應用蟻群演算法於倉儲揀貨路徑規劃

Application of Ant Colony Optimization to Route Planning for Warehouse Robots Picking up Goods

本文主要研究一具備最佳揀料路徑規劃的自動倉儲系統，此系統能依據所要的揀料料件，在手臂存取機限定負荷約束下，達到揀料趟數最少、路徑短和最節省時間的目的。硬體方面，以PLC為基礎，整合人機介面和X-Y-Z軸取放料機械手臂模組，讓機台能進行三軸運動而取、放料件。軟體方面，以GX Works2 軟體平台撰寫PLC 的三軸機械手臂取放料驅動程式。以GT Designer 3軟體平台設計友善的揀料參數輸入和機台監控的人機介面。發展蟻群演算法來規劃倉儲揀料的最佳路徑以達到快速與高效率的要求。再透過Matlab軟體平台撰寫蟻群演算的模擬和實驗程式。最後，以實際的自動倉儲機台，透過蟻群演算法進行最佳路徑規劃的揀貨實驗，來驗證本文所提方法的可行性。

This study developed an automated storage and retrieval system (AS/RS) with optimal path planning for robots picking up goods in a warehouse. The system is designed to determine the most efficient picking path with the fewest trips, shortest distance, and fastest completion time under the load constraints of the robotic arm. On the hardware side, the system is based on a programmable logic controller with a human–machine interface (HMI) with an x–y–z-axis robotic arm module for retrieving and placing goods. On the software side, the PLC programming for the three-axis robotic arm is implemented using the GX Works2 platform. A user-friendly HMI for inputting parameters and monitoring the system is designed using the GT Designer 3 platform. This study developed an ant colony optimization (ACO) algorithm that determines the most efficient picking path. The ACO algorithm was simulated and evaluated in MATLAB and implemented on a physical automated storage system in a real-world evaluation experiment.