基於擴增實境之機械手臂運動規劃技術

本計畫開發一套基於擴增實境（AR）的創新機械手臂運動規劃技術，以提升製造現場的人機協作能力。隨著人工智慧和工業4.0的推進，智慧製造逐漸成為產業重點，但現場人工作業仍難以完全被自動化取代。傳統的機械手臂操作通常需要專業知識和複雜的編程，這對中小型企業在技術應用上的門檻較高。因此，本計畫利用AR技術結合機械手臂控制系統，來實現直觀的操作模式，減少工程人員的學習曲線，並提升操作效率和準確性。運用AR技術作為人與機械手臂的互動介面，透過即時顯示、虛實整合和多感官回饋功能，提升使用者在操作過程中的空間認知與路徑規劃效率。本計畫採用HoloLens 2裝置與Unity 3D開發平台，搭配混合實境開發工具MRTK，創造一個高度直覺化的操作介面。使用者能夠在真實場景中利用AR進行視覺化的機械手臂路徑規劃，並在設計過程中接收即時的視覺與聽覺反饋，進一步提高操作的直覺性。此系統整合逆運動學演算法和虛實座標校正技術，生成機器人指令以驅動機械手臂執行各類操作。研究成果包括提高機械手臂的操作效率與精度，減少專業工程人員的需求，並提升操作的可視性與直覺性。此種互動技術的創新，將有助於推動中小企業在工業5.0轉型過程中實現以人為本的智慧製造，增強其市場競爭力。此系統可提供經濟高效的解決方案，促進智慧製造技術的普及，並實現即時性人機協作的概念。

This project aims to develop an innovative motion planning system for robotic arms based on Augmented Reality (AR) to enhance human-robot collaboration on the manufacturing floor. As artificial intelligence and Industry 4.0 technologies continue to evolve, smart manufacturing is becoming a key focus across industries. However, manual operations on-site are still difficult to fully automate. Traditional robotic arm programming typically requires specialized knowledge and complex coding, posing a high barrier for small and medium-sized enterprises (SMEs) in adopting such technologies. To address this challenge, the proposed system integrates AR technology with robotic arm control systems to enable an intuitive operation mode, reducing the learning curve for engineering staff while improving operational efficiency and accuracy. By using AR as an interactive interface between humans and robots—featuring real-time visualization, seamless integration of virtual and physical environments, and multisensory feedback—the system enhances users’ spatial awareness and path planning efficiency during operation. The system is built using the HoloLens 2 headset and the Unity 3D development platform, along with Microsoft’s Mixed Reality Toolkit (MRTK), to create a highly intuitive user interface. Users can perform visualized path planning for robotic arms directly within real-world environments through AR, receiving real-time visual and auditory feedback throughout the design process to further improve usability. The system incorporates inverse kinematics algorithms and a hybrid coordinate calibration technique to generate robot control commands that drive the robotic arm to perform various tasks. Research outcomes include improved operational efficiency and precision, reduced dependence on specialized engineers, and enhanced visibility and intuitiveness in robotic operations. This breakthrough in interactive technology supports human-centric smart manufacturing and provides SMEs with a cost-effective solution during their transition to Industry 5.0. It promotes broader adoption of intelligent manufacturing technologies and enables real-time, collaborative human-robot interactions that boost competitiveness in the market.