全球定位系統(GPS)已被廣泛地運用於車載系統以及行動裝置上,室外定位 對於當前生活有著不可或缺的存在,但GPS對於室內定位方面目前還無法提供完 善且精準的定位標準。 無線感測網路領域中之無線射頻辨識(RFID)技術,已可透過讀取器(Reader) 與標籤(Tag)間的訊號強度,以及空間中參考標籤的佈設,來計算出誤差低於一 公尺之準確定位。本論文實作主要透過主動式RFID定位方法包含虛擬標籤群組 定位法以及LANDMARC等,再搭配不同的訊號處理方式,經由軟體計算出所需 目標標籤之所在區域後,使用無線網路TCP協定傳送資訊至機器人架構中的 Android行動裝置;Android裝置再透過藍牙通訊傳送定位資料予樂高NXT機器人, 機器人便會移動至目標標籤之所在點。最後經由實驗分析不同定位方法與整個系 統之誤差值之關係,取得有別於理論方面,對於實際定位追蹤運用上最佳的RFID 定位方法。
Although the GPS technology has been widely equipped on vehicles and mobile devices to provide outdoor positioning functionality, in terms of indoor positioning, it is failed due to receiving no satellites’ signal for GPS receivers. The contrary to the GPS technology is utilizing the RFID positioning for indoor applications, where the RSSI (return signal strength index) is the basis to locate where the object/person concerned is in. The objective of this research is to design and implement a robot tracking system based upon active RFID positioning techniques. The system consists of a Lego NXT robot as a tracker, a RFID reader for positioning, a mobile device for bridging the communication between the robot and positioning system, and an object to be tracked (which is a remote-controlled car installed with an active tag). The positioning techniques applied three signal processing methods and three positioning algorithms, and thus nine combinations can be realized and tested to compare the tracking. Both the precision of positioning and the performance of tracking were shown and compared.