The proliferation of unmanned air vehicles and their use in a widening variety of roles has increased the pressure on industry, regulators, the military and even operators to find ways to integrate them into national airspace.
While rules of the air for a whole gamut of manned types are largely considered to be established the world over, introduction of UAVs – ranging from palm-sized systems weighing only a few ounces up to high-altitude aircraft weighing in at some 14t – is proving to be far more complex.
General aviation aircraft such as gliders and microlights operate in national airspace with only the pilot’s visual capacity to prevent them colliding with other aircraft, but having this pilot on the ground, as in the case of a UAV, does not sit easily with a manned aviation world that does not consider unmanned air systems to have the same situational awareness as a pilot sitting in a cockpit.
On the whole, regulations allow for small UAV operators to fly their systems within line of sight of the operator at particular altitudes, but as more sophisticated technology becomes commercialised and the benefits and appeal of operating such systems is recognised, these restrictions are hindering an industry that could be worth billions of dollars in years to come.
Operators want to be able to fly beyond line of sight, so providing the equivalent of a pair of eyes on board the remotely piloted air system to allow it to sense and avoid other obstacles is the holy grail of unmanned airspace integration.
“The intent of sense-and-avoid is for pilots to use their sensors (eyes) and other tools to find and maintain situational awareness of other traffic and to yield the right of way, in accordance with the rules, when there is a traffic conflict,” Adrian Muraru of the Ștefan cel Mare University of Suceava, in Romania, says in a paper entitled Critical Analysis of Sense and Avoid Technologies for Modern UAVs.
Olivier Lapy Sagen
“Since the purpose of this regulation is to avoid mid-air collisions, this should be the focus of technological efforts to address the issue as it relates to UAVs, rather than trying to mimic and/or duplicate human vision.”
Muraru notes that the US Federal Aviation Administration does not provide a definition of what would be required from an sense-and-avoid system, “largely due to the number of combinations of pilot vision, collision vectors, sky background, and aircraft paint schemes involved in seeing oncoming traffic”.
Involved parties are split on what is holding back the introduction of sense-and-avoid technology for UAVs, but Muraru claims that it is a technology restraint, not a regulatory one.
Paul Kennard, trials management officer at 2Excel Aviation – a company that helps other companies develop and test emerging technologies – says the technology required is largely developed, but the funding necessary for its qualification commercialisation is what is missing.
“The real problem with all of this is that there are companies out there that are 70-80% there, but they have to make the decision on whether to commercialise that remaining 20% to produce a marketable product,” he says, adding that this is the expensive part that companies are not willing to invest in without a clear requirement.
Although safety concerns dictate that a sense-and-avoid capability is required, one problem with this is a seeming pressure to apply a “one-size-fits-all” solution to such wide-ranging aircraft; the assortment of UAV types will fly at different altitudes, speeds, in varying locations and will have diverse payload capacities.
Space constraints
Manned aircraft in controlled airspace co-operate with air traffic control through onboard transponders and other communications links that allow for self-separation of aircraft, but the size, weight and power restraints of a UAV make carrying equivalent equipment more difficult.
Google has become an unlikely participant in the realisation of how to integrate UAVs into national airspace, as it unveiled its plans to produce a “ really low-cost” automatic dependent surveillance-broadcast (ADS-B) transceiver.
“We have to answer the question: What does the market find palatable in order to really transform? And that’s where we’re going,” Dave Vos, head of the Google’s Project Wing told an ICAO RPAS conference in Montreal in March. “Think about it: Would you spend $2,000? We have to make it happen.”
Project Wing has previously revealed plans to develop a UAV delivery service, and airspace integration is essential to this.
The UAVs would need to operate in airspace below 500ft, and that is currently used by tens of thousands of general aviation aircraft. Some of those aircraft owners have been reluctant to spend money to equip their aircraft with an ADS-B Out system, but by 2020 the law in the USA will mandate that all aircraft carry the system if it wants to fly in controlled airspace, so making the technology smaller and cheaper will benefit both manned and unmanned aviation.
“Rather than having a one-size-fits-all solution, RPAS sense-and-avoid mechanisms should reflect the operational scenarios in which the aircraft will be flown,” claims a paper on sense-and-avoid safety level requirements from the Guild of Air Pilots and Air Navigators (GAPAN). “Assuming that the platform can comply with the necessary air traffic management requirements, in order to operate exclusively in controlled airspace an RPAS would need only to be equipped like an airliner; it would not need a bespoke sense-and-avoid system.”
However, if it were to operate in uncontrolled airspace, the platform would need to be capable of sensing other airspace users in order to avoid them, it says.
In the case of the UK, which has a relatively complex and dense airspace, general aviation flights in uncontrolled “class G” airspace are not overly safe, statistics show.
Between 1999 and 2008 there were 19 mid-air collisions involving 38 aircraft. This included five military aircraft and 33 GA aircraft, of which 12 were gliders. Hour for hour, GA flying was 2.4 times more likely to involve a random mid-air collision than military flying, GAPAN surmises.
Furthermore, the UK Airprox Board’s database notes that in excess of 70% of the events reported each year occurred in class G airspace, particularly below 3000ft, and in areas around aerodromes. This finding was supported by a similar study carried out in the US that found that the majority of mid-air collisions occurred within 3nm (6km) of airfields.
Any system that is developed should be applied to manned aircraft as well, Kennard argues, based on these accident statistics, so that all aircraft have some sort of self-separation or collision avoidance capability, be it with a full-sized complex radar system, dedicated electro-optical sensor or a software modification to an existing payload. While a person can become distracted when visually flying an aircraft, a UAV by nature can have an unblinking eye.
The UAV does not make decisions based on the data that it collects – this is either stored on board until a later point or is transmitted down to the ground – but adding an algorithm to the sensor could provide it with a capability to either manoeuvre clear of a threat or alert the operator.
As far as the technology is concerned, the solution to this is largely understood to be a sense-and-avoid system that can be scaled up or down depending on UAV size and the operations that it has to carry out.
Kennard notes that a software modification and algorithm development is likely to be the answer, which could be integrated into existing sensors on board the UAV.
Northrop Grumman
GA aircraft do not typically fly at night, so a thermal-based sense-and-avoid capability may not be the most pressing of developments for airspace integration, because collisions when it is dark are less likely. Onboard processing would allow for a UAV to transmit in real time that a collision may happen, and either alert the operator or decide itself on what is the best action to take.
Size matters
Despite their large size, high-altitude, long-endurance UAVs may be considered a more accessible integration than smaller UAVs because they sit at altitudes above all general and commercial aviation routes, and only require airspace access to reach these altitudes.
Germany is one nation that has explored the integration of large UAVs with a sense-and-avoid capability into European airspace through its failed Euro Hawk programme.
This involved modifying a Northrop Grumman RQ-4 Global Hawk HALE UAV with a sensor payload that would help integrate it into European airspace, but the programme was cancelled in 2013 when the cost of developing the sensor escalated.
Berlin is now interested in the MQ-4C Triton maritime variant of the Global Hawk, which overcomes a series of challenges the European variant faced, such as the sense-and-avoid capability and communication links to developing air traffic management systems, and is considering starting a programme to explore this capability versus that of a converted business jet for a signals intelligence role.
The German government is also expecting to begin testing the Euro Hawk sensor again in a series of trials expected to start at the end of 2015, which will continue from where the last phase of testing left off.
“The MQ-4C is an easier air fit, but we still need to develop the sensor,” a German air force command representative tells Flight International. “We are trying to reuse the aircraft that is sitting in Germany. The plan would be that by the end of this year, beginning of next, we’d start flight tests of it again.”
The US Navy is also developing a sense-and-avoid capability for the Triton, and launched a competition in November to provide the HALE UAV with an air-to-air radar that enables it to avoid collisions.
Naval Air Systems Command (NAVAIR) announced that it was looking for a less sophisticated sensor than the Exelis-developed one that the navy had previously contracted to be developed for the Global Hawk – although a programme for it was cancelled in 2013. It will have a “due regard” capacity, which is not as developed as a full sense-and-avoid capability.
NAVAIR said it was taking an evolutionary approach to the development, claiming that is should be modular, scalable, and will be improved as air traffic control and operations develop.
Meanwhile, General Atomics Aeronautical Systems announced on 12 February that it had carried out a flight test of a pre-production, fully-functional variant of its sense-and-avoid air-to-air due regard radar (DRR) on board the MQ-9 Reaper.
The testing took place throughout December at the company’s Gray Butte flight operations facility and Edwards AFB, both in California. The MQ-9 flew a series of planned missions against small and medium-sized manned aircraft, during which the DRR tracked multiple targets while carrying out a wider surveillance of the surrounding area.
The purpose of the trial was to verify the DRR’s functionality on board the Reaper, as well as its integration with the traffic alert and collision avoidance system (TCAS) II that is used on commercial aircraft.
The whole due regard capability is now at technology readiness level (TRL) 7, NAVAIR adds, and is ready for customer operational test and evaluation. The company’s broader system, which integrates the DRR and TCAS II capabilities, is currently at TRL 6.
The integrated sense-and-avoid system will fly aboard NASA’s Predator B-derived Ikhana in 2015 in support of a series of flight tests that will measure the performance of the entire system in a variety of situations.
Technology for collision avoidance does not necessarily have to sit on board an aircraft. The US Army installed the first Ground-Based Sense and Avoid radar in December at Fort Hood in Texas where it bases two General Atomics MQ-1C Gray Eagle UAVs.
Previously the army used visual observers – on the ground or in a chase plane – to provide the necessary sense-and-avoid capacity required by federal regulation, it says. GBSAA will initially support UAVs transiting from airfields to restricted areas where training and testing can occur.
The radar system consists of numerous elements, including multiple 3D radar, data fusion, tracker, as well as classifier and separation algorithms that have been designed and developed solely for sense and avoid.
Fort Hood is the first site to receive the system and the hardware will continue to collect data for a safety analysis and report before becoming fully operational in 2015, the army says. Collecting and analysing the data will allow operators to verify if the radar is seeing everything and give the safety team a good understanding of the airspace traffic.
On the European side, Sagem has also developed its own SAA system, which it tested with its Patroller UAV in an 11-day flight-test campaign in October and November.
Sagem teamed with the French air navigation and safety agency DSNA, the laboratory of the national civil aviation school (ENAC) and Rockwell Collins France to carry out the research under the Single European Sky ATM Research initiative.
To test collision avoidance, ENAC provided a dummy aircraft to act as the airborne obstruction for the sense-and-avoid system to detect and circumvent.
It is made up of a series of sensors including an electro-optic/infrared camera and an automatic risk collision estimation and avoidance flightpath generation module. It was tested in different scenarios and demonstrated an ability to avoid collision without operator intervention.
In parallel the Operational Demonstration of RPAS in European Airspace (ODREA) testing enabled the team to test UAV airport approaches at Toulouse-Blagnac airport while adhering to air traffic control procedures, demonstrating that a UAV can operate near terminal buildings without having an impact on other airport traffic.
This testing led on from a simulation project carried out by the team, and was “a real first in Europe”, according to Sagem.
Source: Flight International