🔬 Active NSTC Projects
Development of a High-Performance Autonomous Navigation System for Unmanned Vehicles Integrating Fiber Optic Gyroscope Inertial Navigation and Multi-Modal AI Visual Perception
Project Overview:
Inspired by human autonomous motion decision-making, this project develops an AI navigation system applicable to sea, land, and air autonomous vehicles. By emulating the spatial awareness and motion decision-making processes of human drivers, the system achieves deeper environmental understanding and cognitive capability. The project aims to make substantive innovations in autonomous vehicle research while advancing Taiwan's All-domain Unmanned Vehicles industry.
Inspired by human autonomous motion decision-making, this project develops an AI navigation system applicable to sea, land, and air autonomous vehicles. By emulating the spatial awareness and motion decision-making processes of human drivers, the system achieves deeper environmental understanding and cognitive capability. The project aims to make substantive innovations in autonomous vehicle research while advancing Taiwan's All-domain Unmanned Vehicles industry.
Digital Twin Cross-Platform Collaborative R&D for Intelligent Automation of Container Terminal Vehicles
Project Overview:
A joint initiative between National Taiwan University and Chenxi Precision has established the "Building and Port Automated Vehicle R&D Sub-Center," focusing on the automation of five categories of industrial vehicles. Built on a digital twin foundation, the project covers drive-by-wire retrofitting, remote control, human-machine switching, multi-modal perception, and cross-platform dispatch. The project also collaborates internationally with the University of British Columbia (UBC) in Canada.
A joint initiative between National Taiwan University and Chenxi Precision has established the "Building and Port Automated Vehicle R&D Sub-Center," focusing on the automation of five categories of industrial vehicles. Built on a digital twin foundation, the project covers drive-by-wire retrofitting, remote control, human-machine switching, multi-modal perception, and cross-platform dispatch. The project also collaborates internationally with the University of British Columbia (UBC) in Canada.
📋 Other Active Projects
Hardware-in-the-Loop (HIL) Research for UAV Hybrid Propulsion Systems
Project Overview:
This project enhances the design and testing capability for UAV hybrid propulsion systems. The focus is on establishing system architecture and mathematical models, and building a Hardware-in-the-Loop (HIL) testing workflow to validate energy management and control strategies. Methods include system analysis, HIL platform construction, and controller interface integration.
This project enhances the design and testing capability for UAV hybrid propulsion systems. The focus is on establishing system architecture and mathematical models, and building a Hardware-in-the-Loop (HIL) testing workflow to validate energy management and control strategies. Methods include system analysis, HIL platform construction, and controller interface integration.
🤝 Collaborative Projects
Heterogeneous Unmanned Vehicle Swarm Semantic Communication System
Project Overview:
Developing an intelligent command brain with deep environmental understanding and decision-making capability, enabling a single operator to intuitively command complex vehicle swarms. Through the SAVLink semantic communication protocol, natural speech commands are precisely translated into logical instructions, maximizing bandwidth efficiency and survivability in contested electromagnetic environments.
Developing an intelligent command brain with deep environmental understanding and decision-making capability, enabling a single operator to intuitively command complex vehicle swarms. Through the SAVLink semantic communication protocol, natural speech commands are precisely translated into logical instructions, maximizing bandwidth efficiency and survivability in contested electromagnetic environments.
🌍 International Collaboration
Physics-Informed Neural Network-Based Material Cyclic Hardening Prediction and Assessment
Project Overview:
In international collaboration with Tohoku University (Japan), this project develops a Physics-Informed Neural Network (PINN)-based prediction system. By extracting monotonic tensile test data, it inversely derives complex cyclic hardening parameters while embedding mechanics governing equations to ensure physical consistency. Monte Carlo simulation is combined with finite element analysis to quantify the impact of material-level uncertainty on macroscopic structural behavior.
In international collaboration with Tohoku University (Japan), this project develops a Physics-Informed Neural Network (PINN)-based prediction system. By extracting monotonic tensile test data, it inversely derives complex cyclic hardening parameters while embedding mechanics governing equations to ensure physical consistency. Monte Carlo simulation is combined with finite element analysis to quantify the impact of material-level uncertainty on macroscopic structural behavior.