Missiles for the Rest of Us? Deep Strike Drones and Countermeasures Unpacked!

What are long-range or deep strike drones? Do they serve a purpose in the force composition of EU and NATO nations? Are they a temporary development or here to stay? Who should take a closer look, and which systems make sense for which mission profiles? What can be done about them? Where is the market headed? In this series, we’ll explore these and more questions for the most common systems. We’re starting with the most popular system, the Shahed in this article the size of a practical paper.

>>>This is an excerpt from my new >10.000 word strategic deep dive on TECH WARS. See the link at the end of this article to check it out!<<<

https://techwars.substack.com/

Electronic Warfare vs. EW-resistant Navigation & Guidance

Electronic warfare (EW) countermeasures like jamming and spoofing are widely being marketed and deployed. They exist in the shape of stationary, mobile and handheld systems, and are used to either overwrite a signal (e.g. giving a drone the wrong coordinates) or to disturb its reception with by introducing noise on the same frequencies where the signal is being received, effectively decreasing the signal-to-noise ratio. The promise of EW countermeasures is that they can make a drone impossible to pilot, either by interrupting or tricking its positioning system, or by interrupting or tricking its communications. Many legacy and lower-priced commercial quadcopters react to these kinds of disturbances by hovering in place, returning to a pre-defined “home” location, or crashing. As we will find out in the rest of this section, modern deep strike drones don’t.

Electronic warfare countermeasures suffer from a painful high-speed arms race/co-evolution dilemma. Just as with the emergence of fiberglass-connected FPVs, which now fly tens of kilometers without interference, deep strike drones have been upgraded incredibly often and rapidly over the past years. For communication with the operator, the use of local (Ukrainian, or, as seen recently, Polish) SIM-cards and Starlink Mini terminals have proven successful. For Global Navigation Satellite System (GNSS; GPS, BeiDou, GLONASS, etc.) connections, the use of multi-band antennae like the already infamous Russian Kometa help counter the countermeasure. Inertial navigation systems (INS), which navigate by using an accelerometer and a gyroscope to calculate movement through space, work to a degree, though they suffer from what’s called “drift”, due to their lack of reference points and inherent inaccuracies of the system.

Visual Navigation and Guidance

The effectiveness of visual navigation like terrain-matching, another countermeasure against spoofing and jamming, is dependent on multiple factors, like the quality and accuracy of the map and the quality of the visual data the drone uses to match its position with its maps. Though advances in edge computing have helped improve the accuracy of terrain-matching, the system also suffers from inherent limitations like sensitivity to lighting and weather conditions, and is inherently limited in flight altitude if there is cloud cover. Some of these can be addressed by using more expensive sensors, but this counteracts the cost-effectiveness at the core of most of these systems.

Computer vision has been more successful in terminal guidance, for instance as observed in Shahed-238 equipped with NVidia Jetson AI computers. The last-mile environments these drones operate in are usually the most EW-contested ones. This is one reason for the many unsuccessful strike attempts and lost drones (with the strike success ratio standing at about 8:2 in August 2025). AI-powered terminal guidance addresses this problem by using machine learning to train a computer vision model to algorithmically identify the shapes and colors of the desired target objects. Once the drone arrives within the kill-box, meaning the pre-designated three-dimensional area where it is authorized to seek and destroy, it locks onto a target and engages it. An important side-note needs to be made: while this is automation, it does not deserve the marketing term autonomy. While these software programs can be quite nuanced, it is important to mention that 99.9% of what is currently being sold at a mark-up because of “AI” is actually nothing but a combination of pre-programmed flight paths and maneuvers, combined with commoditized computer vision algorithms for things like target acquisition. None of this is new, and all of it will be de-bunked in a future deep dive.

How Deep Strike Drones Survive Heavy EW

Let’s go through an example operation. A drone is controlled by an operator, for instance via multiple SIM cards or Starlink, until the connection is interrupted, and take over via visual or inertial navigation if it has not yet reached its kill-box. In this example, the connection is re-established and the operator regains control of the vehicle, basically overriding the internal control functionality, only to see it interrupted again. The drone then takes over control once more, and, in this more sophisticated example of a drone, scans the area for targets using its visual and infrared (IR) sensors while performing a circular movement akin to an airplane’s holding pattern, called a loitering maneuver (this is what OWA UAV get their other name from: loitering munition). Once it has completed its scan, it automatically attacks the softest point...

I write comprehensive strategic deep dives (this one contains more than 10.000 words, and is illustrated with tens of images & short videos) that help decision-makers stay on top of trends, sort the information from the noise, and make better procurement and deal decisions - equipping you with the same level and depth of information of professional analysts and experts, but making it digestible. These deep dives are full of primary research, intelligence and experience I gather myself, from the offices of OEMs and startups to the forward edge of innovation in Ukraine.

They’re helpful for MoDs, OEMs and SMEs, and their investors in PE, VC and banking.

Want to know how the story continues? Head on over to TECH WARS on Substack to read the full deep-dive post. The post is free to read for a limited time - what are you waiting for? GO!

It's truly comprehensive! 👇