立體視覺應用於多自主行動機器人無線感 測網路定位技術之研究

A Study on Wireless Sensor Network Localization for Multiple Autonomous Mobile Robots Using Stereovision

10.6840/cycu201500582

王荷凱

無線感測網路 ； 室內定位 ； 自主行動服務型機器人 ； 立體視覺視差 ； 雲端同步 ； Wireless sensor network ； Indoor localization ； Autonomous mobile service robot ； Stereovision algorithm ； Cloud sync

中原大學機械工程學系學位論文

2015年

碩士

陳冠宇

繁體中文

一般而言，服務型機器人的工作環境是以室內為主，而當多部機器人在環境中執行 任務時，若沒有可以即時監控的工具、精確的定位技術和良好的避障系統，則服務型機 器人很難確實的完成任務。近年來，利用無線感測網路技術建置室內定位系統的應用明 顯增加，乃因其具有架構簡單、小尺寸、低成本、低消耗功率、相互連結性高等特點， 此技術可應用於多部自主行動機器人服務的室內辦公環境中進行文件包裹遞送的物流 作業系統。本文將先前的研究成果加入了雙目立體視覺視差圖和即時雲端同步，首先藉 由建立立體視差圖，讓機器人能自主的完成靜態及動態障礙物的閃避；再者將即時雲端 同步系統架設在各部機器人上，使用者能在遠端得知每部機器人的即時狀態；最後結合 多部機器人在室內工作環境中相遇的避障系統，再使用超音波感測器及ZigBee 無線感 測網路的定位訊號，針對不同的相遇情況使其能各自安全閃避與執行任務。

Typically, service robots are mainly designed to work in indoor environments. In fact, if there is the lack of real-time monitoring tool, precise positioning technology and good obstacle avoidance system, service robots can not actually perform tasks well. Recently, due to its simple structure, small size, low cost, low power consumption, and easy to connect with each other, the wireless sensor network has been widely used for constructing the indoor locatization system. This positioning technique can be applied to a logistics operating system for controlling multiple autonomous mobile robots working in an intelligent office environment for the delivery of documents and packages. In this paper, the methods of binocular disparity map and real-time cloud sync are integrated into the results of previous studies. Firstly, we measure the depth of each obstacle by computing the disparity map from stereo-pair images to let a service robot can avoid both static and dynamic obstacles. Next, users can remotely monitor real-time status of every robot by installing the cloud sync system on each robot. Finally, by integration of ultrasonic sensors and WSN’s positioning signals into an obstacle avoidance system for multiple robots, these robots can avoid each other in different situations.

1. 謝宗浩(2016)。以OpenCV發展立體視覺線掃描攝影機平台之三維影像重建技術。中原大學機械工程學系學位論文。2016。1-57。