| 摘要: | 伴隨橋梁結構老化與載重需求增加，傳統依賴人工作業的分佈式光纖佈設
效率低落、品質易受人為影響，且高空作業風險甚大，難以滿足當前結構健康監測（SHM）對大範圍、高精度佈設的需求。本研究提出並實現 ROAD （Remote-controlled distributed Optical Fiber Automated Deployment Robot）Robot—一套基於ESP32 + FreeRTOS 的遠端控制自動佈設系統，專為橋梁 DFOS 施工設計。系統整合閉環張力控制、自動膠黏釋放與紅外循跡導航三大核心模組，並透過 MQTT協定實現遠端參數設定與即時監控，使使用者可於智慧裝置上動態調整張力目標、膠黏時序與行走速度，同時監看張力曲線、路徑偏差。
張力控制模組利用高精度負載元件配合 PID 算法，在光纖放卷過程中即
時維持張力於可控範圍；膠黏單元則以事先校正的時間–體積模型精準塗佈矽膠，確保黏結品質穩定；循跡導航採用雙通道 TCRT5000 紅外感測器結合差動輪速演算法，達成毫米級路徑追蹤。經多點校正與去抖動優化後，於實驗室台架與模擬橋面測試中，ROAD Robot 在最佳速度設定下展現出行走偏差小於 3 mm 、膠量輸出重現性高，且整體佈設時間較人工短，完美消弭高空與狹窄環境的危險作業需求。
研究結果充分驗證了本系統在自動化 DFOS 佈設領域的可行性與優越性，
為未來橋梁 SHM 自動化部署提供了安全、高效、可擴充的技術路徑。後續工作將聚焦現場長期耐久試驗與多場景適應性開發，期許推動智慧基礎設施監測技術的廣泛應用與持續進化。;As bridge infrastructures age and loading demands grow, manual Distributed Fiber Optic Sensing (DFOS) installation remains slow, inconsistent, and hazardous. This thesis introduces ROAD (Remote Optical-fiber Automated Deployment) Robot, a remote-controlled system built on an ESP32 with FreeRTOS, designed for automated fiber deployment on bridge surfaces. ROAD integrates three core modules—closed-loop tension control, automated adhesive dispensing, and infrared line-following navigation—and employs MQTT for remote parameter configuration and real-time monitoring. Operators can adjust tension setpoints, dispensing intervals, and travel speed via a mobile interface, while observing live tension profiles and path deviations.
The tension module uses a high-precision load cell and PID algorithm to maintain stable fiber tension during payout. The dispensing module applies a calibrated time–volume model to ensure consistent silicone adhesive deposition. Infrared navigation relies on dual TCRT5000 sensors and differential wheel-speed control to achieve millimeter-level path fidelity. After multi-point calibration and debounce filtering, bench and mock-bridge tests demonstrate that ROAD Robot can consistently deploy fiber with lateral deviations under 3 mm, deliver repeatable adhesive volumes, reduce installation time, and eliminate dangerous manual work in elevated or confined spaces.
These results validate the feasibility and performance advantages of an automated DFOS deployment approach, offering a scalable, safe, and efficient solution for future SHM applications. Future work will focus on long-term field durability trials and adaptation to diverse structural scenarios, aiming to advance intelligent infrastructure monitoring practices. |
