Field report S-83 / M2 · sheet /strategy

Corporate StrategyFiled from the territory

No Rules of the Road: Why War Robots Out-Engineer Self-Driving Cars

War robots face a problem self-driving cars never do: no roads, no rules, no stable terrain. Ukraine's machines stay mostly remote-controlled — for now. By Elena Vasquez.

Grid
S-83 / M2
Surveyed
November 15, 2026
Reading
4 min read
Correspondent
Elena Vasquez
The race to build autonomous war robots faces a problem that self-driving cars don't. There are no rules of the road.
Fig. S-83 / M2The race to build autonomous war robots faces a problem that self-driving cars don't. There are no rules of the road. — jarmoluk / pixabay

Field notes

  1. Most ground robots in Ukraine remain remote-controlled, with autonomy limited to niche functions such as waypoint following, ARX Robotics executive says.
  2. ARX CEO predicts battlefield management systems directing vehicles within one to two years; no company has deployed it yet.
  3. ARX's GEREON robots are used by Ukraine and procured by the UK; DCE's X-series robots are procured by the British Army.
  4. Ukraine's Brave1 grant program seeks autonomous logistics robots that can return to base and operate when communications are lost.

Battlefield robots still cannot drive themselves into combat, and the reason is structural: unlike civilian roads, war has no lanes, no markings, no rules and no guarantee that tomorrow's terrain resembles today's.

Militaries are fielding ground robots in fast-growing numbers, above all in Ukraine, where uncrewed vehicles fire weapons, attack positions and evacuate wounded soldiers while their operators stand farther from danger. Defense manufacturers now want to push those machines toward greater autonomy — and are running into a problem self-driving carmakers do not face.

The commercial logic is straightforward. Greater autonomy would let fewer soldiers operate more vehicles, allow troops to stay farther from the front, and keep robots moving when jamming cuts communications.

What makes a battlefield harder than a street?

David Roberts, CEO of the British arm of German defense company ARX Robotics, whose GEREON robots are used by Ukraine and procured by the UK, said civilian developers work in an environment that is fundamentally easier.

"Roads are structured," Roberts said. "They've got rules. Roads are wide enough for cars to drive down. They have markings to tell a vehicle what to do and when to do it."

In war, "there are no real rules," he continued. "You don't quite know if someone's going to step out in front of you or be in the long grass in front. It's a very challenging environment in the first place to build autonomy."

Mike McGinty, CEO of British firm Digital Concepts Engineering (DCE), whose X-series ground robots the British Army has procured, framed the contrast in similar terms. A street is "flat, of a guaranteed surface. It's marked. It's a track. It has a left and a right. It's signed," he said — and it persists: "The same street's there tomorrow."

Battlefields offer none of that stability. Robots must cross muddy tracks, forests and dangerous terrain while under attack, with terrain, conditions and targets changing constantly.

"True autonomy is an extremely hard problem on cross-country ground," McGinty said. "If you see a line on the ground, is it a hole, a log, a burn mark, a shadow, a crocodile, a person?"

He acknowledged that building a self-driving car or taxi remains hard — "the truth is it's still extremely difficult" — but said autonomous military ground vehicles are a far more daunting task.

How much autonomy do Ukraine's robots actually have?

Not much, by the executives' own account. Roberts said most robots in Ukraine are simply remote-controlled, with autonomy confined to "very niche areas" such as following waypoints when the operator connection drops.

ARX's GEREON robots run on Mithra, a software layer connecting sensors, AI and autonomy functions. The company says its "supervised autonomy allows a single soldier to task and oversee multiple systems simultaneously, delivering more mission output with fewer personnel."

The machines can already follow waypoints, detect people, avoid collisions and make safety stops, Roberts said. The "ability to tell a vehicle to go to a certain place on the battlefield is very much where we are working at the moment." He predicted that within the next year or two, battlefield management systems will be able to direct vehicles where to go — a capability he believes no company has deployed yet, though several are working on it.

McGinty said remote operation is the autonomy level DCE has "mastered." He was blunt about the next step: "Full autonomy is still a very difficult technological step. So someone will take it, but it won't be us because we're not the software company doing that."

DCE's robots can be fitted with a Marionette kit enabling remote control and supporting autonomous features, including software built by other companies.

What comes next?

Oleksandr Yabchanka, head of robotic systems for Ukraine's Da Vinci Wolves Battalion, said ground robot autonomy in Ukraine "is still in the early stages of development," but demand is real: "The future of both aerial drones and unmanned ground systems lies in the development of autonomy."

Multiple programs are underway. Ukraine's state-backed innovation platform Brave1 runs a grant program calling for "autonomous logistics" robots that can independently return to base and keep moving when communications are lost.

Ukraine has already integrated AI into aerial drones — the war's more prominent uncrewed system — as well as fixed weapons, and is increasingly embedding it in ground robots. The aim across these programs: one soldier overseeing many machines at once.

The battlefield, not the highway, will decide who solves it first.

via feeds.businessinsider.com (Source)

Share this article:

More from Elena Vasquez

Elena Vasquez

Show full bio

Staff writer covering companies, deals and strategy in corporate strategy and chief-executive leadership at CEO Frontier.

486 articles

Filed under

Adjacent sheets

« Previous articleNext article »

End of sheetS-83 / M2 · November 15, 2026