Losing Starlink Is the Best Thing That Happened to Ukraine's Drones

Losing Starlink Is the Best Thing That Happened to Ukraine's Drones

Everyone is panicking over the wrong antenna.

Read the breathless coverage whenever U.S. political friction threatens Starlink access, and you get the same tired script. Fragile supply chains. Geopolitical leverage. Blind drones falling out of the sky. It makes for great morning briefings. It is terrible engineering analysis.

I have spent years watching defense contractors and logistics teams blow millions building single-point-of-failure architectures because a commercial vendor made life easy in month one. Easy is expensive in month eighteen. When a battlefield system depends on a commercial satellite constellation controlled by a foreign boardroom and subject to diplomatic temperature swings, you do not have a resilient force. You have a rented tether.

A signal outage hurts. Nobody in a command post enjoys watching telemetry drop when a link flickers. Pretending a temporary loss of Starlink is a fatal blow to Ukrainian drone operations, however, confuses convenience with capability. Worse, it blinds observers to the brutal selection pressure that actually forces real autonomy out of cheap hardware.

The Lazy Consensus Around Satellite Tethers

The lazy consensus assumes modern long-range and tactical drones are helpless without continuous, high-bandwidth satellite command-and-control. If the pipe narrows or goes dark, the mission fails.

That assumption comes from peacetime doctrine and wealthy procurement habits. In an environment saturated with electronic warfare, directional jamming, and spectrum monitoring, relying on an external commercial RF link is already a compromise. Commercial LEO constellations give operators impressive throughput, low latency, and rapid deployment. They also paint an electronic footprint, introduce policy choke points, and encourage operators to treat drones like remote-controlled cameras rather than disciplined systems.

When ties fray and access wobbles, commentators treat the disruption like pulling the oxygen mask off a patient. In reality, many drone formations have been fighting with intermittent links, degraded bands, and spotty connectivity for years. Electronic warfare already breaks links daily without any help from Washington.

To understand what actually happens when Starlink drops, we have to separate command-and-control from navigation and terminal guidance.

  • Command-and-control is the operator talking to the airframe: steering, retargeting, adjusting speed, or pulling live video.
  • Navigation is knowing where the airframe is relative to the earth and the target: inertial measurement units (IMUs), terrain contour matching, optical odometry, or celestial/visual landmarks.
  • Terminal guidance is the final phase into the target: visual lock, onboard object recognition, or pre-programmed waypoints.

When analysts lump all three together under "lost signal," they sound authoritative while missing the mechanics. A drone that loses a satellite link does not automatically lose its mind. Only a drone designed like a flying webcam does.


Why Dependency Breeds Bad Engineering

I have seen defense startups burn through seed rounds chasing shiny satcom modules when their airframes could barely hold a straight line in crosswinds. Satcom feels like a shortcut. Need beyond-line-of-sight reach? Bolt on a terminal. Need live HD feed from deep behind the line? Route it through orbit.

Shortcuts carry hidden interest.

First, every extra gram, watt, and dollar spent on external comms eats into payload, range, or unit count. In attrition warfare, unit economics and mass matter relentlessly. A system that costs three times as much because of proprietary comms hardware is a system you field in smaller numbers.

Second, continuous operator-in-the-loop control creates a cognitive bottleneck. One pilot staring at one feed cannot scale effortlessly when hundreds of airframes go up simultaneously. Human attention is scarce. Bandwidth does not fix human fatigue; it often feeds it by tempting operators to micromanage flight paths that should be automated.

Third, geopolitical dependency is a design flaw disguised as a partnership. Any external dependency—whether a commercial satellite network, a single munitions line, or a foreign export license—gives an outside actor a throttle. Even if the vendor never intends malice, policy shifts, terms-of-service updates, billing disputes, or shifting diplomatic priorities can alter coverage maps overnight.

Classical military theorists like Carl von Clausewitz wrote extensively about friction—the gap between plans on paper and reality on the ground. Modern warfare adds policy friction right on top of physical and electronic friction. Ignoring policy friction during system design is malpractice.

Architecture Approach Comms Vulnerability Unit Cost & Scale Operational Bottleneck
Satcom-Dependent High to policy & jamming shifts Higher per unit; slower surge Operator attention & link latency
Autonomous / Edge-Guided Lower mid-flight; hardened Lower recurring comms cost Compute reliability & terrain data
Hybrid / Mesh / Fallback Moderate; graceful degradation Balanced; modular upgrades Integration complexity & training

Notice the middle row. Moving away from constant satellite dependence forces engineers to solve hard problems they otherwise put off: drift compensation in cheap IMUs, lightweight onboard vision models, resilient radio mesh relays, and pre-mission route planning that accounts for wind and jamming zones.

Those problems are ugly. They require messy field testing, failed prototypes, and humbling data from real flights. Yet once solved, they produce systems that do not blink every time a political headline turns sour.


Answering the Questions People Keep Asking the Wrong Way

Whenever Starlink disruptions hit the news, the same search queries spike. Most of them bake a false premise right into the question.

"Can Ukraine's drones still operate effectively without Starlink?"

Yes—for specific missions and well-engineered platforms—and no for lazy designs that were never built to survive contested or severed links.

Ask that question as a binary yes-or-no and you get useless answers. Long-range strike drones flying pre-planned profiles toward static infrastructure do not need live satellite video every second to reach a coordinate. Short-range FPVs rely on local RF links or fiber-optic spools rather than LEO satellites anyway. Where Starlink hurts most is mid-range reconnaissance, dynamic retargeting, and flexible command relays where operators want real-time two-way data over long distances.

Even there, "less convenient" is not "useless." Teams adapt by shifting to staged relays, tightening mission profiles, using stored imagery, and accepting delayed reporting rather than live streams.

"How do drones navigate when GPS and satellite signals are jammed or lost?"

By combining boring math with stubborn sensors, rather than praying for a clean sky signal.

Imagine a scenario where a reconnaissance drone crosses a heavily jammed corridor. Satellite positioning is spoofed or drowned out, and the satellite data link drops completely. Inside the airframe, a micro-electromechanical IMU tracks acceleration and rotation thousands of times per second. Left alone, small sensor errors compound over time—a phenomenon known as inertial drift.

To keep drift from ruining the run, the system pairs inertial data with other cues: barometric altitude, airspeed, magnetic heading where reliable, or downward-facing cameras matching ground features against stored maps. Even simple terrain-following rules and dead reckoning get a well-tuned airframe surprisingly far. It will not match surgical precision over a hundred miles with zero updates, yet it often gets close enough to reacquire a target zone or complete a leg.

Only as an interim bridge, not as a permanent religion.

Swapping one commercial satellite provider for another—whether Eutelsat/OneWeb, regional geostationary capacity, or a bespoke military constellation—solves the immediate coverage gap. That matters tactically. Do not dismiss tactical relief when troops are in the field.

Still, treating another satellite logo as the final answer just moves the leash to a different post. Commercial constellations are built primarily for commercial economics and commercial orbits. Military needs—rapid tasking, sovereign control, anti-jam resilience, surge capacity—frequently clash with commercial business models.

The smarter play is graceful degradation. Build systems that use commercial satcom when it is available, fast, and cheap, while remaining fully capable of completing core missions when that link vanishes. Treat orbit as a turbocharger, not the engine block.


What Happens Next in the Trenches

Let us be honest about the cost of this contrarian view. If you strip away easy satellite links and push harder on edge autonomy and local relays, things break.

Drones will miss targets more often at first. Onboard computer vision gets fooled by fog, shadows, seasonal foliage changes, and decoys. Edge compute draws power and generates heat in tight fuselages. Training operators to trust autonomous modes without abandoning manual oversight takes time—and mistakes cost hardware.

Anyone claiming that losing Starlink carries zero downside is selling fantasy. You lose bandwidth. You lose rapid mid-flight course corrections in some scenarios. You force operators and engineers through a painful adjustment period.

Yet pain is part of the curriculum in modern warfare. Electronic warfare pioneers like those who shaped modern signals intelligence and suppression of enemy air defenses learned decades ago that every frequency you rely on becomes a target or a liability. The same rule governs external networks. If an adversary or a third party can sever your link, you have handed away part of your sovereignty.

Forces that treat a signal loss as a crisis stay reactive. Forces that treat a signal loss as a design constraint build tougher machines, sharper doctrine, and supply chains that cannot be held hostage by a single contract.

Stop mourning the tether. Start building drones that do not need one.

HH

Hana Hernandez

With a background in both technology and communication, Hana Hernandez excels at explaining complex digital trends to everyday readers.