Naver Cloud proposes AI framework to link mixed military robots
The concept uses spatial AI and data-linking to help drones and ground robots share a common operating picture and divide missions.

Naver Cloud presented a defense artificial intelligence and robot cooperation concept on Nov. 5 at the K-Defense Commercial Robot Technology Roadshow at KINTEX in Goyang, according to Edaily.
Choi Soo-hyuk, a Naver Cloud leader, proposed supporting drones and ground robots so they can understand the same battlefield situation and divide missions using AI, spatial information and data-linking technologies.
A remaining challenge is that robots from different platforms process information in different formats and meanings, so they can interpret the same situation differently, Choi said.
Naver Cloud presented three foundations for the concept: AI on individual equipment, a data base that links multiple systems, and AI infrastructure to support learning and improvement.
Field equipment needs lightweight AI that can recognize and judge its surroundings even when communications are cut off, because drones and tactical vehicles cannot always send information to the cloud and wait for answers, Choi said.
Naver is working on small models that can understand video, images, voice and documents together, according to Choi’s remarks reported by Edaily.
Choi said Naver’s spatial reconstruction technology could use photos or drone video to restore terrain and spaces in 3D, with potential uses in mountain, DMZ and underground-tunnel reconnaissance and battlefield-environment analysis.
The presentation included a scenario in which four drones share 3D spatial information, integrate their maps into common coordinates, divide search areas, hand off targets and coordinate routes, while engagement decisions require human approval.
Naver would use a data fabric to connect scattered data and a defense ontology to organize information relationships and military doctrine for commanders’ decision support, Choi said.





