IDDK Advances Toward 'Autonomous AI Experiments' in Space, Co-Files Two Patents for an Autonomous Experiment Platform Using LLMs

IDDK Inc., a provider of privately-led space experiment services, announced on June 10, 2026, that it has jointly filed two patents with Space Seed Holdings Inc., its subsidiary RegeneSome Inc., and Space Nhome Institute Inc. The patents concern a device and control technology that utilize machine learning models, including Large Language Models (LLMs), to autonomously conduct cell cultivation and various scientific experiments.

The two co-filed patents are for an "Integrated Autonomous Experimental Device for Cell Experiments and its Control Method" and an "Experiment Module Integrated Control Device and its Control Method."

The company views this filing as a crucial milestone, not just for automating cell culture devices based on its core one-chip microscopy technology, but for evolving into a platform that can autonomously execute and analyze space and terrestrial experiments in diverse scientific fields like microalgae, drug discovery, and materials science.

**Background: The 'Absence of Researchers' and Communication Barriers in Space Experiments**

In terrestrial research facilities, scientists can look through a microscope and adjust experimental conditions in real-time based on the state of the cells. However, in space environments like on-orbit or on the lunar surface, researchers cannot stand next to the equipment. Furthermore, space presents a high barrier with extremely limited resources (power, volume, communication) and unavoidable communication delays or interruptions with the ground.

As we enter the post-ISS (International Space Station) era and the lead in space experiments shifts to the private sector, there is a pressing need for a highly standardized, autonomous experiment platform that does not depend on astronaut work hours or communication environments.

**IDDK's One-Chip Microscopy (MID) Becomes the 'Eyes' for AI**

The company's goal is not merely to 'automate' experiments. It aims to integrate its core technology, the Micro Imaging Device (MID), into the experimental apparatus as the 'eyes' of the AI. This allows the 'observe, judge, operate' process, traditionally handled by humans, to be completed within the device itself.

This will enable the creation of an "autonomous space lab" that can autonomously detect changes in cells and continue optimal experiments even in extreme environments with interrupted communication.

**AI Transforming Scientific Research and the Aim of These Patents**

The world is seeing rapid progress in "Self-Driving Laboratories," which combine robotics, machine learning, and literature databases, allowing AI to autonomously cycle through literature review, experiment planning, device operation, results analysis, and determining the next conditions. Examples include the AI system "Coscientist" for chemistry research and the robotic lab "A-Lab" for autonomously searching for new material synthesis conditions, where AI is connected to experimental devices to drive research in a closed loop.

However, to extend this concept to the resource-limited environment of space, a mechanism is needed that can flexibly link and reconfigure multiple experimental instruments and observation modules with a common control layer, rather than individual devices that only automate specific experiments.

This joint filing targets two technological layers: the "autonomous control of cell experiments" incorporating the company's microscopy technology, and a "higher-level integrated control" that links different experimental devices with a common system.

**About the Co-filed Patent Technologies**

**Patent 1: Integrated Autonomous Experimental Device for Cell Experiments and its Control Method**

This is a space-compatible, autonomous bio-experiment device that integrates a culture vessel, reagent delivery mechanism, and environmental control (temperature, pH, gas, etc.) centered around the company's core MID technology. The key breakthrough is that an AI (like an LLM) integrates and analyzes high-precision microscopy images from the MID and other sensor data in real-time.

Two Innovations: - From mere 'automation' to 'autonomy' that makes its own judgments: Unlike conventional automated devices that follow a pre-determined program, the AI interprets morphological changes in cells observed by the MID (growth, abnormalities, etc.) on the spot to continuously and autonomously alter and optimize reagent dosage, timing, and temperature. - Autonomous response to space-specific troubles: In space, the behavior of liquids and bubble contamination can determine an experiment's success. When the system detects issues like bubble contamination from images, the AI autonomously executes recovery actions such as adjusting fluid speed, pausing, or purging based on the situation.

**Patent 2: Experiment Module Integrated Control Device and its Control Method**

This is a higher-level system platform technology that goes beyond cell experiments to connect and integrally control different types of experimental equipment (observation, analysis, liquid delivery, material synthesis, etc.) with a common interface. An AI (like an LLM) acts as the commander for the entire experiment by interpreting image and numerical data from all modules. This is a concept akin to an "Operating System (OS) for scientific experiments."

Two Innovations: - Adaptable to any experiment by simply swapping modules: As long as a new experimental module developed by a third party complies with the common protocol, the LLM will automatically recognize its functions and specifications. This allows for flexible adaptation from drug discovery to materials science without extensive reprogramming. - 'Edge-Cloud Coordination' that doesn't stop in space: The system uses high-performance AI models on the ground or in the cloud when communication is good, and automatically switches to a local AI model on the spacecraft when communication delays or interruptions are anticipated, ensuring experiments continue nonstop.

**Integrating the Expertise of Four Companies in Space Bio-Material Experiments**

This patent filing was made possible by the fusion of cutting-edge research from the four companies. The introduction of this technology aims to advance all these areas toward "autonomous AI experiments."

- **Microalgae Euglena Space Experiment (Space Nhome Institute x IDDK):** The MID technology will be installed on the 'Aoba' satellite, scheduled for launch in the latter half of 2026. This will evolve into microbial experiments that autonomously adjust light and temperature based on the organism's state. - **Cell/Exosome Research (RegeneSome x IDDK):** This will advance the joint development of devices for cell experiments and exosome production in space.