Research & Development

The ultimate goal of this project is to automate the extraction of fixture coordinates, generation of machining-area coordinates, and pose alignment for semi-structured workpieces using AI-based 3D scanner analysis, thereby enabling fully unmanned operation across the entire Loading–Scanning–Cutting–Unloading process.

To achieve this, an AI segmentation model will be developed to automatically distinguish the product region from the material-removal region. In addition, scan-to-CAD registration software will be developed using partial point cloud data and full CAD models to automate pose recognition and coordinate extraction.

The key quantitative targets are a coordinate and pose calculation time of less than 2 seconds, AI-based pose alignment accuracy within 200 μm using only the main system, and an AI segmentation accuracy of at least 92% IoU. After applying the ICP-based post-processing registration algorithm developed by the consortium, the final target is a positional error within 100 μm and an angular error within 0.2°.

This development aims to address the increasing heat flux of inverter power modules driven by the wider adoption of SiC power semiconductors and the expansion of 800V high-voltage electric vehicle platforms. Conventional CNC-machined inverter coolers have structural limitations because their uniformly arranged pin-fin structures cannot effectively respond to non-uniform heat generation, while multi-part assemblies using bolted joints and gasket sealing inherently involve leakage risks. The target of this development is the coolant flow-path section of a cold plate for a 150 kW-class passenger EV inverter, excluding the power electronics circuit and power module package from the development scope. The external housing and internal heat exchange structure will be manufactured as a single integrated component using the L-PBF metal additive manufacturing process. By applying a DfAM-based variable-density flow-channel design, the development aims to respond effectively to localized heat concentration and achieve quantitative targets of thermal resistance of 0.136 K/W or less, pressure drop of 20.0 kPa or less, and temperature uniformity ΔT of 8.0°C or less, thereby mitigating thermal bottlenecks in high-power SiC inverters and securing next-generation additive-manufactured inverter cooling component technology.

Development of an on-premise Edge AI-based predictive maintenance system for CNC machines that analyzes real-time equipment data to reduce unexpected failures, minimize downtime, and improve manufacturing productivity.

This project aims to develop a variable-transparency window technology for mobility sun visors featuring high-speed optical response and low haze characteristics. The technology is designed to dynamically control light transmission while maintaining high visibility and optical clarity, thereby improving passenger comfort, safety, and energy efficiency in next-generation mobility applications.

The objective of this project is to improve workplace safety and enhance productivity by introducing robotic automation into the rotor pin press-fitting process for automotive HVAC components used in electric vehicles. Through the automation of repetitive and labor-intensive operations, the project aims to increase process consistency, reduce manual workload, and improve overall manufacturing efficiency.

This project aims to develop specialized 3D printing technologies for key motor materials, enabling the production of free-form motors that cannot be manufactured using conventional methods. By reducing production costs and improving motor performance, the project seeks to support high-mix, low-volume manufacturing of motors and related components for a wide range of advanced industries.

This project aims to develop a metal additive manufacturing factory solution that leverages artificial intelligence (AI) to optimize metal additive manufacturing processes and apply them to the production of high-carbon steel tooling components. The project utilizes energy image generation and AI-based analysis technologies derived from unstructured process data to monitor and optimize the build process in real time. In addition, it seeks to develop metal powder traceability and management technologies to improve build quality, enhance productivity, and reduce labor requirements in additive manufacturing operations.

CY Autotech was selected for the 2024 AI Voucher Support Program, a government initiative designed to accelerate digital transformation and enhance industrial competitiveness through the adoption of artificial intelligence technologies.

The objective of this project is to contribute to national defense capabilities by manufacturing defense components using 3D printing technology. Through this initiative, the project seeks to demonstrate and validate domestically produced metal additive manufacturing parts, promote localization of critical defense components, and strengthen cost competitiveness in the defense industry.

In 2024, CY Autotech was selected for the Automotive Parts Supplier Competitiveness Enhancement Support Program. As part of Hyundai Motor and Kia's supplier support initiative, the project is expected to strengthen the company's security infrastructure and proactively address future cybersecurity challenges.

This technology focuses on the development of an ultra-lightweight, ultra-thin film cell with variable transparency in the viewing area to improve the visibility of augmented reality (AR) devices. By dynamically controlling light transmission, the technology enhances image clarity and user visibility, making it suitable for applications in AR glasses, automotive glazing, and architectural glass systems.

This project aims to develop the industrial design of a professional-grade 3D printer through close collaboration with a specialized design firm. The goal is to enhance product competitiveness, usability, and brand identity while reflecting the advanced technological capabilities of the system.

This project seeks to lay the groundwork for a successful business by investigating market demand and technological trends related to smart factory solutions for powder management in mass-production metal additive manufacturing. The study will support the development of commercialization strategies and future growth opportunities in the additive manufacturing industry.

The objective of this project is to build datasets for shape correction of composite resin 3D-printed parts and to establish data combinations that correlate material, equipment, and process parameters with mechanical properties. These datasets will be used to predict and compensate for the quality of products manufactured through the Powder Bed Fusion (PBF) metal 3D printing process.

Development of an Automated Post-Processing System for Unmanned Machining with Workpiece Coordinate Recognition Accuracy Within 50 Micrometers

Development of Process Inspection Software for Metal Additive Manufacturing Using a 3D Metal Printer

This feasibility study evaluates the need for developing a PBF-based metal additive manufacturing system equipped with in-situ process monitoring software and an interchangeable cartridge module. The project aims to enhance productivity, process reliability, and scalability, thereby enabling the commercialization and mass production of metal 3D printed components.