Domesticating the Drone: The Demilitarisation of Unmanned Aircraft for Civil Markets

RPAS refers to a system, extending beyond the unmanned vehicle to include ground stations (where control units and remote pilots are based) and communications infrastructures. Within the broad definition of RPAS lie a diverse range of systems and vehicles. Some differences between these RPAS are immediately apparent, such as the size or weight of vehicles. Other differences are more subtle, such as the medium of communication between the vehicle and the ground station. These technical differences are particularly important in regulations and the logistical management of RPAS, including the definition of the phases of airspace integration.

The distinction between RPAS and other UAS is defined by the extent of autonomy and automation delegated from the pilot to the system. Automation levels range from those that are fully piloted from a remote location to those that are fully automated. There are also several points in-between, with some manoeuvres triggered automatically through autonomous monitoring of conditions. Depending upon system priorities, autonomous manoeuvres may have priority over, or be overridden by, the commands of a remote pilot. The International Civil Aviation Organisation (ICAO) and current EC plans will only permit autonomous manoeuvres to override pilot command in extraordinary circumstances such as communication failure or imminent collision risk. The UAS technologies beyond this definition, featuring greater autonomy, are also quite well developed and, while integration is not currently planned, it could plausibly follow a successful period of development in the RPAS sector.

RPAS vehicles are often organised by their weight, with 150 kg defining the boundary between heavy and light vehicles. In Europe, light vehicles are regulated at a national level while heavy vehicles are regulated at a European level. The 150 kg threshold has been described as arbitrary (Finn and Wright 2014). The difference in regulatory treatment above and below this threshold could fragment development and, ultimately, have a distorting effect upon the market. Two variables are often used to capture the complexity of RPAS operations; the operating altitude and the line of sight (LOS) from the remote pilot station (RPS). Unlike the weight threshold, these are not arbitrary but capture the complexity of the operational requirements for flight. At low-altitudes, up to 150 m (below the level where manned aircraft fly), RPAS are further differentiated by their LOS from the RPS. In the simplest scenario, the vehicle remains in direct visual LOS from the RPS, usually within 500 m. Where this is not possible, extended visual LOS describes operations where the vehicle is not directly visible to the pilot and a support crew based in a second location is used to maintain visual LOS. Where this is not possible, the situation is described as beyond visual LOS, and support technology is required to monitor movement. At high-altitudes, 150 m, RPAS may share airspace with manned aircraft and so safe integration is crucial making the situation more complicated than lower altitude flights. In a radio LOS, a direct connection is maintained between the vehicle and the RPS. Where this is not possible, beyond radio LOS, more sophisticated communication channels such as satellite are required.

These distinctions are based upon the specifications of the system, and are often used to support the management of the technology in regulatory or logistic contexts. Many discussions about RPAS, however, are less focussed upon the physical specifications of the technology and more upon its applications, as well as the motivations for and consequences of their development. The uses of the technology are also particularly important when evaluating it in terms of ethical principles, social values, and grand visions such as RRI. Applications of RPAS can be organised into four major categories; military; non-military governmental; commercial; and personal/recreational. There are several examples within each category, some already routinely performed, many at a pre-market demonstration stage and others in the category of predictions, promises and imaginaries of varying plausibility. While an exhaustive list would be unwieldy and futile, the following paragraphs present an overview of each category of use with some key examples.

By far the most visible and debated applications of RPAS fall under the military category (M-RPAS), particularly surveillance and combat applications in the War on Terror. M-RPAS can be piloted from an RPS at the other side of the world, distanced from immediate danger, as well as the mechanisms and consequences of their actions. This clearly presents a lower risk to combatants, a key advantage of M-RPAS over manned aircraft. However, some critics have argued that they may contravene the just war principle that an army should not risk reduce risk to their own soldiers by means that increase risk to civilians within enemy territory (Billitteri 2010). Others have suggested that their efficiency makes military operations more attractive, effectively promoting a more aggressive foreign policy, including military action in foreign territories that are not officially defined as warzones, such as Pakistan and Yemen (Sparrow 2009). Counterarguments hold that military drone strikes comply with international law and are justified by the principle of self-defence and are an appropriate response to the characteristics of modern warfare, whereby enemy combatants are not readily distinguishable from civilians (Billitteri 2010). These, and several other ethical and legal arguments for and against M-RPAS, are often debated in policy, academia and the mainstream media (for a European summary, see Dworkin 2013).

The state may also deploy RPAS in non-military, civil operations. These C-RPAS are most often associated with security and surveillance applications, such as policing and border patrol. Other areas where governments have used C-RPAS include forest management, fire services, air sampling, search and rescue, and infrastructural maintenance. The motive is often value for money, with C-RPAS achieving the same (or higher) performance standards as manned aircraft more efficiently and economically. Furthermore, they present options for performing operations in dangerous or uncertain contexts, such as emergency response, with a lower risk to personnel.

Commercial and industrial applications could include internet provision, agricultural monitoring, entertainment, advertisement, security and surveillance, infrastructural maintenance, and delivery services. Some applications offer new capabilities, while others allow safer, more efficient, or ‘sexier’ performance of existing tasks. RPAS are also well suited to so-called ‘dull dirty and dangerous’ where unmanned solutions are preferred. Some such applications have achieved media interest, such as the rather fantastical, headline-grabbing reports about drone delivery services for pizzerias and bookshops. These stories have attracted some mainstream debate about privacy, safety and liability.

Several applications of C-RPAS are outside commercial and state activities, such as recreation, creativity or some other community or individual purpose. Many enthusiasts are organised into substantial communities with websites, forums and meetings.² Many participants appear to have the technical skills to build and modify their own C-RPAS from prefabricated kits or their own original designs, depending upon their level of expertise. These communities are generally supportive to other members new and old, and an ‘open source’ attitude to innovation is adopted, whereby ideas and solutions are shared across open platforms. Some such RPAS applications have community oriented aims, such as education, creativity, empowerment and activism. Despite the legality of personal uses of RPAS often being unclear, there is a growing market for off-the-shelf products for responsible and rogue users alike.

There is substantial crossover amongst these categories of applications and technologies developed and sold for one purpose can be used for another. For example, an RPAS designed to monitor air may be used by the military to identify potential chemical threats to ground troops, by the government to ensure compliance with air quality regulations, by industry to provide information and forecasting services and by individual ‘citizen scientists’ to monitor their industrial neighbours. RPAS base vehicles are highly customisable and can be configured with different payloads to deliver various marketable services for government, commercial, recreational and military purposes. Several niche applications that may not otherwise have attracted sufficient investment to fund their own development are made financially viable by the provision of a core technological base that can be modified for a wide range of mainstream and niche purposes. An illustrative product is Microflown AVISA’s vector sensors, which can detect sounds, locate their source and initialise actions. The devices are promoted for landing support and mid-air anti-collision systems, valuable in many application domains, as well as for target acquisition systems, detecting gunshots and artillery fire and pinpointing their sources as potential targets (Microflown AVISA 2012). It is clear that innovations such as these, as well as greater miniaturisation and economies of scale, will provide benefits across a wide range of RPAS applications for both civil and military purposes.

In Europe, C-RPAS above 150 kg are controlled at a European level by the European Aviation Safety Agency (EASA) while those under 150 kg remain under national rules. While some Member States authorise flights in normal airspace, the EC has responded to pressure to support the development of a European C-RPAS market by harmonising rules. This started with a three-part consultation on RPAS from 2009 to 2012. The first stage being a hearing on light unmanned aircraft led by the erstwhile DG Energy and Transport, the second a high-level conference organised by the EC with the European Defence Agency and the third a set of five workshops organised by DG Enterprise and Industry (DG ENTR) and DG Mobility and Transport (DG MOVE) in 2011–2012. Following these consultations, the insights were brought together into a Working Document (European Commission 2012), and the European RPAS Steering Group (ERSG) was established to plan a series of activities that would lead to the initial integration of RPAS in European airspace by 2016. The ERSG roadmap (ERSG 2013b) was delivered in June 2013, setting out several challenges, issues and potential remedial steps. It is not law, but a unified call for action produced by a range of key stakeholders under the guidance of several Commission services, wherein it remains a key document. The following subsections describe these steps in more detail, highlighting the development of ideas raised early in the consultation process to actions recommended in the roadmap, in particular the strategies for managing public acceptance described in its appendix on social issues (ERSG 2013a).

As a dual-use technology, the innovation and economies of scale anticipated for C-RPAS will also benefit the M-RPAS sector. As mentioned in previous sections, these synergies were recognised early in the EC consultation process and were encouraged by a wide range of actors. In this section, we explore this recognition in greater detail. First we revisit the C-RPAS consultation period to highlight how dual-use was encouraged throughout the process leading to the publication of the roadmap before exploring narrative and structural support for dual-use in wider settings; military analyses of C-RPAS and recreational RPAS activities. We do not consider these synergies between C- and M-RPAS, nor the recognition of them, as critique in themselves. Rather, we present them as a point of contrast with the strategies for managing public acceptance.

Alternative visions of C-RPAS, to replace that of the ‘killing machine’, are not articulated, although we may draw some inspiration from the two other motives and expected consequences of C-RPAS development (aside, that is, from their benefits to the M-RPAS sector). These are their functional benefits and economic potential. The first can be captured as a Robot Helper vision of C-RPAS as assistants in the tasks that are too dull, dirty, dangerous, difficult or expensive for human presence. Here, RPAS are the enablers of services that deliver social value more effectively and at a lower cost. The second can be captured in a New iPod vision of C-RPAS as a revolutionary technology that will spearhead innovation and the inception of new services in other sectors. These predictions go beyond direct, specific advantages—such as improvements in communications or aviation technologies—extending to further unspecified spin-offs on a scale described as beyond prediction through comparisons with other technological milestones including space flight (European Commission 2009), the iPod (European Commission 2012) and the internet (European Commission 2014d). These New iPod and Robot Helper visions are quite compatible and are often deployed together, for example in a recent Communication from the EC to the European Parliament and the Council (European Commission 2014a).

While the roadmap does not develop detailed plans for replacing the killing machine vision of C-RPAS, it does introduce some strategies for public communications that, they anticipate, could reduce the possibility of an adverse public reaction. At this point, the roadmap shifts from its previous focus on the different technical specifications of C-RPAS to their different possible uses, arguing that “it is likely that the public will make a distinction between those operations that provide a “greater good” (for example fire fighting or search and rescue) and those operations, which have only limited community benefits” (p. 39). The strategy is further elaborated, underlining the need for better explanation of the potential benefits of C-RPAS by highlighting some uses that are considered more benign while avoiding discussion of potentially controversial uses. For example, they recommend the C-RPAS community to “stress the roles RPAS have in conducting humanitarian operations or in testing for airborne toxins, rather than focusing only on the military and security applications” (pp. 39–40). Finally, it recommends establishing an ad hoc RPAS public acceptance facilitator group, which would be tasked with understanding public perspectives on C-RPAS, conducting short surveys, and identifying the means to gain their acceptance.

Strategies for managing visions of C-RPAS may include strategies to manage the vocabulary associated with it, such as refraining from using the word ‘drone’ because it is considered to be too strongly associated with controversial military operations and that this association would damage the C-RPAS sector. In some quarters, the ‘d-word’ (as coined by Gosztola 2013) has become taboo (Grossman 2013), and its use is even disputed in military contexts (see Hammond 2013). Communities of enthusiasts appear to recognise the role of vocabulary in broader visions of technology, and have debated strategies of how to refer to C-RPAS in this context.⁶

Many actors have attempted to manage the vocabulary, often with disappointing results. The Association for Unmanned Vehicle Systems International (AUVSI), which lobbies on behalf of both M-RPAS and C-RPAS sectors, has organised a workshop titled Privacy, Ethics, Perception and Dealing with the “D” Word (AUVSI 2014). Their president argued that the “average person on the street, and even intelligent and informed people, when they think of the word ‘drone,’ they think of the military, they think hostile, they think weaponised, they think large and they think autonomous” (Whittle 2013). Perhaps unsurprisingly, these efforts have often failed, and have collectively yielded several articles about the ‘drone industry that dare not speak its name’. There are several specific examples of attempts to convince journalists to avoid using the word ‘drone’ backfiring. For example, a CEO’s pleas for a journalist to avoid the dreaded nomenclature being cited in the article’s opening line—“whatever you write, please don’t call it a drone” (Terris 2013). These episodes illustrate how difficult it is to control language, particularly in the presence of media interest. When proponents are sensitive to the connection between C-RPAS and M-RPAS, they attract attention to the very thing they wanted to disguise and, perhaps more damaging, they attract attention to the fact that they attempted to disguise it.

The EC have mostly referred to RPAS and UAS in their consultation documents and ‘civil drones’ in public-facing communications (e.g. European Commission 2014b, d, f). Some other policymakers, however, are also considering the role of vocabulary as an important part of the wider vision that is communicated to publics. For example, a UK parliamentarian stated that “using the term drone on smaller UAVs muddied the debate” and “invoked images of the large scale UAVs operated by the military” (All Party Parliamentary Group on Drones 2013), and a UK Defence Committee consultation on military and civil aspects of RPAS development positioned nomenclature at the top of the agenda; “defining the terms RPAS, UAS and “drone”” (Defence Committee 2013).

The propriety of M-RPAS and the ethics of dual-use technology development are important topics, but discussion of these issues is not necessary to show that the strategies for managing public acceptance of C-RPAS, as described in the roadmap, do not conform to the concept of RRI. The strength of the relationship between C- and M-RPAS is recognised by stakeholders across the board, with the benefits for M-RPAS consistently identified by policy actors as one of the motives for, and predicted consequences of, the development of C-RPAS. However this contrasts with the visions of and motives for C-RPAS development as presented in the strategies for managing their public acceptance articulated in the ERSG roadmap. As such, the strategies appear disingenuous, as they accept that citizens’ perspectives on technologies are important, but decline to present a full understanding of all the motivations for and consequences of development, most notably that a developed C-RPAS sector would benefit the M-RPAS sector. This problem can be captured in the context of RRI.

Applying RRI to specific research and innovation practices is not straightforward, although guiding frameworks are emerging from academic and policy circles. In the introduction, we outlined three key points from this literature. These are explored in more detail here with reference to the problematic strategies for managing public acceptance of C-RPAS. The first point was that responsibility goes beyond the processes or outcomes of a development, extending to the substance and transparency of its purposes and motivations. In developing a framework for applying RRI to specific projects, Stilgoe et al. (2013) described moving beyond questions about the products and even the processes of innovation, to interrogate the purposes. This means that, in addition to the direct outputs of an innovation process (the product) and the procedures associated with its management (the process), we must also consider the motivations for an innovation process (the purpose) and their transparency. Proper application of their ‘stage-gate’ method of assessing projects would require the cooperation of agents in control of C-RPAS development with an independent assessment panel. Nonetheless,  we can consider two of their criteria as particularly relevant here; clear communication of the nature and purpose of the project, and mechanisms identified to understand public and stakeholder views (Stilgoe et al. 2013, p. 8). These require the assumptions, commitments and framings of the project to be reflected upon and communicated in dialogic engagement with a wide range of stakeholders and publics, a process that has certainly not occurred during the consultation processes described here.

The second point was about the role of knowledge as a precondition for responsibility. A recent report (European Commission 2013) proposed that if the consequences of an action are not known—perhaps because of unpredictability or misinformation—then the actor cannot be held accountable for them. This ‘criteria excuse’ does not apply to deliberate ignorance or professional negligence, but it does mean that responsible innovations must be designed and managed in “such a way that it provides the relevant persons and stakeholders with relevant knowledge” (p. 57). While the strategies for managing public acceptance of C-RPAS described here have recognised the importance of public awareness and support, the approach has not been to present a complete and balanced understanding.

The third point was that pushing a technology without sufficient dialogue at early stages is irresponsible, and can damage development. In identifying types of irresponsible research and innovation, von Schomberg (2013) described technology push with the example of genetically modified (GM) food in the 1990s. Echoing the current discussion of C-RPAS, the irresponsible dimension he identified was not the GM technology itself, but insufficient dialogue amongst actors at an early stage and a regulatory focus upon safety aspects without considering the broader contexts, including social aspects. RRI requires greater stakeholder involvement at earlier stages of development. The innovator must respond to societal needs, which must be defined by societal actors. Such an approach requires the innovator and the societal actors to be co-responsible for the innovation process.

Considering the story of C-RPAS development to date in the context of these warnings, we can point to some specific problems as well as potential remedial action. The EC’s C-RPAS consultation process, described in the “C-RPAS Development in Europe” section, is well documented in a central portal which is regularly updated and includes details including the minutes of meetings and even the slides used during the presentations (European Commission 2014c). Contributions offered to the process were clearly influential in the structure and content of the eventual roadmap, but were limited to expert communities, largely from relevant industry and policy communities with some limited academic involvement. The use of a range of experts and documentation of the process is laudable, but it also exposes the notable absence of public consultation. A second point is that the recommendations for managing public acceptance of C-RPAS rest upon several untested assumptions, including that RPAS are seen as ‘killing machines’, that this is because of media hype, and that this will negatively affect acceptance of C-RPAS. In fact, we know very little about public reactions to C-RPAS as there have been no public consultations. The relatively small number of studies performed to date (in North America) indicate that citizens hold sophisticated understandings of C-RPAS that are not well captured as opposition or support but, rather, depend upon the specific application area and context of use (Bracken-Roche et al. 2014; Eyerman et al. 2013). While the ERSG strategy suggests studies of public attitudes, their aim would be to support the design of better PR campaigns to achieve acceptance more effectively, treating citizens’ objections as obstacles to be overcome (Hayes et al. 2014).

Responding to these points, we recommend a different approach to managing public acceptance of C-RPAS. At present, the burden of acceptability is with the citizen, the target of information actions that are designed to transform them from the passive role of opponent to the passive role of acceptor. In a responsible process, the citizen’s role would be that of a fully informed, active agent, (co-)responsible for development, and the burden of acceptability should be with the technology. As such, strategies to manage public acceptance should seek to transform the technology so that it is acceptable to citizens, not to transform publics into acceptors of the technology. Attempts to replace the killing machines vision with a more palatable alternative may be misguided, if not irresponsible. Following the lessons of those that sought to avoid the use of the word ‘drone’, it may also increase the chance of public opposition. Instead of changing visions, at this early stage of development, a responsible strategy would seek to promote transparent debate and dialogue about a wide range of visions and explore how publics can become part of the co-production of new visions and responsible development paths.