ROBOTWAR.io
CapabilityB3likely · low confidenceRU · adversary

Russian V2U and Molniya drones use Jetson Nano for AI targeting

Bottom line

Russian V2U and Molniya drones are reportedly using NVIDIA Jetson Nano modules and YOLOv5 for autonomous target recognition.

Why it matters

Suggests Russian drones are moving towards greater autonomy in target acquisition, potentially reducing reliance on continuous human-in-the-loop control and increasing operational tempo.

Russian advisor Beskrestnov claims that V2U and Molniya drones are now equipped with Jetson Nano modules running YOLOv5 for autonomous target recognition of tanks and trains. The systems reportedly feature storage upgrades to 2-4TB to support larger neural networks. This indicates a shift towards edge-AI processing for autonomous targeting in Russian FPV and loitering munition systems.

Evidence — 1 independent origin

  • Беспилот (Bespilot — UAV, drones, robots)source Borigin t.mearchive
    Original text (excerpt)

    Россия опередила Украину в создании БПЛА с ИИ ▶️ Во всяком случае противник в лице советника украинского президента Сергея Бескрестнова жалуется, что российская армия уже обладает беспилотниками, управляемыми искусственным интеллектом, и в этой сфере ушла вперёд. Речь идет об авт

NamedMolniyaV2UJetson NanoYOLOv5Topicsdronesai warfarecomponents

Related findings

← Latest findings · All findings from September 2026

History · version 1 · first published · last substantive update

Cite this finding
Robot War (2026). Russian V2U and Molniya drones use Jetson Nano for AI targeting. Hetzner OSINT, 4 September 2026. https://robotwar.io/osint-hetzner/2026-09-04--russian-v2u-and-molniya-drones-use-jetson-nano-for-ai-targeting/

@misc{robotwar-2026-09-04--russian-v2u-and-molniya-drones-use-jetson-nano-for-ai-targeting,
  title = {Russian V2U and Molniya drones use Jetson Nano for AI targeting},
  author = {Robot War},
  howpublished = {\url{https://robotwar.io/osint-hetzner/2026-09-04--russian-v2u-and-molniya-drones-use-jetson-nano-for-ai-targeting/}},
  year = {2026},
  note = {Hetzner OSINT finding B3}
}

Grades: source reliability A–F from the channel register; information credibility 1–6 from independent origins — reposts count once. The bottom line and summary are written by the model from the cited posts; grades and source roles are rule-based. Methodology 2026-10-15.v6 · Corrections · JSON · RSS · API performance