit of founding members including Qualcomm and Intel. 3D Robotics moved along with Yuneec International providing cheaper, better, and more reliable UAV software. For example, the Yuneec DataPilot™ software, integrated in the new H520 hexacopter, is tightly coupled to the PX4 flight control architecture. The DataPilot™ is a complete solution for planning survey and waypoint-based UAV flight. Over 500,000 units of Pixhawk have been manufactured. In February 2017, Meier helped Yuneec to establish its R&D (research and development) center in Zurich, and since then has become an advisor to the company in high-tech development. In September 2017 Meier was named as one of top under-35 innovators by MIT Technology Review.
Although the drone was born by scaling down technology, during the past few years the technology of the drone has been scaled up in the form of light e-airplanes. Tian Yu, CEO of Yuneec (Unique) International Co., is based in Kunshan, Jiangsu Province, China, where his factory employs 1,800 workers to assemble 1 million drones per year. When Tom Peghiny of South Woodstock, Connecticut, developed his Flightstar e-Spyder ultralight, he then went into collaboration with Yuneec to develop the GreenWing International eSpyder (GW280), a single-seat electric airplane sold as a kit for construction as an amateur-built airplane, as well as the two-seat GW430. It has a Yuneec 24-kW electric power system that can lift a 220-lb (100 kg) payload at an initial climb rate of 375 feet (110 m) per minute. Maximum speed is 56 mph (90 kph); economy cruise is 37 mph (60 kph). Flights as long as one hour are possible, with a 30-minute reserve still available upon landing.
The first flight of the GreenWing E430 took place from the Yuneec factory near Shanghai on June 12, 2009. In kit form, it was then shipped for further testing to Camarillo, California. On July 14, 2009, the prototype aircraft was registered in the USA as N386CX, and on July 18, 2009, it was given a Certificate of Airworthiness by the Federal Aviation Administration. Further test flights were carried out, totaling 22 hours. The prototype E430 was then shipped by truck to Wisconsin and displayed at EAA AirVenture Oshkosh in July 2009. The two-seater high-wind composite is powered by a 42 kW Yuneec motor, powered by a three-pack Yuneec OEM Li-Po battery weighing 13 kg (28.6 lbs.) and producing 66.6V (30 Ah). With the three-pack, the airplane is expected to have an endurance of 1.3 to two hours cruising at about 53 knots. With a five-pack battery configuration, endurance is said to be 2.25 to 2.5 hours. It takes about three hours to charge the batteries from regular AC power outlets. Batteries are expected to last about 1,500 hours. The innovative battery system allows battery packs to be easily and quickly swapped so that the plane can be flown with one battery while another one is being charged simultaneously. Tian Yu announced that his company was building a 260,000 ft2 (25,000 m2) factory to produce the aircraft in Shanghai, that was expected to open in October. Yuneec is working on developing a solar-cell installation for the wings that will recharge the aircraft’s batteries.
In 2010, the E430 was named the winner of the Lindberg Prize for electric aircraft at AirVenture. In the same year it was named Brit Insurance Design of the Year in the transport category. But by December 2012, a total of only two examples had been registered in the United States with the Federal Aviation Administration. The first one was the initial prototype shipped to the U.S., registered in the Experimental–Exhibition category on July 14, 2009, although its registration expired on March 31, 2012. The second was registered in the Experimental–Research and Development category on January 26, 2011, to Flying Tian of Monterey Park, California. German DULV certification was awarded in February 2013. GreenWing began taking orders at the 2013 EAA AirVenture show in Oshkosh and kits were scheduled to begin delivery the end of 2013. But as of August 2015, GreenWing International had stopped trading.
Almost in parallel, Huazhi Hu, based in the Guangzhou Province of southern China, had already been making camera and hobby drones, such as the Ghostdrone1.0, when he decided to “scale up” and make a passenger-carrying autonomous drone (PAV), which could transport a person via air the same way as Google’s self-driving automobile via road. After two of his friends, his CEO Ji Chen and his helicopter coach, were tragically killed in air crashes, Huazhi Hu determined to make an absolutely safe transporter. The idea is that a passenger hops in, enters his destination on a 12-inch (30-cm) touchscreen, and hits the takeoff button. Weighing 440 lb. (200 kg), the EHang 184 AAV looked like a small helicopter although with four “lifter booms,” apparently a quadcopter. It was in reality an “octocopter” because each boom had two independent high-power electric motors and composite propellers. Thus, if any of the motors or props were to fail, there was still adequate lift and control to bring the EHang 184 to a safe landing. The electric-powered drone could be fully charged in two hours, carry up to 220 lb. (100 kg), and fly for 23 minutes at sea level. The cabin fitted one person and a small backpack and even had air conditioning and a reading light. It was designed to fly about 1,000 to 1,650 feet (300 to 500 m) off the ground with a maximum altitude of 11,500 feet (3,500 m) and top speed of 63 mph (100 kph). If any components malfunctioned or disconnected, the aircraft would immediately land in the nearest possible area to ensure safety. Its propellers folded inwards as it landed so it could fit in a single car parking space. The communication was encrypted and each AAV had its independent key.
The EHang 184 was unveiled at CES International in Las Vegas on January 6, 2016. U.S. authorities were just starting to lay out guidelines for drone use, and a human-passenger drone seemed certain to face strict scrutiny. On June 8, 2016, the EHang 184 was given clearance for testing by the Nevada Institute for Autonomous Systems (NIAS) and the Governor’s Office of Economic Development (GOED) to put the drone through testing and regulatory approval. The program will take place at Nevada’s FAA-approved test site, one of six such drone-testing locations across the U.S. There was no mention of whether this testing would cover the vehicle’s capacity to carry humans, but the stated intention is to gain the EHang 184 a Certificate of Airworthiness, an FAA document that gives it authority to fly. “I personally look forward to the day when drone taxis are part of Nevada’s transportation system,” the institute’s business development director, Mark Barker, told the Las Vegas Review Journal.
From 2014, Huazhi Hu, a drone millionaire based in the Guangzhou province of southern China, developed the Ehang184 (Ehang).
During 2016, EHang continued its progress; a handful of 184s made some 200 test flights, some fully autonomous, to further improve the technology, but only in China. Monitoring of these flights was carried out by the very first flight command center in Guangzhou; the building was converted to remotely track live data—including speeds, altitude, individual propeller power, location, drone camera feed and video feed of the passenger—plus communicate with passengers and schedule air traffic. EHang planned 2- and 4-seater versions and large-scale working drones for agriculture. In February 2017 at the World Government Summit, Mattar al-Tayer, the head of Dubai’s Roads and Transportation Agency, announced that, following experiments in the skies above their city, a blue and white livery EHang 184 would go into regular operation. This was to be integrated into the project of Sheikh Mohammed bin Rashid Al Maktoum whereby by 2030 one-quarter of public transport will be autonomous. But in an ecosystem where winds can go up to 40–50 knots (46–58 mph), with both sand and fog, certain critics pointed out that its ballistic parachutes would not have the time to deploy at altitudes below 100 ft. (30 m), and even if the chutes did deploy, the PAV could crash into power lines, trees or water. Unfortunately, with the recent lack of sales of its Ghostdrone, EHang had to lay off about 70 employees and appeared to be dealing with fiscal problems resulting in missed payments to suppliers; one of the victims was the EHang 184. When Dubai launched its drone-powered taxi service, EHang did not appear.
During 2015, students from the National University of Singapore (NUS) built and tested the single-seater eVTOL Snowstorm, under the auspices of FrogWorks, a collaboration between the NUS faculty of engineering’s Design-Centric Programme (DCP) and the University Scholars Programme (USP). Snowstorm was fitted with 24 motors, each driving a propeller of 76 cm diameter with 2.2 kW of power. Its hexagonal frame was made up of anodized aluminum beams, carbon-fiber plates, and tubes with Kevlar ropes. The pilot seat was positioned at the center of the machine, its weight supported by six landing gear legs, the bottom of which is an inflated ball that adsorbs shock when landing. Three independent rechargeable lithium battery sets provide a total power of 52.8kW. Snowstorm was never commercialized.7
The number of near-misses and collisions between drones and aircraft is on the rise. According to the UK’s Civil Aviation Authority (CAA), there were 23 drone-related incidents at UK airports between November and April 2017, including 12 near-misses. A clutch of British companies have developed a counter-drone technology, the Anti-UAV Defence System (AUDS), which sends out drone-jamming signals and will soon be tested out at U.S. airports by the Federal Aviation Administration (FAA). According to a recent report, the FAA had close to 600 drone-sighting reports in the period between August 2015 and January 2016. The agency said it receives more than 100 reports every month regarding unmanned aircraft flying dangerously close to an airport or airplane.
Recognition of the importance of the vital importance of controlling the air traffic goes back to the dawn of commercial aviation. In 1921, Croydon Airport, London, was the first airport in the world to introduce air traffic control, using wireless telegraphy. In the USA the number of planes using Cleveland Airport jumped from a few thousand in its inaugural year to nearly 20,000 by 1929. A new terminal building constructed that year contained the world’s first air traffic control tower, a tall, glass-enclosed structure with a 360-degree view of the airfield. Soon after its construction, two-way radio was installed in the tower, the first time this had been used in the aviation field. This proved to be an important addition, as in its early years the airport used the “allway” landing mat process, which allowed multiple planes to land simultaneously on different parts of the airfield, a process designed to prevent pilots from having to wait mid-air for space to land. Following the crucial use of radar in World War II to control the arrival and departure of warplanes during the Battle of Britain and other theaters, the system was adopted to monitor and control the busy airspace around larger airports. With the exponential growth of drones and runway-free eVTOL taxis, the pressing challenge has been to monitor traffic at an altitude and even below, halfway between automobile traffic and conventional air traffic at 33,000 ft. (10,000 m) and so prevent accidents.
From late 2015, extensive work was carried out by engineers at NASA’s Ames Research Center in Moffett Field, California, on UTM for Civilian Low-Altitude Airspace and Unmanned Aircraft System Operations drone traffic control. The majority of flight testing occurred at Crows Landing, a remote, closed, private-use airfield, 18 miles (29 km) southwest of Modesto, California. Prior to flight test, the team deployed a 100-ft. (30-m) weather tower, small weather stations, microphone, Automatic Dependent Surveillance-Broadcast (ADS-B) in a ground relay station for air traffic feeds, and a radar station for flight test monitoring and data collection. The Technical Capability Level One system will enable UAS operators to file flight plans reserving airspace for their operations and provide situational awareness about other operations planned in the area. Technology Capability Level 2 (TCL2) has focused on flying small drones well beyond the pilot’s line of sight over sparsely populated areas at five of the FAA test sites: Virginia, North Dakota, Texas, Alaska and Nevada. The team worked with over 250 partners throughout the industry that plan to use drones in their businesses in the airspace above buildings and below crewed aircraft operations in suburban and urban areas. So NASA’s NTX research center is exploring how flight corridors can work without voice interactions. This includes improved “sense-and-avoid” technology that will allow drones to communicate with other passenger aircraft to avoid one other.
On October 25, 2017, President Donald Trump directed Secretary of Transportation Elaine Chao to launch the unmanned aircraft systems (UAS) Integration Pilot Program, an initiative to safely test and validate advanced operations for drones in partnership with state and local governments in select jurisdictions. This program will evaluate a variety of operational concepts, including night operations, flights over people, flights beyond the pilot’s line of sight, package delivery, detect-and-avoid technologies, counter-UAS security operations, and the reliability and security of data links between pilot and aircraft. It is designed to provide regulatory certainty and stability to local governments, communities, UAS owners and operators who are accepted into the program. In less than a decade, the potential economic benefit of integrated unmanned aerial systems into the nation’s airspace is estimated to equal up to $82 billion and create up to 100,000 jobs.
Another way to tackle the threat posed to commercial air traffic by wandering drones has been tested by Dubai with a Drone Hunter, developed by the UAE–based tech firm Exponent. Once the pilot confirms the location, the hunter is ready to take off in three minutes at a speed of 110 km per hour. Elsewhere, in the Netherlands, the Dutch National Police have taken a lower-tech approach by collaborating with a private company called Guard from Above, which trains birds of prey, such as juvenile bald eagles imported from North America, to intercept drones.
The AgustaWestland Project Zero is a tiltrotor/fan-in-wing airplane. In December 2010, the management of AgustaWestland in Farnborough, England, approved the formation of a team under James Wang with the intention of producing a technology demonstrator incorporating as many innovations as possible on a single airframe. Various companies in Italy, the UK, the U.S. and Japan worked on the design and/or manufacturing of elements of Project Zero, including four different branches of Finmeccanica. Ansaldo Breda designed a custom-built electric motor inverter and accompanying motor control algorithm, while Selex ES provided the High-Integrity Flight Control Computer and the Actuator Control Unit. Lucchi R. Elettromeccanica custom-built the axial flux permanent magnetic motors; Rotor Systems Research LLC worked in conjunction with AgustaWestland on the aerodynamic design of the rotor blades. Lola Composites produced the composite material from which most of the exterior surfaces are made of carbon-fiber-reinforced polymer (CFRP), while UCHIDA manufactured the composite structure for the rotor blades, shrouds, and spokes. Stile Bertone worked to develop the aesthetic and aerodynamic styling of the aircraft. In June 2013, Project Zero was publicly displayed at the Paris Air Show. Flight testing has also continued, but focused on ⅓-scale models rather than the full-scale demonstrator due to the limited flight endurance it possesses. In February 2016, it was announced that, until batteries increase their energy density, a hybrid drive system would be installed on the full-scale aircraft; this is aimed at extending the flight endurance from 10 minutes to 35–45 minutes.
Another example of scaling-up has been made by Hirobo Electric Corporation of Hiroshima, well known and respected for over forty years for its line of high-quality radio-controlled helicopter models such as the Hirobo Eagle 3 EP. In 2010, Hirobo decided to use their expertise to present a personal electric all-composite helicopter, the Hirobo HX-1 BIT, which was launched at the International Robotics Exhibition in Tokyo in 2013. The Japanese company’s compact IMU-05 attitude sensor is one such technology that has been developed to enable autonomous flight, and offers the company a serious competitive advantage in that area. The IMU-05 collects a wide range of information such as attitude angle, acceleration, angular rate and magnetic direction, and enables the helicopter to be remarkably stable in blustery winds—it is a key enabling technology of autonomous flight for helicopters. The company is not expecting to deliver the passenger-carrying variant before 2021, citing expected bureaucratic problems.
Joe Ben Bevirt of Santa Cruz, California, whose motto is “To found, build, invest in, and guide companies which improve the world,” had already founded Joby Energy, Inc., to develop airborne wind turbines to harness the immense and consistent power in high-altitude wind to provide reliable and low-cost renewable energy. In 2015 Bevirt founded Joby Aviation with aeronautical engineer Alex M. Stoll. Bevirt and Stoll then applied for a patent for an “Aerodynamically Efficient Lightweight Vertical Take-Off and Landing Aircraft with Pivoting Rotors and Stowing Rotor Blades.”8
Alex Stoll is well qualified. Following a spell in the Aircraft Aerodynamics and Design Group Stanford University (2008–2012), he worked with Mark D. Moore, William J. Fredericks and Nicholas K. Borer of NASA Langley Research Center to investigate drag reduction through distributed electric propulsion. The two-seater Joby S2, powered by lithium-nickel-cobalt-manganese-oxide batteries, uses 12 tilting electric propellers to provide multirotor-style balanced VTOL capabilities. Once it reaches cruising speeds, these rotors fold away into aerodynamic bullet shapes, and the aircraft can reach speeds of up to 200 mph (322 km/h) and ranges of up to 200 miles using four additional cruise-optimized props on the backs of the wings and tail fins. The two-seater Joby S2 Electric VTOL PAV should soon begin trials.
During 2010, as part of his doctoral degree, Dr. Mark D. Moore, the chief technologist at NASA’s Langley Research Center, Hampton, Virginia, working with a team from Massachusetts Institute of Technology, the Georgia Institute of Technology, the National Institute of Aerospace (NIA), and M-DOT Aerospace, came up with the concept of a hover-capable, electric-powered, low-noise, personal VTOL technology-concept, prop-rotor aircraft. Called Puffin, it would be capable of flying a single person at a speed of 150 mph (240 kph). Range is expected to be less than 50 miles (80 km) with initial battery technology. The design has a 14.5-foot (4.4-m) wingspan and stands 12 feet (3.7 m) tall on the ground in its takeoff or landing configuration. In August 2010, the one-third-scale model of the Puffin was on display at the NASA Langley campus for filming for the Discovery network series Dean of Invention. The concept was also presented to the American Helicopter Society conference on aeromechanics.
In 2010, ONERA (the French National Aerospace Research Laboratory), led by Claude Le Tallec, an expert in remotely piloted aircraft systems, launched the idea of the P-Plane (Personal Plane Project), essentially a European program. Like the Paris-based Velib (bicycle) and Autolib (electric car) fleets, the P-Plane concerns a similar network of ATC electric air taxis. Its R&D brings together a consortium of five research centers, four universities and a range of businesses from nine European countries and Israel.
The Sikorsky Aircraft Corporation of Stratford, Connecticut, also launched Project Firefly to build and flight-test the world’s first large-scale all-electric tilt-rotor. Sikorsky announced the existence of their new aircraft on July 19, 2010, at the Farnborough International Air Show in the United Kingdom and displayed it for the first time at the Experimental Aircraft Association’s AirVenture 2010 convention in Oshkosh, Wisconsin, on July 26. The Firefly is a modified Sikorsky S-300C helicopter with its engine replaced by an electric motor and digital controller from U.S. Hybrid and two Li-Po battery packs. Integrated sensors provide real-time aircraft health information to the pilot through a panel-integrated, interactive LCD monitor. Eagle Technologies LLC executed the custom airframe modifications and assembly of the demonstrator aircraft. The helicopter can hold only the pilot, no passengers, and operate for 12 to 15 minutes. It has a top speed of 80 knots, or about 92 mph. It was expected that the helicopter would make its first flight in late 2010 or in early 2011. The project was put on hold until more energy-dense batteries could be located.
Without the resources of a big corporation, on August 12, 2011, Pascal Chretien, a French-Australian electrical/aerospace engineer and helicopter pilot, built and flew his e-helicopter Solution F for 2 minutes 10 seconds up to a maximum height of 1 meter at Venelles, near Aix-en-Provence, France. As a conventional tail rotor drains somewhere between 8 percent and 10 percent of total hover power, Chretien modeled and built a coaxial design with two counter-rotating rotors on top—a torque-balanced design that can fly without the need for a tail rotor to stop the aircraft from rotating out of control—instead, it just needs a simple, lightweight tail fin. This concept was taken from the conceptual computer-aided design model on September 10, 2010, to the first testing at 30 percent power on March 1, 2011—less than six months. In place of the typical cyclic control, which uses an ingenious variable blade-tilting system to control which way the helicopter tilts and advances, Chretien chose an extremely simple weight-shifting system—a big set of handlebars (incorporating the collective control) that literally tilt the main weight of the aircraft underneath the rotors—as his steering assembly. The rechargeable battery cells are Li-Po pouch cells, with an energy density of 160 watt-hours per kg. Chretien initiated a few tethered test flights to test the action and torque balance of the rotor controllers, the weight-shift directional tilt system and the ground effect behavior of the aircraft.
During the following weeks, Solution F logged a total 99.5 minutes of flying time in 29 flights. A typical flight lasted four minutes, with a demonstrated maximum of six minutes. Actual forward flight above translational speed (15 to 20 knots) was never experimented. Although very low speed was tried, the test flights were just hovers, outside ground effect. Chrétien and Solution F filed patents relating to a “serial hybrid” helicopter concept. The principle is built around an engine that produces electric power via a generator. The generator feeds batteries, which enable a distributed stack of electric direct drives to turn the blades. In 2012 Pascal Chrétien formed a company in the USA based on his patented technology, Tetraero.
The challenge of flying a conventional electric helicopter has been taken on by Philippe Antoine working with the ENAC HCI engineering school with the removable-wing Volta, an MC1 microcopter. The maiden flight, lasting a record 15 minutes with Antoine at the controls, took place on December 2, 2016, at Castelnaudary runway near Toulouse, France. Further funds to increase battery energy are en route.
On November 11, 2013, American singer Lady Gaga (aka Stefani Joanne Angelina Germanotta) launched ARTPOP, her third studio album, at the ArtRAVE party inside a building at the Brooklyn Navy Yard by piloting “Volantis,” the world’s first flying dress, designed by Benjamin Males of the London company Studio XO and built by TechHaus. She was strapped into a white carbon-fiber bodice in the shape of a woman, above which six twin-motor fan units were surrounded by white columns arranged hexagonally and connected to a central node above the bodice, which rested on the ground using a circular stand when not in flight. The batteries and associated control and radio link systems were housed in a central hub at the top of the column, in order to minimize the weight of copper used for the electrical power cabling. The hex 12 drone propelled Lady Gaga half a meter (20 inches) off the ground and hovered forward for several meters. “Hopefully, one day you’ll own a Volantis of your own,” she commented. ARTPOP became the ninth best selling album of 2013 with 2.3 million copies worldwide.9
The use of drones as MAVS (Micro Air Vehicles) has an interesting history. In December 1992, the RAND Corporation conducted a study for the Defense Advanced Research Projects Agency (DARPA) that considered a wide variety of micro-devices for defense applications. This study projected that it would be possible to have flying vehicles with a 1 cm span and less than 1 gm payload in ten years. In 1993 the RAND Corporation performed a feasibility study on very small controlled or autonomous vehicles. A more detailed study followed and was performed at the Lincoln Laboratory in 1995. This led to a DARPA workshop in the fall of 1995. Developing 15.24-cm (6-in.) flying vehicles was proposed in the fall of 1995 by R.J. Foch of the U.S. Naval Research Laboratory (NRL) and M.S. Francis (DARPA). Vehicles of this type might carry visual, acoustic, chemical or biological sensors. They became of interest because electronic detection and surveillance sensor equipment were miniaturized so that the entire payload weighed 18 grams or less.
It was not long before AeroVironment’s Paul MacCready had become involved. In 1996, they were funded by DARPA with a Phase I SBIR contract to study the feasibility of a 15.24-cm MAV. They concluded that a vehicle of this size was feasible and received a Phase II SBIR contract in 1998 that resulted in the Black Widow, one of the smallest and most successful MAV systems that could carry a useful payload. This vehicle was electrically powered by one 10-watt DC motor with a four-inch propeller; had an aspect ratio of 1.0, a wing span of 15.24 cm, and a total mass of ~80 g; and could carry a color video camera and transmitter. It also had a 3-gram fully proportional radio control system. A pneumatic launcher and a removable pilot’s control unit with a 10.16-cm (4-in.) liquid-crystal display, in a briefcase, were also developed to complete the system. In 1999 the AeroVironment MAV team led by Matt Keenon received awards from DARPA and Unmanned Vehicles Magazine for the Black Widow. The Black Widow set several records for an outdoor flight of a micro air vehicle on August 10, 2000, including an endurance of 30 minutes, a maximum range of 1.8 km (2 mi), and a maximum altitude of 234.39 m (770 ft.). The success of the Black Widow led AeroVironment to the development of a somewhat larger “flying wing” MAV, the Wasp. The Wasp had a root chord of 21.33 cm (8.4 in.) and a wing span of 36.57 cm (14.4 in.), and weighed 181.43 g (6.4 oz). It was powered by one 10 w DC electric motor, was designed to fly between 40.23 kph and 48.27 km/h (25 and 30 mph) at a maximum altitude of 91.44 m (300 ft.), and had a color video camera and transmitter. The Wasp was hand-launched, and had an autopilot, an endurance of 60 minutes, and a range of 4 km (2.5 mi) line-of-sight. This vehicle eliminated the need for the pneumatic launcher of the Black Widow and was easier to fly. A stripped-down RC Wasp set an endurance record of 1 hour and 47 minutes on August 19, 2002.10
In January 2010, Tamkang University (TKU) in Taiwan realized autonomous control of the flight altitude of the Golden Snitch, an 8-gram (0.3-oz), 20-cm (8-in)-wide, flapping-wing MAV. The MEMS Lab in the TKU has been developing MAVs for several years, and since 2007 the Space and Flight Dynamics (SFD) Lab has joined the research team for the development of autonomous flight of MAVs. Instead of traditional sensors and computational devices, which are too heavy for most MAVs, the SFD combined a stereo-vision system with a ground station to control the flight altitude, making it the first flapping-wing MAV under 10 grams that realized autonomous flight.
In 2008, the TU Delft University in the Netherlands developed the smallest ornithopter fitted with a camera, the DelFly Micro, the third version of the DelFly project that started in 2005. This version measures 10 cm (4 in) and weighs 3 grams (0.1 oz), slightly larger (and noisier) than the dragonfly insect on which it was modeled. The importance of the camera lies in remote control when the DelFly is out of sight. However, this version has not yet been successfully tested outside, although it performs well indoors. Researcher David Lentink of Wageningen University, who participated in the development of previous models, DelFly I and DelFly II, says it will take at least half a century to mimic the capabilities of insects, with their low energy consumption and multitude of sensors—not only eyes, but gyroscopes, wind sensors, and much more. He says fly-size ornithopters should be possible, provided the tail is well designed. Rick Ruijsink of TU Delft cites battery weight as the biggest problem; the lithium-ion battery in the DelFly MAV, at one gram, constitutes a third of the weight. Luckily, developments in this area are still going very fast, due to demand in various other commercial fields.
Was there a limit to the miniaturization of camera-equipped quadcopter drones? In 2014, Guangdong Cheerson Hobby Ltd., located in Shantou City, Chenghai District, Fengxin industrial zone, presented their CX YouCute Tiny. It measured only 2.3 × 2.3 × 2.0 cm (1 in. × 1 in. × 0.8 in.) and weighed only 7g (0.2 oz), being launched from two fingers of a human hand.
For the autonomous navigation of miniaturized robots (e.g., nano/pico aerial vehicles), MIT engineers have designed a computer chip that uses a fraction of the power of larger drone computers and is tailored for a drone as small as a bottle cap. The new methodology and design has been termed Navion. The team, led by Sertac Karaman, the Class of 1948 career development associate professor of aeronautics and astronautics at MIT, and Vivienne Sze, an associate professor in MIT’s department of electrical engineering and computer science, developed a low-power algorithm, in tandem with pared-down hardware, to create a specialized computer chip. Karaman says the team’s design is the first step toward engineering “the smallest intelligent drone that can fly on its own.” He ultimately envisions disaster-response and search-and-rescue missions in which insect-sized drones flit in and out of tight spaces to examine a collapsed structure or look for trapped individuals.11
At the General Robotics, Automation, Sensing and Perception (GRASP) Laboratory at the University of Pennsylvania, mathematicians, computer scientists, and engineers have been working to create these complex robots and to operate them. The GRASP team prefers the term “robot” to “drone” altogether because of the latter’s violent stigma. The GRASP laboratory received a $5 million grant from the DoD to study swarming groups of networked autonomous robots—the Scalable Swarms of Autonomous Robots and Mobile Sensors, or SWARMS project. The goal is to create a swarm of bug-like quadrotors so intelligent it will re-form. Multiple vehicles can fly as a formation on “search and rescue and disaster recovery” missions, from acting as first responders in dangerous situations to gathering intelligence in a hostage crisis. The Pelican, another model, is equipped with seven-inch-long arms and a laser scanner and camera. It can fly into a building, scope out the digs, and construct real-time 3D maps, identifying features like doorways, people, and furniture, estimating its position with respect to these features one hundred times a second, and navigating. Another micro-drone research project is called the Micro Autonomous System Technologies Collaborative Technology Alliance, also funded by the Army Research Lab, thanks to a $22 million grant—the single largest grant in the history of the university’s engineering school. The stated intent is “to help create the fundamental networks and technologies that will put unmanned machines on the front lines of battle.”12
Another example is the Perdix drone developed by MIT engineering students in 2011. Perdix, named after a character from Greek mythology who was changed into a partridge, has a wingspan of 12 in. (30 cm), operates autonomously, and shares a distributed self-healing brain. It carries out its mission without human piloting, but can talk to other drones to collaborate on getting the job done: surveillance. Because every Perdix communicates and collaborates with every other Perdix, the swarm has no leader and can gracefully adapt to drones entering or exiting the team. It operates in cooperative swarms of 20 or more. The technology was first modified by the U.S. military in 2014. The Perdix program became known from March 2016, when the Washington Post revealed footage of an F-16 fighter releasing 20 Perdix over Alaska. At the time, however, the Post stated the drones had already been undergoing flight testing for two years. On October 26, 2016, the U.S. military launched a swarm of 103 Perdrix drones from an F/A-18 Super Hornet fighter jet during a test over Naval Weapons Station at China Lake in California. Launching was from small pods on hard points on both fighter-plane wings. The Perdix are capable of withstanding ejection at speeds of up to Mach 0.6 and temperatures as low as minus 10 degrees Celsius. On October 26, the drones formed up at a preselected point and then headed out to perform four different missions. Three of the missions involved hovering over a target, while the fourth mission involved forming a 100-meter-wide circle in the sky. The Perdix may revolutionize aerial warfare.
Further progress was made with AIAUVs during the summer of 2017, when Ashish Kapoor, Andrey Kolobov and a team at the Adaptive Systems and Interaction Group at Microsoft Research, Redmond, Washington, tested two artificially intelligent model sailplanes above the desert valley surrounding Hawthorne, Nevada. The Styrofoam machines with their 16.5-foot (5-m) wingspan navigated the skies on their own, guided by computer algorithms that learned from onboard sensors to monitor air temperature, wind direction, altitude, and other metrics, in addition to speed and location data from GPS, which enabled them to predict air patterns. The AI pilot can detect when the sailplane is suddenly gaining altitude, indicating it has located a rising thermal, and plan a route forward. The Microsoft Aerial Informatics and Robotics admitted that prototypes were still dependent on an electric motor to get off the ground in the first place, while the servos that move the sailplane’s flaps and ailerons so that it can steer towards and stay within a thermal are all powered by an onboard battery. But eventually solar cells on a larger gilding aircraft’s massive wings could provide all the power it needs, as could wind-powered generators incorporated into its fuselage, and enable it to fly indefinitely, or in other words, an infinite soaring machine.13
Since 2011, engineers led by Marcus Fischer at the Bionic Learning Network of FESTO, an industrial control and automation company based in Ostfildern in southern Germany, have developed the Li-Po battery 23-watt 3.70 m (12 ft) SmartBird, resembling the herring gull and capable of flapping its wings to take off, to fly and to land without the aid of other devices to provide lift. The natural wing beat of a bird was emulated by using bionics technology to decipher bird flight. The Festo Network includes TU Berlin, Delft University of Technology, TU Ilmenau, Friedrich Schiller University Jena, Christian-Albrechts University in Kiel, University of Arts and Industrial Design Linz, University of Oslo and Akershus, Department of Product Design, University of Applied Sciences Ravensburg-Weingarten, University of Stuttgart, CIN University Tübingen, University of Ulm, Fraunhofer IPA.
The honeybee (Apis mellifera), which pollinates nearly one-third of the food we eat, has been dying at unprecedented rates because of a mysterious phenomenon known as colony collapse disorder (CCD). The situation is so dire that in late June 2014 the White House gave a new task force just 180 days to devise a coping strategy to protect bees and other pollinators. The crisis is generally attributed to a mixture of disease, parasites, and pesticides. Inspired by the biology of a bee, researchers, led by engineering professor Robert Wood at the Microrobotics Lab of the Wyss Institute, Harvard University, are developing Autonomous Flying Microbots aka RoboBees, man-made systems that could perform myriad roles in agriculture or disaster relief. A RoboBee measures about half the size of a paper clip, weighs less than one-tenth of a gram, and flies using “artificial muscles” compromised of materials that contract when a voltage is applied. To construct RoboBees, researchers at the Wyss Institute have developed innovative manufacturing methods, so-called pop-up micro-electromechanical (MEMs) technologies that have already greatly expanded the boundaries of current robotics design and engineering. A RoboBee can lift off the ground and hover midair when tethered to a power supply. After two years of R&D, in 2016 the Wyss team announced that their RoboBees can now perch on objects from any angle, using an electrode patch and a foam mount that absorbs shock to perch on surfaces and conserve energy in flight—like bats, birds or butterflies. The new perching components weigh 13.4 mg, bringing the total weight of the robot to about 100mg—similar to the weight of a real bee. The robot takes off and flies normally. When the electrode patch is supplied with a charge, it can stick to almost any surface, from glass to wood to a leaf. To detach, the power supply is simply switched off.
But they still need to be able to fly on their own and communicate with each other to perform tasks like a real honeybee hive is capable of doing. The researchers believe that as soon as 10 years from now these RoboBees could artificially pollinate a field of crops, a critical development if the commercial pollination industry cannot recover from severe yearly losses over the past decade. RoboBees will work best when employed as swarms of thousands of individuals, coordinating their actions without relying on a single leader. The hive must be resilient enough so that the group can complete its objectives even if many bees fail.14
Another approach has been taken by Anna Haldewang, a 24-year-old industrial design student at Savannah College of Art and Design (SCAD) in Georgia. Haldewang created 50 designs of a bee drone before landing on the final model, Plan Bee, which does not resemble a bee at all. The drone consists of a foam core, a plastic-shell body and two propellers. There are also six sections of the drone that meet at the bottom, all of which have tiny holes that let the machine gather pollen while it hovers over plants. It can then release the pollen at a later time for cross-pollination. Haldewang noted that Plan Bee is still in its early stages, but she has filed a patent for the technology and design. Its application in backyards as a teaching tool has potential, but the drone could conceivably be used in large-scale farming, even in hydroponic farming.
During 2016, a team led by Mirko Kovac at Imperial College London’s Aerial Robotics Lab at the South Kensington campus, again taking inspiration from nature, developed an aquatic micro aerial vehicle (AquaMAV) which dives like a gannet and launches like a flying fish to collect water samples. The drone only weighs 200 grams (7 oz) and can currently achieve speeds of around 30 miles per hour from a starting point beneath the water. It can make the aerial leap even if conditions on the surface are rough. The researchers state that using waterjet propulsion and energy from a 3.5 g (0.1 oz) Li-Po battery, AquaMAV can currently fly around five kilometers to and from an analysis. The team says the aerial range would enable those analyzing the samples to be at a safe distance away from a potentially hazardous situation. The Imperial College team has also developed another nano-quadcopter which acts like a spider, whereby tools can be attached to the drone’s moving arm, which could allow it to repair and examine buildings in hard-to-reach or dangerous areas; it can even spin a simple web of fine thread.
In agriculture, farmers are now using infrared camera–carrying drones to pinpoint problem spots with insects and aphids in vast fields and ranchlands. Based on the mapping, another drone then drops a “cocktail” of predatory insects, transported in a sock attached to the underbelly of the drone and containing a mixture of vermiculite and insects, onto grapevines and citrus trees to combat pests. By focalizing pest control, they prevent spread and save money. One example of this is the AgriDrone developed by Saga University in Japan and IT firm OPTiM. Agri Drone uses a suspended bug zapper which delivers a glowing, insect-enticing electric payload to the points at which the pests are congregating in harmful numbers. Primarily used at night, it utilizes infrared and thermal cameras to shoot targeted doses of pesticides where insects are congregating. A similar project has been developed by the Department of Life Sciences and Computing at Imperial College London with agriculture services company Agrii of Grantham.
Researchers at the University of Illinois at Urbana-Champaign and Caltech have developed a self-contained robotic bat—dubbed Bat Bot (B2)—with soft, articulated wings that can mimic the key flight mechanisms of biological bats. The B2 possesses a number of practical advantages over other aerial robots, such as quadrotors. Bats have more than 40 active and passive joints and the B2 robot has only 9 (5 active and 4 passive). The compliant wings of a bat-like robot flapping at lower frequencies (7–10 Hz vs. 100–300 Hz of quadrotors) are inherently safe because, replaced by rapid-spinning propellers, their wings comprise primarily flexible silicone membrane materials and are able to collide with one another, or with obstacles in their environment, with little or no damage.
Instead of designing microdrones to mimic insects, an alternative is to equip live insects with electronic navigation systems. In 2009, DARPA funded a joint project by the University of California, Berkeley, and Nanyang Technological University in Singapore to control wirelessly the flight of the giant flower or rhinoceros beetle using a 6-cm (2.4-in.), 8-gm (0.3-oz) backpack powered by a 3.9 volt lithium battery connected to six micro-electrodes implanted into the insect’s wing-folding coleopteran muscles. In 2017, a team led by Jesse J. Wheeler at the Charles Stark Draper Laboratory, Inc., in Cambridge, Massachusetts, announced their DragonflEye project whereby, in conjunction with the Howard Hughes Medical Institute, they had “equipped” live dragonflies with aoptrode navigational systems, even smaller than optical fibers. In addition, real dragonflies are unbelievably nimble and quick, compared to clumsy man-made drones, with the ability to maneuver turns as sharp and fast as 9-Gs. Another advantage is that, as long as they have food, water and sunlight, they have a greater autonomy than battery-powered mechanical drones. Draper is also planning to use the same system on bumblebees for pollination.15