Skip to main content Skip to footer
RESEARCH PROJECTS & TEAM

Smart Manufacturing

Applies AI vision-language-action models and deep defect detection to automate quality control in smart manufacturing, improving yield and efficiency.

Smart Manufacturing

Related
Technology
Vision-Language-Action Models Deep Visual Defect Detection AMR Auto-Calibration Real-time Quality Control Optimization

Research Project

Demonstration-driven Process Knowledge Translation

This research focuses on demonstration-driven process knowledge translation for intelligent assembly tasks. The proposed system converts human operation demonstrations into robot-executable assembly instructions. By integrating visual perception, action understanding, atomic action parsing, and modular low-level skills, the system automatically identifies atomic actions, operation sequences, and key task parameters in assembly procedures, and translates them into executable robot control workflows, thereby reducing the need for manual programming and repeated tuning in conventional robot deployment.

Project Details

Principal Investigator

Ching-Hung Lee Distinguished Professor

Research Field: Artificial Intelligence Development and Applications, Design of Intelligent Systems, Control and Applications of Robot Manipulator, Intelligent Control, Smart Manufacture, Drone Applications

PI Details

Research Project

VR Multi-Robot Team Teleoperation

This subproject focuses on remote control, task collaboration, and wireless communication for multi-robot teams, establishing a VR operation system that supports collaborative work among heterogeneous robots. The system enables the operator to control different platforms through intuitive hand gestures in an immersive interface, while using a digital twin to understand each robot’s position, status, and task relationships in real time. The overall architecture emphasizes functional task allocation and communication capability, allowing the robot team to exchange control commands, image streams, and status information within the same network, while each platform retains its own computation and execution capabilities. The project has been validated on multiple robotic platforms and has completed load testing of compressed RGB-D streams from multiple robots, confirming that the system can support multi-robot data transmission. This system is expected to serve as a foundation for robot collaboration, allowing additional robots to be integrated into the same robot team at lower cost in the future.

Project Details

Principal Investigator

Ching-I Huang Assistant Professor

Research Field: robot-learning methodologies and virtual reality technologies aimed at enhancing human-robot cooperation

PI Details