Ukraine’s Robot Vehicle Completed 93% Alone, but Starlink Has No Quick Replacement
A Ukrainian unmanned ground vehicle completed 93% of a 20-kilometer test route autonomously in October 2026, a striking demonstration of how onboard computing can reduce the need for a continuous data link. Yet the test also frames the harder communications problem: Ukraine still has no near-term alternative that matches SpaceX’s Starlink across satellite coverage, connection speed and ease of deployment.
The Lava Unmanned Systems Regiment of Ukraine’s National Guard said an Apex AI-controlled vehicle handled 18.6 kilometers by itself, leaving an operator to control it for the remaining 1.4 kilometers. The system navigated with cameras, artificial intelligence and a three-dimensional terrain map created with lidar before the run.
That architecture moves part of the workload from the network into the vehicle. Instead of requiring an operator to issue continuous steering commands, the machine can identify obstacles and follow a mapped route locally. The communications link can then carry supervision and higher-level instructions rather than every movement.
Autonomy, however, reduces bandwidth demand rather than eliminating communications. Current systems cannot fully replace human operators during complex missions in rapidly changing environments. A pre-mapped ground route is also a narrower challenge than maintaining dependable connectivity across many mobile platforms, command functions and operating areas.
A satellite launch starts a much longer process
Ukraine is pursuing a sovereign space layer as well. Himera and Denmark’s GomSpace launched the OBRIA NANO experimental satellite on Oct. 1. By Oct. 27, the companies said it was operating in orbit and undergoing tests, according to SpaceNews. The project is intended to inform a future constellation providing communications coverage across Ukraine.
One test spacecraft is an engineering foundation, not a replacement network. Himera and GomSpace have not disclosed the satellite’s capacity, the number of spacecraft planned or the project’s financing. Building an independent satellite communications infrastructure would take several years at minimum, including the development of spacecraft, ground terminals, network management and supporting personnel.
That time scale helps explain Starlink’s enduring advantage. Its value is not confined to any individual satellite or terminal. The practical capability comes from a large constellation, broad coverage and inexpensive terminals that can be deployed with comparatively little specialized setup. Ukrainian military personnel interviewed by Oboronka said that ready access to those terminals has also slowed domestic communications development a classic consequence of relying on a capable commercial system that is faster to field than a sovereign alternative.
Terrestrial backups trade reach for infrastructure
Mesh networks and conventional radio links offer a more immediate backup. Systems including Silvus StreamCaster and HIMERA can pass data through interconnected devices, allowing traffic to take another path if an individual node is lost. This can maintain local communications without a constant satellite internet connection.
The constraint appears when that local network must span hundreds of kilometers. Greater distance requires more relay stations and supporting infrastructure, along with funding and specialized personnel. Those relays can also be disrupted, so mesh communications improve resilience without reproducing the geographic reach of a satellite constellation.
Commercial cellular service provides another supplementary path. Drones and other connected platforms can use 4G or LTE networks where service is available, but coverage gaps and device-access restrictions prevent cellular links from guaranteeing continuous connectivity. Cellular infrastructure is therefore useful as one layer, not a universal substitute.
The same boundary applies to more autonomous aircraft. Computer vision and terrain imagery can let some systems navigate or continue portions of a mission after losing a link. That capability limits the effect of an interruption, but it does not provide the broader command, coordination and data-transfer functions supplied by satellite communications. Replacing Starshield connectivity for high-speed drones remains especially difficult.
Ukraine’s most practical path is consequently an architecture rather than a single competing product: sovereign satellites for selected coverage, mesh and radio links for local networking, cellular service where available, and onboard autonomy to reduce how often vehicles need operator input. Major Ihor Bohdan has advocated this decentralized approach instead of searching for one direct Starlink replacement.
Each layer covers a different failure or coverage gap, but every added layer also brings integration, maintenance, training and procurement demands. OBRIA NANO can begin proving technology in orbit, while the 93% autonomous ground test shows how much communications demand can move onboard. Neither result closes the larger gap: a Ukrainian-controlled system with Starlink’s breadth remains a multi-year undertaking.
| More aerospace and engineering stories, right in your MSN feed. Follow AMI on MSN |
By David Whitaker — Associate editor for AMI’s aerospace and drone systems desk, translating flight systems, aircraft programs, spaceflight, and UAV developments into accessible technical stories.
