AI technology company Crowdworks said on April 14 it has joined a consortium led by the Ulsan National Institute of Science and Technology (UNIST) and was selected as a participating organisation for a task to develop a large-scale industrial AI system for the shipbuilding sector.

The project is part of the "Large-Scale Industrial AI Research Support Project" backed by the Ministry of Science and ICT and the Institute of Information & Communications Technology Planning & Evaluation (IITP). The UNIST consortium will serve as the lead organisation. Total funding is 40.3 billion won, including 28.5 billion won in central government funds, 2.5 billion won from Ulsan city and 9.3 billion won in corporate contributions, and it will run through 2030.

The goal is to develop a large-scale AI foundation model based on shipyard field data and apply and validate it in actual industrial sites to drive an AI transition in shipbuilding.

Crowdworks will work with UNIST, HD Hyundai Heavy Industries and HD Korea Shipbuilding & Offshore Engineering to take charge of building large-scale industry-specific data and developing training datasets. A core task is processing multimodal data, including document data such as design drawings and work orders, as well as on-site video and sensor data.

To that end, it will introduce its intelligent data preprocessing solution, the Alpy Knowledge Compiler. It plans to convert unstructured data such as complex shipyard design drawings into a form AI can use immediately. It also plans to build an operating system based on a closed network (on-premises) to manage key design data without external leaks.

A Crowdworks official said it was meaningful to join a major project aimed at driving AI innovation in shipbuilding, a national strategic industry. The official said the company will use its proven data engineering capabilities to create a successful case of large-scale AI optimised for the shipbuilding industry.