Following a twentieth century that led to the dominant role of aviation in military operations, the twenty-first is sure to see a notable increase in remotely-piloted military aviation. These airborne systems, often at a lower cost of ownership than their manned equivalents, were originally designed to carry out missions considered ‘dull, dirty and dangerous’. In the air, their main raison d’être was to gather intelligence but their use has rapidly expanded as a result of progress made over the past twenty years, and they have become truly multi-role intelligence assets and valuable tools for tactical coordination. The Air Force now has considerable experience in the use of intelligence-gathering drones, and in particular, theatre or MALE drones.⁽¹⁾ Carried along by technological developments, remotely-piloted military aviation is moving progressively into new fields. Despite this increased call upon technology, man must remain the guarantor of supervised employment of his combat system and of its ethical use, and hence he must receive training appropriate to these challenges.

Development of drones and the advent of permanent airborne reconnaissance

Although from the nineteen-seventies on, some other forces operated drones like the Nord Aviation CT.20, Canadair CL-289 and Sagem Crécerelle, these unmanned vehicles then followed a flight plan programmed on the ground. Their mission was IMINT⁽²⁾ reconnaissance of sites in support of the artillery, looking for enemy military units. The mission was conducted in contested airspace and remained conceptually close to the reconnaissance missions of the First World War. Here, the drone replaced man in what was considered ‘dangerous’.

The Air Force’s first steps towards the use of drones were made in the nineteen-nineties with the Israeli Hunter system. It was fitted with an optronic ball and a laser illuminator but was limited by its communications system to an operational range of 200 km (125 miles). Its heavier successor, the EADS/IAI Harfang, announced our country’s membership of the MALE drone users’ club, freeing us from limitations of operational range linked to line-of-sight communication. With this type of aircraft dedicated to low-intensity or asymmetrical operations, the low level of air threat in some theatres led to designing an vehicle ideally adapted to long-endurance surveillance with medium-altitude flight for discretion and reduction in fuel consumption, low speed and straight, glider-like wings to increase endurance, a dome for the satellite aerial to allow piloting beyond line of sight (BLOS) and real-time transmission of information. Its level flight, low speeds and light load meant that the structural reinforcement needed for a combat aircraft, for example, could be dispensed with, allowing a greater fuel load. Moreover, the absence of aircrew saved weight and overcame the human physiological limits of over 20 hours’ flying time. This type of drone therefore covered the ‘dull and dirty’ aspects.

Increased activity of the various types of drone operated by the Air Force is evidence of the success of the surveillance drone concept. After the Hunter (1,500 flying hours), the Harfang saw a dramatic increase in activity (15,000 hours in 10 years), which the MQ-9 Reaper built upon (25,000 hours after five years of use). In operational theatres, the need to gather intelligence totally discreetly, and without any change in the environment, is continually increasing. The increased capability of the drone that allows the permanent surveillance over areas of interest means that several crews are needed for each to allow for reliefs: this presents a major challenge to find the number of operational crews required. The next step will be the arming of drones in order to seize opportunities to act or to protect without delay during these surveillance phases. Such a capability was initiated by the US Air Force on its MQ-1 Predators at the beginning of the century.

Beyond the increase in drone activity lies a revolution in their use, which has expanded from the simple, occasional reconnaissance mission to seek out military materiel to that of constant surveillance of an entire environment, including the people in it.

Since operations in Afghanistan, all theatres now set us against irregular adversaries in counter-insurrection and anti-terrorist operations. Modes of action employed by the adversary include dilution among the population, hiding of depots and training centres and the combined use of rudimentary material, civil technologies such as GPS, and modern communication methods. It is no longer a question of simply performing reconnaissance of military materiel to detect and identify them to establish an order of battle but of using the persistence of long-endurance drones to survey areas and track individuals over long periods in order to understand their organisations and intentions. The operators’ work therefore centres on the use of wide-field sensors to detect activity in incompletely covered areas, then establishing the patterns of life of individuals or groups of people with narrower-field sensors. The lack of anti-air threats coupled with this need for surveillance over the long term has ensured the primacy of the drone in intelligence gathering for low-intensity military operations. But when the adverse ground-air threat reappears, as it did in Georgia in 2008 and Ukraine in 2014, the use of this type of drone becomes far more complicated because of its vulnerability.

From surveillance to strike

In the case of asymmetrical operations, permanent occupation of airspace by surveillance or strike assets puts the enemy under immediate pressure and precludes any mode of action that is too obvious. For this, the MALE drone is a game changer, since it upsets the element of uncertainty: the insurgent who previously could secretly prepare an ambush by assembling forces or planting an IED without being seen, now risks permanent discovery. Arming of these aircraft has added a capability for immediate action that is complementary to other aircraft such as fighters and helicopters and allows for immediate strike if the latter are unavailable. The long endurance drone also allows the choice of the most opportune moment for an air attack against a transient, mobile enemy who exploits urban areas for hiding and therefore avoiding strikes because of the presence of civilians. In this, the rules of engagement that apply are identical for both manned aircraft and those whose crews are ‘remote’.⁽³⁾

The crew at the centre of the MALE drone’s capability

A crew of four people man a French Reaper for the conduct of its missions: a pilot, a sensor operator, an image interpreter and a mission commander. The first two take care of what is happening in the present time whereas the other two analyse the information gathered and prepare future action. During long missions, several crews relieve each other in order to remain attentive and efficient. The environment of a cockpit on the ground, being far less constraining than that of a combat aircraft, means much more analysis equipment can be incorporated; liaison officers can be present, too, in the case of certain operations. The crew is connected to different levels of command of forces via a wide array of methods and also has to hand an unequalled range of digital and informational resources to support its analysis capability, which often gives it a central role in intelligence and coordination of action.

Lethal operations of MALE drones are regularly challenged, broadly on the basis of two arguments, the first being the ease with which they would be able to achieve targeted neutralisations outside any standard legal framework. Yet the means (the drone) must not be confused with the one giving the order. The second argument is the ‘robotisation’ of the system, since these drones would lead to automating the decision-making processes leading to any neutralisation, and therefore dehumanising them. Now, drones like Reaper call upon certain automatic flight functions only—and even they are quite basic compared with those in airliners: the important functions of intelligence and targeting are entirely conducted by the crew. Man is permanently in the loop in real time, from identification of the target through to the strike, which is far more than for a cruise missile strike or artillery fire onto map coordinates. This advantage of the man in the loop has also led to a significant proportion of attacks conducted by combat aircraft and helicopters in the Sahel being guided by drone crews. In this way, the technological developments that are entering service will contribute to boosting further the role of the unmanned aircraft in air-ground missions in permissive environments and to diversifying their use.

Rapid progress and more varied uses

There is considerable technological profusion in the field of surveillance drones, both small and large, and there is rapid progress in conceptual fields—those of new uses, platforms, automation of piloting, recovery, information handling and insertion into airspace. The Air Force is also interested in small drones for reasons that include a less rigid regulatory framework that facilitates innovation, civil and military uses generating real dynamism, and rapidly increasing performance in endurance, flight envelope and sensor quality which usefully complements that of the MALEs. It is envisaged that they could cover the particular operational needs of the Special Forces, as well as protection of sites, the anti-drone battle and exploration of technologies essential to future swarm systems. The essential tenets of their employment—since they are common to all drones—would also form the basis for operator training.

Much progress is also being made on optical and radar captors, or sensors, from the original narrow-field to the wide field now afforded by systems such as Wide Area Motion Imagery—WAMI, and also means for listening, systems for integration in airspace, armament and, of course, connectivity. In parallel with them, new capabilities are appearing: equipment for offensive EW, powerful lasers, acoustic buoys for maritime surveillance and more. Automatic data handling, especially through the use of Artificial Intelligence (AI), will also ease the crews’ analysis work and their manipulation of sensors, which will lead to their even greater effectiveness.

The most symbolic area of interest, and with great potential for growth, is clearly that of unmanned combat air vehicles (UCAV). Nevertheless the survivability of platforms poses a crucial challenge that requires major compromises to be made in their design comparable with those of combat aircraft, and very precise coordination with other aircraft in flight. The timescale for the arrival of these drones is rather longer.

To be able to take on these developments, the MALE is, and must remain, a low-cost carrier of high-tech captors, or sensors: the flying vehicle itself does not present any major technical challenge in terms of flight envelope or carrying of passengers, in contrast to the captors and communication systems. The interest lies in fitting the widest possible variety of captors on a single platform, such as ELINT, SAR-GMTI⁽⁴⁾ and optronic balls, thereby sharing implementation costs and greatly increasing the available capabilities for conducting a mission. It follows that it is essential to design platforms with reserves of volume and energy so that they can be fitted with extra facilities later during their service lives and also to reduce the time needed for qualification in order to introduce new technologies more rapidly. All these technical developments will lead to a greater range of their activity, which will include coordination of reconnaissance and strike missions, jamming, maritime surveillance, destruction of some short-range ground-air sites, logistic transport and in-flight refuelling.

There is greater diversification in the use of mini drones, too. The forces are using them more and more to improve their capacity for surveillance and investigation and to guide commandos in their action. Aside from use in theatres of operation, the Air Force is acquiring them for use in ensuring protection of its sensitive sites.

The drone will also lead to the use of airspace virtually devoid of aviation today: the stratosphere, for example, offers undeniable operational advantages from which the Air Force could benefit. At these high altitudes of over 20 km (66,000 feet) it is possible to maintain electrically powered drones in flight for several weeks and in a few days reach areas far removed from the starting point. Since functions such as communications relays would become conceivable, they could complement our space capabilities over permissive or semi-permissive theatres.

This extension of the operational domain of unmanned aviation means a rethink of captors is necessary to ensure multi-sensor integration from the lowest levels of information gathering and instantaneous retransmission of pertinent intelligence both to the forces and to the highest strategic levels.

In spite of this profusion of innovations, the ability to use unmanned aviation in as flowing, flexible and reactive a manner as a manned aircraft does not exist other than with mini drones. Integration of drones into ever more congested airspaces and into air traffic generally therefore requires major effort to ensure the military freedom of action of our systems. The technological trend is also towards the reduction of some of the barriers to the use of small and large drones in terms of altitude, radius of action, autonomy and therefore of employment, even if the conditions of implementation related to their size will continue to maintain a degree of segmentation. One might then question what can be automated and what should be kept under human control? The question of automation is closely linked to the training of the crews.

Training: the crucial challenges

The first attempts at regulation⁽⁵⁾ segregate drones by their size, even though technological developments are sure to render such segregation obsolete by the increasing ability of small drones to reach higher altitudes—over 2,000 metres (6,500 feet). Moreover, a large part of airspace is densely used and is subject to strict rules of navigation both over national territory and on operations, for which the operators have to receive adequate training. Piloting of unmanned systems must therefore be reconsidered: such is the objective of the Air Force’s Cyclope project to create a joint Centre for drone crew initial and continuation training (Centre d’initiation et de formation des équipages de drones—CIFED) on its air base at Salon-de-Provence. This will offer common aeronautical training independent of the size of drones but adapted to the types of airspace in which they will operate and to the intended mission profiles. It is essential for air safety that airspace is managed according to common rules, hence the need for training that instils the fundamentals of coordination so that the operators can launch their drones into controlled airspaces.

The second element of training consists of specialised instruction on the conduct of systems, to give drone operators the ability to detect attacks on the integrity of their systems. Resilience will rely more than ever on the airmen at the controls who masters the precise functioning of his system: he or she is the one who can recognise failures and can take back manual control when faced with potential threats such as cyber attack, jamming, directed energy weapons, jamming of GPS and physical strikes. Cognitive science and ergonomics will be fundamental here: experiments being conducted at the Mont-de-Marsan Air Warfare Center (Centre d’expertise aérienne militaire—CEAM) need to put the human cognitive factor at the centre of man-machine interface design.

Man must also be at the centre of digital transformation, which is intended to add value for those who benefit from it—he armed forces and the centres of decision-making. Connectivity, collaborative combat and data links must not end up in a position where the data scientist or some other person in the development team becomes responsible for military action simply because he designed the AI-based interpretation system. Man must remain at the heart of the system, so that he is the decision-maker. The ergonomics and degree of autonomy afforded to the unmanned airborne vehicle must retain a human at the heart of the system—in the loop—so that through his combatant’s ethics he is the guardian of the use of his combat system and of the level of violence employed.

* * *

All these developments are promising, but we need to remain modest in any prediction of the future of military drones around 2035, since 90 per cent of the innovations to come in the fields of AI and robotics are still unknown. That said, there is a distinct trend towards increased use of drones in air, land and sea military systems. The over-complexity of systems, cyber threats and disputes in extra-atmospheric space (regarding communications and navigational positioning, for example) will have effects on all things entirely remotely operated or automated. We therefore need to learn how to manage the evolution of the technologies concerned and to guide the men and women who will operate these systems. The challenges to safety and security of operations will require robustly designed systems and training of military personnel in aeronautics, conduct of remotely-controlled systems and the ethics of the military in operations and, moreover, the time and means needed to do it. ♦

(1) Medium Altitude, Long Endurance.
(2) Intelligence from imagery sources.
(3) Florence Parly, Minister for the armed forces, said in her closing speech of the Université de la Défense on 5 September 2017, that she wanted from the outset to refute any possible confusion and to dispel possible fears. No, an armed drone is not a killer robot. They are two systems quite unlike each other. This decision [to arm Reaper] changes nothing in the rules concerning the use of force with regard to the law of armed conflict. (www.defense.gouv.fr/).
(4) ELINT=intelligence from sources of electromagnetic radiation/transmission; SAR=Synthetic Aperture Radar; GMTI=Ground Moving Target Indication.
(5) Decree of 17 December 2015, concerning the use of airspace by aircraft circulating therein with no person on board (www.legifrance.gouv.fr/).