基於ROS開發工業應用之無人搬運車安全及強健移動式機器人導航策略

Abstract

現今自主式移動機器人已成為製造工廠邁向智慧化的重要元素，本論文主要為提出無人搬運車(AGV)的設計與開發，並整合該無人搬運車將軟硬體的設計至以機器人作業系統(ROS)環境為主的車載控制系統中。為了滿足製造過程中具備有高度彈性化物料運行的要求，本文開發的AGV整合了多種功能，可自由的在站點之間進行導航，並能閃避障礙物，以及與遠端監控中心進行協調。除此之外，AGV系統中還包括即時定位與建圖（SLAM）、路徑規劃、路徑追蹤、搬運貨料架、自動充電以及數據記錄等技術，並在類產業環境下成功進行了測試。為了確保AGV在導航過程中是否能安全移動，透過動態窗口避障法（DWA）在路徑追蹤時可閃避障礙物並回到原本規劃之導航路徑上。另一方面，使用者可通過設計好的可視化界面遠端監控AGV的運行狀態，並自行規劃AGV的運行流程。本文提出的AGV自動化物料搬運方案，在實際工廠生產線環境中作進一步驗證，實驗結果證實，本論文所開發AGV之相關技術確實提供了有效且安全的解決方案，能夠實際運用於在工廠生產線實際環境中。Nowadays, autonomous mobile robots are becoming as the important elements in manufacturing facilities moving toward smart factory. This thesis presents the design and development of an automated guided vehicle (AGV) that incorporates hardware and software designs into the on-board control system based on robot operation system (ROS). In order to operate within manufacturing processes with highly flexible material flow requirements, the developed AGV incorporates several capabilities to navigate the freely between stations while avoiding obstacles and coordinating with the remote control center. Additionally, techniques such as simultaneously localization and mapping (SLAM), path planning and tracking, docking and navigation, handling cargo racks and automatic charging, and data recording are also included and tested successfully under the industrial scenario. To ensure the safety during the navigation process, dynamic window approach (DWA) is employed to utilize online motion planning to avoid the recently revealed obstacles. On the other hand, users can remotely monitor the operation status of AGV, and set up the process flow for AGV through the designed user interface. The AGV with the proposed method for the automatic configuration of material handling is furtherly validated in the production line of industry. The experimental results depict the related technologies of the developed AGV as a promising solution to viable, effective, and safe operation in industrial environments.