TIM ROBINSON reports on some of the highlights from 2012 ASTRAEA conference held in November as the second phase of the UK’s civil UAS project prepares to conclude in March. [caption id="attachment_7636" align="alignnone" width="322"] ASTRAEA is focused on proving the necessary steps for UAS and manned aircraft to share the same airspace. (ASTRAEA).[/caption] Six years ago, when the UK’s £62m ASTRAEA (Autonomous Systems Technology Related Airborne Evaluation & Assessment) UAV project was launched, few had even considered the potential civilian uses for UAS, UAV or, in popular public and media parlance, ‘drones’. Today, the situation is much changed. The ongoing war on terror has brought UAS (and the debates that accompany them) mainstream. Indeed, there are now a growing number of civil UAS stories appearing, from fears of privacy, to spotting pollution, to even a ‘name a drone’ contest with BBC’s children’s TV show Blue Peter. New peaceful applications, such as ‘drone journalism’, wildlife photography or using UAVs to monitor environmental concerns (such as ash tree dieback or countering rhino poaching), are coming thick and fast, emphasising the interest in this new disruptive technology from a variety of areas. All the signs are that a pent-up civil UAS market may be ready to explode, with predictions ranging from it being worth some $62bn a year, or up to $400bn annually, if all services are included. The future, then, seems bright.  

An over-hyped start?

[caption id="attachment_7637" align="alignnone" width="344"] Surrogate aircraft, such as this Jetstream have been used to test UAS 'detect & avoid' technology. (BAE Systems).[/caption] However, those within the community and outside observers are aware that there has been a previous ‘false dawn’. The formation of ASTRAEA in 2006, for example, meant enthusiasts predicted that, by 2012, UAS in civil airspace would be used to bolster Olympics security, or to help the police, coast guard and emergency services in the English Channel. Yet these proved over-optimistic.  One source within ASTRAEA confirms that the programme perhaps peaked too soon — with potential operators who were excited by the possibilities of UAS soon disappointed that the airspace regulations and ‘sense and avoid’ challenge remained obstacles.  

The ASTRAEA consortium

However, ASTRAEA was never set up to produce a commercial ‘approved’ UAS that could be bought off-the-shelf. Instead, it has been investigating and exploring the technologies and procedures needed to allow UAS and manned aircraft to co-exist safely in the same airspace. It has to be remembered that, while manned aviation has evolved organically over the past 100 years (and in parallel with the development of airspace rules), UAS have to integrate seamlessly with the existing airspace regulations. ASTRAEA, which consists of aerospace companies such as BAE Systems, Thales, Rolls-Royce and QinetiQ, as well as SMEs and academia, was set up to tackle this highly complex problem. Yet, in parallel to the work done by ASTRAEA, there has been an explosion in the smaller UAS types for civil uses. These light UAVs under 150kg and line-of-sight only are now rapidly proliferating and are being used as subistitute cherry pickers, cranes or ladders for aerial photography, inspection and other tasks. It could be argued that ASTRAEA’s goal, to allow the routine operation of UAS in civil airspace, has already been achieved. This is for the mini- and micro UAV market, of which some 170 operators now been licensed by the UK CAA, using existing rules. This initial market experience is feeding into ASTRAEA’s work on larger UAS.  

‘Virtual certification’

[caption id="attachment_7638" align="alignnone" width="374"] Cassidian used a small fleet of Minis to test a highly robust mobile communications and data network suitable for civil UAS command and control. (ASTRAEA)[/caption] Key to ASTRAEA’s approach in attempting to understand such a complex, interrelated challenge has been ‘virtual certification’ and breaking the required tasks down into the novel parts unique to UAS to be solved. For instance, tests by BAE Systems and Thales have used turboprop aircraft simulating UAVs to test sensors and computer algorithms for the ‘detect and avoid’ challenge. Meanwhile, Cassidian installed radios in Mini Coopers (and a Welsh steam train!) in order to test an ultra robust, secure mobile network that is able to dynamically reconfigure on the fly to ensure data and communications get through. This capability of dealing automatically with lost or degraded links, is expecially critical for civil UAVs. Another part of ASTRAEA has seen Cobham investigate the challenge of autonmous air-to-air refuelling with a probe and drogue system. Simulations have now progressed to a physical ground rig using two robotic arms to test the relative motions. If successful, this could be a major enabler for indefinitely airborne UASs — perhaps to undertake long-duration scientific or ocean monitoring.   

In-built airmanship

[caption id="attachment_7639" align="alignnone" width="371"] Sophisticated shared decision making software tools for civil UAS could also benefit manned aircraft. (ASTRAEA)[/caption] Another presentation during the day looked at the UAS computer ‘brain’ or Autonomous Integrated Mission System (AIMS). This would comprise ‘detect and avoid’, diversion planning and weather awareness, as well as  seeking to include elements of ‘airmanship’. For instance, not only might the autonomous pilot ‘know’ places for an emergency forced landing (e.g. football fields, or parks) from internal geographic databases, it would also be able to cross check with an IR camera to avoid the ‘car boot sale’ problem of humans suddenly being present in an open space. ASTRAEA has also been looking at giving the AI ‘icing awareness and avoidance’ — something that, tragically, human pilots still have difficulty with.  Indeed, ASTRAEA is not just about the technology but also about the procedures and new ways of thinking needed. One fascinating presentation during the conference looked at the issues of human factors and UAS operation, to test the hypothesis of how many UAS?air vehicles can a human safely control? Simulated flights in a synthetic environment allowed the tests to incorporate multiple ATC regions and handovers, shift changes and even pile on the pressure with a malfunctioning UAV. Exploring and de-risking these issues of ‘trust’ in the machine are crucial to regulatory and operator acceptance.  

Progress elsewhere

[caption id="attachment_7641" align="alignnone" width="384"] ASTRAEA is not alone in tackling the challenges of UAS in civil airspace. (ASTRAEA).[/caption] But ASTRAEA is not alone. There is also increased interest around the world in  unlocking this potentially lucrative civil UAS market. In Europe, for example, MIDCAS project tests are aimed at developing a UAS ‘sense-and-avoid’  system, with manned surrogate UAV flights to lead to a demonstration with an Alenia Sky-Y. Meanwhile, in the US, which along with Israel has led the field in military UAS development, the situation is more unclear. The FAA had been mandated to prepare for the introduction of shared airspace by 2015. Indeed, the plan was for six national UAS test sites to be selected, to allow for a variety of different environments for UAS to be trialled. This, however, has now been postponed. Finally, as recently as November, a new website for the JARUS (Joint Authorities for Rulemaking on Unmanned Systems) group went live at www.jarus-rpas.org. The aim of  the group, which is drawn from national aviation regulators and EASA, is ‘to recommend technical, safety and operational requirements for the certification and safe integration of UAS/RPAS into airspace and at aerodromes.’ Clearly much groundwork has already been done by ASTRAEA.  


Yet, there still remain challenges to unlocking solutions. Part of this is maturity of the technology (currently TRL6, needs to be TRL9) but others have arisen due to the efforts of ASTRAEA and other programmes thinking more deeply about the civil use of UAVs. For example, ‘equivalency’ — the idea that UAS in controlled airspace should be treated no differently under regulations than manned aircraft. However, this itself presents paradoxes. Why, for example, would a UAS operator sat on the ground still need a pilot’s style medical? And, since all UAS flights are essentially IFR, does the VFR airspace class even make sense with their introduction?  

Public perceptions

There is also the thorny matter of public perceptions and mislabelling that may hamper general acceptance of civil UAS. The connection in the public’s mind between military surveillance and/or armed platforms, such as Reaper, with civil UAS is especially strong — not helped by the anti-drone lobby who believe all uses of this technology are worth opposing. In addition, the fear of flying robots, ‘killer drones’ or airborne ‘terminators’ is also present and difficult to counter. Indeed, at the conference it was admitted that the industry had not done itself any favours with its choice of terminology. While engineers may be happy with using ‘autonomous systems’ to describe aerial vehicles which could in some sense be classed as merely having a very sophisticated autopilot, one suggestion was that ‘shared decision making’ between the human/UAS might be a better term. Having come up with the concept of ‘autonomy’, the UAS sector now finds itself rowing back somewhat to explain that true fully autonomous systems are both undesired and unachievable — at least in the near to medium future. As Lambert Dopping-Hepenstal, ASTRAEA programme director notes: “Although engineers use ‘autonomy’ to describe future UAS operations — there will always be a human supervising.” The changing professional terminology from UAV, to RPAV to UAS has also allowed the use of ‘drone’, with its unthinking, cyborg connotations, to take firm hold in the media and public. Allied to this, recent news stories about technology and privacy (e.g. phone-hacking) means that any legal and ethical issues need to be addressed up front. Says Dopping-Hepenstal: “We’ve got to engage the public.”  

What’s next?

[caption id="attachment_7640" align="alignnone" width="403"] Will an ASTRAEA Phase 3 see a greater amount of research move out of synthetic environments? (ASTRAEA).[/caption] The second phase of ASTRAEA is set to conclude in March, leading to speculation as to whether the project will be continued or whether it will fall victim to the funding ‘valley of death’ — the gap between R&D and bringing products to market. So far, the tests have included simulated flights in synthetic environments, testbench demonstrations and surrogate uses of manned aircraft (or even, in the case above, steam trains). The next obvious step, according to Dopping-Hepenstal would be a national flying UAS demonstrator. Yet, there also may be other spin-off benefits beyond just opening civil airspace to UAS. Telemedicine or robotic surgery, for example, might be one spin-off, or new applications for the airborne/mobile dynamic reconfigurable IP network. Finally, the work done on the autonomous decision support system, for example in detecting icing, could have major benefits for manned aviation, perhaps giving GA pilots a ‘virtual co-pilot’ to assist in emergencies or bad weather. ASTRAEA has wider benefits than many might think.  


In summary, as phase 2 of ASTRAEA concludes, there is much to be excited about, despite the challenges still outstanding. Essential research in ‘detect and avoid’, robust communications networks and human factors, to give just three examples, along with the whole ‘virtual certification’ means that many more the issues are now being understood. But, more importantly perhaps, is that the word is being spread — civil UAS use is now firmly on the public’s radar and wider interest is growing. The applications then, that this disruptive technology may one day allow (e.g. Amazon parcels delivered directly to your location) probably haven’t even been thought of. As one attendee noted, when the question of ‘whether there was really a market’ was asked: “Henry Ford was told that there would only ever be a market for 100 automobiles, since that was the number of chauffeurs in New York city.” It may seem to be cautious but ASTRAEA’s step-by-step approach is key to safely unlocking this new untapped future for aviation.

Tim Robinson
4 January 2013