以視覺聯網輔助機器人在非結構化環境中執行定位與挑揀任務

機構典藏 > College of Engineering > Graduate Institute & Department of Mechanical and Electro-Mechanical Engineering > Thesis > Item 987654321/111408

Please use this identifier to cite or link to this item: https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/111408

Title:	以視覺聯網輔助機器人在非結構化環境中執行定位與挑揀任務
Other Titles:	Mobile robot localization and manipulator bin-picking in unstructured environments using internet of vision
Authors:	黃品叡;Hwang, Pin-Jui
Contributors:	淡江大學機械與機電工程學系碩士班王銀添;Wang, Yin-Tien
Keywords:	機器人聯網;機器人物件挑揀;機器人定位;連續適應均值移動;非結構自動化;internet of robotic things;robot bin-picking;Robot Localization;continuously adaptive mean-shift, CAM-Shift;unstructured automation.
Date:	2016
Issue Date:	2017-08-24 23:52:02 (UTC+8)
Abstract:	本研究發展視覺聯網系統，並且應用於零售業的倉儲自動化。目的是以聯網的視覺輔助行動機器人於非結構化環境中進行定位任務，以及使用視覺輔助偵測與辨識貨品，進而導引機器手臂夾取貨品。本論文的研究議題包括自動倉儲網聯技術、機器人定位問題、與視覺式貨品挑揀等。自動倉儲網聯將機器人、視覺與感測器等以網際網路連結，以便進行任務規劃與控制命令傳輸等。機器人定位則是使用低功率藍芽(Bluetooth Low Energy, BLE)定位與固定攝影機QR-code定位，進行行動機器人在工作區範圍內與範圍外的定位任務。視覺輔助貨品挑揀則是以影像色彩直方圖搭配連續適應均值移動(continuously adaptive mean-shift, CAM-Shift)演算法，進行貨品的偵測辨識與追蹤，並且導引機器手臂進行夾取貨品的任務。本研究將在PC-based控制器內建立所需的發展環境，以Visual C++程式語言整合各種感測器、SCARA手臂控制、網路伺服器、影像偵測辨識演算法等。最後以整合的系統執行聯網視覺輔助機器人定位與貨品挑揀之任務，以評量所提演算法的效能。 This study investigates the issues of internet of robotic things. The purpose is to localize the robot in environment automation using internet of visual sensor and present a vision-based robotic bin-picking system. The issues of the study include internet of robotic things, robot localization and vision-based robotic bin-picking system. Automated storage and retrieval system connect robot and vision sensor with internet in order to planning and give instructions. Robot localization divide into two parts, first is using Bluetooth Low Energy (BLE) localization or known as iBeacon localization. Another localization is QR-code localization whit IP camera. The first part is for outside the workspace and the QR-code localization is for the workspace. Vision-based robotic bin-picking system is using color histogram and continuously adaptive mean-shift (CAM-Shift) to identify the product and guide the robot arm to pick the product. The software program of this study investigates is developed in a PC-based controller using Microsoft Visual Studio C++. The resultant system is used to perform the internet of visual sensor localization and vision-based robotic bin-picking system.
Appears in Collections:	[Graduate Institute & Department of Mechanical and Electro-Mechanical Engineering] Thesis

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	130	View/Open

Loading...