Collaboration with Spacelintec
Verifying Fault Diagnosis and Response Through Launch in the Second Half of the Year

Korea Aerospace Industries (KAI) has launched a technology demonstration project to verify the autonomous decision-making capabilities of artificial intelligence (AI) in space environments.

Business Agreement for Satellite AI Demonstration Cooperation. Center left: Hyunseok Seo, Executive Director of Satellite Research Lab at Korea Aerospace Industries, Center right: Haksun Yoon, CEO of Spacelintek. Provided by KAI.

Business Agreement for Satellite AI Demonstration Cooperation. Center left: Hyunseok Seo, Executive Director of Satellite Research Lab at Korea Aerospace Industries, Center right: Haksun Yoon, CEO of Spacelintek. Provided by KAI.

View original image

According to the aerospace industry on March 23, KAI signed a "business agreement (MOU) for CubeSat AI demonstration cooperation" with Spacelintec on March 20. The signing ceremony was held at the Daejeon Research Center and was attended by key officials from both companies.


The core of this agreement is to verify whether a high-performance AI module independently developed by KAI can be mounted on a CubeSat and autonomously diagnose and respond to satellite anomalies in orbit. The CubeSat platform is being jointly developed by Spacelintec and Yonsei University.


Based on this collaboration, the two companies plan to launch a CubeSat equipped with the AI module into space in the second half of this year as part of a demonstration project. The key focus of the project is to verify "AI on-board processing" technology, where the AI autonomously analyzes data, determines the satellite’s status, and derives response measures without instructions from the satellite’s main computer (OBC).


The demonstration will be carried out in stages. First, a specific fault signal is sent from the ground station to the satellite, and the AI module detects the signal, analyzes potential fault causes and scope, then derives optimal response measures and generates a technical report, which is sent back to the ground station.


This process, which simulates hypothetical fault scenarios to check the reliability of the AI algorithm, is considered a key step toward developing "fully autonomous satellites" capable of self-operation without ground intervention in the future. Currently, most satellites transmit data to the ground station when an anomaly occurs and receive analysis results and action instructions. If AI-based autonomous response becomes possible, it is expected not only to reduce communication costs but also to improve response speed through real-time decision-making.


KAI explained that, as a comprehensive satellite systems integrator, it developed the AI module by combining fault prediction and predictive maintenance technologies accumulated during aircraft development with its satellite expertise. The AI module was trained on various potential fault scenarios that could occur during satellite operation to enhance its autonomous decision-making capabilities.


In particular, the AI module incorporates a neural processing unit (NPU) developed by the domestic startup Mobillint. This is expected to contribute to the self-sufficiency of defense semiconductors as well.



Industry experts are paying close attention to the possibility that, if this demonstration succeeds, AI-based autonomous operation technology could become standardized across all satellite systems developed by KAI in the future. As the era of new space advances with the cluster operation of ultra-small satellites, securing autonomy is becoming a core element of competitiveness.


This content was produced with the assistance of AI translation services.

© The Asia Business Daily. All rights reserved. Unauthorized AI training and use prohibited.

Today’s Briefing