運用樹莓派技術模擬工業產線之研究

A Study on Simulation of an Industry Product Line Using Raspberry Pi

林佳慷(LIN,CHIA-KANG)

指導教授 ： 陳冠宇

中原大學/工學院/機械工程研究所/碩士(2019年)

https://doi.org/10.6840/cycu201900783

摘要

隨著科技發展，微控制器(microcontroller unit, MCU)及單板電腦(single-board computer, SBC)在近年來有長足的進展，其中Arduino及樹莓派(Raspberry Pi)分別為MCU及SBC最具代表性的產品。二者最主要的差別在於樹莓派為一部小巧、功能完整，但處理速度較慢的電腦，因此具備網路與外接攝影機等Arduino沒有的功能。本文的研究目標為模擬一條具有自動組裝功能的小型產線，包括由樹莓派控制的一具機械手臂及一條輸送帶，另外藉由機器視覺結合卷積神經網路判別組裝結果。根據實驗結果顯示，本文所設計發展的小型產線，雖然初步可以完成設定的目標及功能，但仍有很大改進的空間。

關鍵字

樹莓派 ； 機器視覺 ； 卷積神經網路 ； 單板電腦

並列摘要

With the development of technology, microcontrollers (MCUs) and single-board computers (SBCs) have made great progress in recent years, where Arduino and Raspberry Pi are the most representative products of MCUs and SBCs, respectively. The main difference between the two is that Raspberry Pi is a small, full-featured, but slow computer. That is to say, Arduino does not support some features, such as the Internet and external cameras. The goal of this thesis is to simulate a small production line with automatic assembly function, including a robotic arm and a conveyor belt controlled by Raspberry Pi, and to recognize the assembly result by using machine vision combined with a convolutional neural network. According to the experimental results, the small production line designed and developed in this thesis can initially achieve the setting goals and functions, but it still have much room for improvement.

並列關鍵字

Raspberry Pi ； machine vision ； convolutional neural network ； single-board computer

參考文獻

[1] Z. Zhang, “Flexible camera calibration by viewing a plane from unknown orientations,” Proceedings of the 7th International Conference on Computer Vision, Corfu, Sep. 1999, pp. 666-673.

[2] 劉瑞禎, 于仕琪, OpenCV教程─基礎篇, 北京航空航天大學出版社, 2007.

[3] G. Bradski and A. Kaehler, Learning OpenCV: Computer Vision with the OpenCV Library, O’Reilly, 2008

[4] H. Hirschmuller, P. R. Innocent, and J. M. Garibaldi, “Real-time correlation-based stereo vision with reduced border errors,” International Journal of Computer Vision, Vol. 47, No. 1/2/3, pp. 229-246, 2002.

[5] H. Hirschmuller, F. Scholten, and G. Hirzinger, “Stereo vision based reconstruction of huge urban areas from an airborne pushbroom camera (hrsc),” Proceedings of the 27th DAGM Symposium, Vienna, Austria, 2005, Vol. LNCS 3663, pp. 58-66.