Starting with the Nagorno-Karabakh conflict, and continuing with the current Ukraine war, virtually anyone can see the effectiveness of loitering munitions (LMs) on the battlefield for themselves. Technologically, LMs have been mature for a long time, with early examples having entered service decades ago. In Germany, however, these types of munitions have long been undesirable for political reasons relating to fears around weaponising drones. Since the conclusion of the drone debate, the German defence industry has re-evaluated the potential of LMs and has already demonstrated its first systems. One such loitering munition is Diehl Defence’s ‘Libelle’ (ENG: ‘Dragonfly’), which the manufacturer claims possesses unique capabilities.
The term ‘loitering munition’ is used to describe remotely operated precision munitions which can be launched without specific target coordinates, then circle or linger over a target area for extended periods of time until a worthwhile target is detected, and then engage it. In addition to target engagement, loitering munitions can therefore also be used for reconnaissance purposes. Loitering munitions can be broadly divided into three different categories based on their design:
The first type corresponds to the design of a classic surface-to-surface attack LM. This category of LMs are typically associated with long range and loitering times, with relatively low optical and acoustic signatures. Such LMs can only attack their targets only at shallow dive angles, due to their aerodynamic construction. Obstacles and cover in the immediate vicinity of the target can therefore be used to protect against strikes from these types of LMs. Additionally, the reusability potential of these systems tends to be limited due to their reliance on a parachute or net for recovery.
The second LM design is based on a typical guided missile design, albeit featuring larger, X-shaped wings and propeller engines instead of a rocket motor. Such designs can be used to achieve both long ranges and loiter times, but are also capable of engaging their targets at very steep dive angles, permitting top-attack engagements against their targets, and therefore possessing greater potential to defeat against armour than their shallower-diving classically aerodynamic counterparts.
The third type is the copter drone design. These LMs systems can take off vertically and, if necessary, land again independently. Since young soldiers in particular may be familiar with copter-type drones from their leisure time, the controls are quite easy to train. These systems are capable of attacking targets directly from above, so positions in partial cover offer little protection. The ability to hover in place or land and wait on rooftops of buildings makes LMs of this design particularly well-suited for urban combat. The disadvantages of such rotary-wing LM systems includes their typically slower speeds, lower ranges, and shorter flight/loiter times (often <1 hour), and greater susceptibility to wind conditions compared to the first two LM types discussed. In addition to these, they are easier to detect acoustically, since both forward movement and lift are provided solely by the rotors, they must turn at very high speeds, which translates to a relatively loud signature.
Diehl Defence first publicly presented its concept for an LM in Autumn 2021. Since then, the concept has been further developed and, according to reports, a flight-ready system has already been demonstrated to interested customers, and will soon be ready for the market. For this to succeed, Diehl Defence engineers have combined already mature technologies and products in the creation of the Libelle.
According to Diehl Defence, Libelle will be available in two variants – a smaller version for dismounted use and a larger variant for mounting on vehicles. Both are reusable and designed for anti-tank applications. According to the manufacturer, the Libelle is absolutely fail-safe and designed to ensure easy operation even under combat stress. The total weight of the smaller dragonfly designed for dismounted use is said to be less than 13 kg. This means it can be easily stowed in a rucksack and carried by a single operator.
Strictly speaking, the Libelle is more of a classic munition shape with two coaxial contra-rotating two-blade propellers, rather than a more conventional copter design. The body of the Libelle consists of three main sections. The upper sections house the drive assembly with the electric motor, while the middle assembly directly below the lower propeller houses the electronics, power supply, and a fibre optic bobbin which can be used as a means of interfacing with the munition in the absence of radio. The lower assembly houses the warhead and the sensors, with the latter protruding out from the main body of the munition near the base. The sensors each appear to be provided with a cable conduit running up the side of the munition for power supply and interface purposes.
The Libelle’s operating principle can perhaps best be compared with the ‘SMArt 155’ family of munitions, which one or more sensor-fuzed submunitions. The warhead is strongly based on the explosively-formed penetrator (EFP) warhead used in SMArt, but some differences exist, such as in the fuzing system used. Whereas the SMArt relies on a combination of an infrared (IR) sensor and a millimetre-wave radar sensor to form the fuzing system, the Libelle uses cameras supported by modern image recognition algorithms. Beyond simply activating the warhead, these are also capable of discriminating between targets.
In contrast to SMArt, the Libelle is not fired from a gun, but makes its way to the target under its own power and can hover in place or fly search patterns until a worthwhile target is detected. Using intelligent image recognition, the target can then be tracked, and the munition can fly to within engagement range of the warhead. Once within range, the Libelle’s base is automatically aligned with the target to achieve maximum hit probability, and the EFP warhead is then activated. This provides sufficient penetration to defeat even main battle tanks if employed against their weaker roof armour.
According to Diehl, controlling the Libelle is as simple as it gets. For example, the LM can autonomously approach the potential target area following the operator simply setting a marker on a digital map, and it will then loiter there. Once a suitable target has been identified, it is selected by the operator, also by placing an on-screen marker on the target, and the rest of the engagement is automated. All in all, this means that the system can be deployed in a targeted and safe manner even under combat stress and without requiring extensive training.
The control software developed by Diehl is app-based, meaning that no specific control unit or ground control station is required for the Libelle. The user can install the app on their smart device, and use it to interface with and control the Libelle. Interface between the operator and the munition can be radio-based, adapted to the user’s radio systems, or via the fibre optic cable.
The fibre optic link has a length of several kilometres, and is immune to jamming, allowing the user to employ the Libelle even in on battlefields where the electromagnetic spectrum is contested. Moreover, in this interface mode the Libelle does not emit any radio or microwave signals, allowing it to avoid detection by hostile direction-finders. At a time when armed forces are increasingly investing in drone defence systems, this capability should not be ignored.
As soon as it is ready for the market, the Libelle certainly has the potential to complement the anti-tank capabilities of modern armed forces, finding its particular niche in urban or complex terrain where line-of-sight restrictions hinder the employment of anti-tank guided missiles (ATGMs). In these environments, the Libelle would offer an effective tool for both reconnaissance as well as strike, and thanks to the option of using a fibre optic interface, it can be equally effective under conditions of jamming by friendly or hostile forces.