ROBOTWAR.io
CapabilityB3likely · moderate confidenceUA · friendly

PLA Navy tests GJ-21 drone on Type 076 amphibious ship

Bottom line

PLA Navy successfully tested the GJ-21 drone takeoff and landing on the amphibious assault ship Sichuan (Type 076).

Why it matters

Demonstrates China's rapid advancement in integrating UCAVs with amphibious platforms, potentially altering power projection capabilities in the Indo-Pacific and challenging US naval dominance.

Satellite imagery and reports confirm that the PLA Navy successfully conducted takeoff and landing tests of the GJ-21 unmanned combat aerial vehicle on the amphibious assault ship Sichuan (Type 076) on 30 October 2026. This marks a significant step in the integration of unmanned systems with China’s new generation of amphibious warfare vessels, enhancing their strike and reconnaissance capabilities.

Evidence — 1 independent origin

  • Telegram channel:Victory Dronessource Borigin t.mearchive
    Original text (excerpt)

    🔥 Китай провел успешные испытания по взлету и посадке на палубу десантного корабля «Сычуань» Тип 076 ВМС НОАК палубного дрона «Гунцзы-21». 🕔 30.10.2026 г. ✈️ @china3army МАХ ✍

NamedGJ-21SichuanTopicschinadrones

Related findings

← Latest findings · All findings from October 2026

History · version 1 · first published · last substantive update

Cite this finding
Robot War (2026). PLA Navy tests GJ-21 drone on Type 076 amphibious ship. Hetzner OSINT, 30 October 2026. https://robotwar.io/osint-hetzner/2026-10-30--pla-navy-tests-gj-21-drone-on-type-076-amphibious-ship/

@misc{robotwar-2026-10-30--pla-navy-tests-gj-21-drone-on-type-076-amphibious-ship,
  title = {PLA Navy tests GJ-21 drone on Type 076 amphibious ship},
  author = {Robot War},
  howpublished = {\url{https://robotwar.io/osint-hetzner/2026-10-30--pla-navy-tests-gj-21-drone-on-type-076-amphibious-ship/}},
  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 2026-10-18.v7 · Corrections · JSON · RSS · API performance