Like in previous years, this year has shown the potential of unmanned systems on the battlefield. The war in Ukraine has also shown us that there is still a lot of potential to be uncovered in unmanned warfare. With unmanned warfare, I am not only referring to the continuing weaponisation of various types of commercial airborne drones or the development of unmanned ground platforms. The term unmanned warfare also includes the systems developed to protect against multiple systems in the air, on the ground, or at sea, as well as the growing industry involved in delivering solutions with short lead times. It is an industry mainly centred around small groups of tech-savvy individuals who have transitioned from modifying existing platforms and building with parts bought online to designing and constructing electronic parts and circuits. We should not forget that software developers are continuously upgrading the firmware to various commercial platforms like the DJI Mavic or building their machine-vision targeting software for FPV. With this year drawing to an end, one can conclude that the war has shown not only the importance of engaged volunteers to supply the armed forces with weaponry but also that the cycle of adaptation to the present situation on the battlefield needs to be short (months) where a fast, simple solution that is “good enough” has precedence over an advanced optimal solution that takes longer time to produce.
But what can we expect for unmanned warfare in 2025? I have previously covered the evolution of unmanned systems in Ukraine.[1] As such, I don’t intend to repeat what has already happened but focus more on what we might see in the coming months. However, to do so, we need to highlight some recent developments in Ukraine that have laid the foundation for what will come, both technical and tactical.
Let us begin with the apparent development of artificial intelligence (AI) augmentation in unmanned platforms – or, to be more accurate, automatic tracking and interpreting sensor data. AI has been a buzzword for several years, and the war in Ukraine is no exception. In some regards, it has also become a weapon in information warfare. Claim having drones with AI targeting capabilities, and the enemy must act accordingly. However, when viewing the available video material, one would deduce that it is not an AI as a problem-solving brain adapting to its environment (as defined by the EU[2]) that is commonly used but rather a machine strictly following its programming.[3] Although there is some processing of collected sensory data, it mostly resembles standard tracking functionality in larger military-grade systems. As such, it is not autonomous either but rather an automatic process.[4] That does not downplay the significance of the targeting functions in the various platforms. Even the simplest tracking algorithms increase the probability of the drone (or other platforms) striking its target even while jammed. Using easily accessible micro-computers like Raspberry Pi as targeting computers has made previously expensive technology notably cheaper and more accessible.[5] Platforms with automatic targeting systems will likely become a common sight next year, both on the ground and in the air.[6]
Another technical development likely to become more common next year is platforms that can switch frequencies automatically or by the operator when suppressed by electronic warfare. There are already indications that Ukraine has operational systems where a drop in signal strength triggers a change in frequency without losing video or telemetry. This will require the ground control station to have multiple modules and antennas instead of the standard of one dedicated antenna. However, one Ukrainian solution utilising commercially available transmitter modules has been showcased on social media. Using commercially available modules from well-established manufacturers guarantees compatibility with most first-person view drones today. It presents a cheap and quick way of dodging specific counter-uas (CUAS) applications focusing on one frequency. Most handheld CUAS systems fielded by the Russian side can only suppress one frequency at a time, and having to overpower two frequencies requires more energy than these systems may be able to generate without an added power source. It also adds complexity when a drone is carrying a repeater, which, in theory, would mean that the CUAS operator has four different frequencies to suppress in different directions.
Unmanned ground vehicles (UGVs) have become more common throughout the year. They usually serve three distinct roles: as a mine layer, logistical transport, or vehicle-borne IED (VBIED). Especially from the Russian side, there seems to be no shortage of companies ready to produce various types of UGVs for the Russian armed forces. Although the more prominent platforms like Uran-9 have yet to be fielded, the Russian units seem eager to use whatever platforms provided, even the crude ones consisting of converted hoverboards.[7] Ukraine has also caught on to the development and bought buggy-like vehicles to supply their soldiers and evacuate the wounded.[8] Both parties have utilised UGVs as remotely controlled weapon platforms with varied success[9], and it is almost certain that we will continue to see small- to medium-sized UGVs used in various support roles next year as a means to minimise losses and hold terrain. We might see dual-use platforms, such as combined weapon and supply vehicles. Some ground vehicles will probably use automatic driving functions and the ability to detect and engage targets (much the same way as the airborne platforms). We might also see automatic teaming between platforms (air and ground) with a master/slave relationship during 2025.
Although a first-person view (FPV) drone is simple in its construction, it has shown its worth on the battlefield as both an anti-armour and anti-personnel weapon. However, success cannot be attributed only to the technical platform. We must acknowledge the tactical development required to gain the upper hand in an ever-changing environment. The latest revealed tactic is the “pop-up ambush”, where an FPV drone is put on standby in the field, waiting for a target to strike.[10] Although it is uncertain how effective this tactic is, it is at least a way of prolonging the flight time of FPV drones, which commonly have a very short flight time (30 minutes or less) compared to other platforms. Instead of searching for a target and spending precious flight time, one could use a recon drone with a longer flight time, which also can function as a repeater. Combine this with swarm technology or teaming between air and ground platforms, and you will have a combined weapons ambush conducted without the risk of losing soldiers. Tactical adaptations like this are vital when hit rates decline to around 30-50%.[11] We will probably see the ambush tactic combined with machine-vision targeting during next year, where an FPV can be prepositioned and put on standby to become active when certain conditions are met.
A cheap and relatively common tactic against FPV is using analogue video receivers bought online to detect the video feed from drones nearby. Both Russian and Ukrainian soldiers have been documented using this method throughout the war, especially to verify the effectiveness of electronic warfare engagements. One broadly available solution to the problem is using a digital video link instead of an analogue. There are several available “plug-n-play” solutions that FPV hobbyist use to convert their analogue equipment to digital. Getting the parts needed is no more difficult than acquiring analogue receivers and transmitters. Digital video means higher resolution and a clear picture, especially in infrared (IR), which has several benefits in a military setting. And the use of digital links renders analogue video receivers useless. However, while digital may be preferred in a civilian setting, it has some drawbacks when used as military equipment. The most prominent is that a digital video feed that encounters distortion or starts to get a weak signal between the platform and the operator will sometimes become pixelated, distorted and even frozen. It’s just like streaming your favourite movie on a lousy internet connection. An analogue video feed will become static, but the feed will continue without losing video segments.
The more expensive solution, which Russian units have been said to be using successfully between 6 and 10 km behind the front, is the FPV with a fibreoptic cable spool. Although the length of the cable limits the range and has some impact on the speed, the fibreoptic FPV has proven to be agile enough to pose a threat against soldiers and vehicles. Ukraine has picked up the thread and started to develop similar platforms. The benefit of fibreoptic cable is that it can’t be suppressed or detected with electronic warfare. And just like the previously mentioned digital video link, the operator receives a high-resolution image. This is one possible reason Russian drones first do a flyby is to gather imagery of the target area before striking. A strike that is by far more expensive than the FPV platforms we are used to. An FPV with a fibreoptic spool containing a 10km cable can cost between $1500 (Ukrainian version) and $2500 (Russian version) per unit.[12] An ordinary FPV used in Ukraine costs between $200 and $500 per unit. Although fibreoptic protects against electronic warfare, it seems less likely to become a familiar sight than FPV with digital video links. Both Ukrainian and Russian drone units rely on monetary donations to be able to buy parts and build drones. Some units will be hard-pressed to choose their solution, especially when consuming 3000 drones monthly.[13]
On that note, the energy battle is something to observe next year. A Russian trend throughout 2024 has been a shortage of batteries for DJI Mavic. This has resulted in Russian units requesting donations for batteries, buying cheaper battery clones, modifying existing batteries, and even trying to get more capacity out of the battery through software and hardware modifications. Meanwhile, Ukrainian units are said to have experienced that electronic warfare equipment in the trenches runs out of power after hours of continuous combat and is hindered from recharging batteries when subjected to artillery fire. Within the war is a war of energy endurance; whoever can keep their equipment online (or airborne) the most prolonged gains the upper hand. The previously mentioned tactic with pre-positioned drones is one way to save energy and keep an asset active longer. Solutions for maintaining equipment dormant, especially jammers, have been showcased recently, and we likely will see more solutions being battle-tested in the coming months.
These are some highlights of what to look for next year. One could describe several more areas where we are likely to see changes. For example, I have not mentioned unmanned surface vessels (USV), the potential reorganisation of forces in Russia or the potential of swarms. I will almost certainly get around to covering these areas during 2025. However, these picks represent areas that this author thinks will impact the outcome of combat engagements in the upcoming year.
