智慧型控制理論之機器人定位系統開發

The development of Robot Positioning System with Intelligent Control

洪博文(Bo-Wen Hong)

指導教授 ： 黃英哲

元智大學/電機通訊學院/電機工程學系/博士(2012年)

摘要

近年來，無線射頻與影像處理系統技術被廣泛應用於智慧住宅與大型醫療院所建置之整合的室內即時定位系統。本文提出新的具影像辨識與模糊類神經網路之RFID室內定位架構系統，解決智慧型機器人之室內即時定位問題，該方法建立於系統之接收訊號強度判斷基礎上，系統架構採用2.4GHz主動式RFID元件，該系統具備讀寫器與標籤的長距離傳輸優點，以及搭配模糊類神經網路控制器的強健性與適應性，並於範圍面積大的室內進行定位，在此將提出的影像辨識搭配模糊類神經RFID方法用以解決一系列的智慧型機器人之室內即時定位問題，研究工作建構於兩種架構上，透過實驗結果成功的表現可以得知該系統具有很好的精確度以及具備適應性，在此可以提供智慧型充電機器人與家庭保健輪椅機器人定位與導航的方式，該架構同時可以降低系統成本與減少系統運算複雜性。

關鍵字

無線射頻辨識系統 ； 室內即時定位 ； 影像辨識 ； 模糊類神經網路 ； 智慧型機器人

並列摘要

In recent years, Radio Frequency Identification system (RFID) and image processing system technology has been widely applied to intelligent houses and large medical institutions to build real-time positioning system. This dissertation proposes a fuzzy neural network based on image processing system for real time navigating and RFID indoor positioning the intelligent robots. The technology is based on received signal strength indication method. The proposed system structure includes 2.4GHz active RFID devices and camera. Since active RFID devices have a prominent capability on long distance communication, it is the reason why we apply to the localization for a intelligent mobile robot. The fuzzy neural network controller provides robustness and adaptive rule for environment. Further, we solve the problem of real-time indoor positioning for intelligent robot with RFID-based fuzzy neural network. In our research work, two structures developed the successful experiment results show that the proposed system architecture and positioning system provides very good accuracy. In the meantime, it makes intelligent self-charging robot and home health care wheelchair robot positioning system available for navigation and guidance. The proposed indoor localization system also has two advantages of cost reduction and simple computation.

並列關鍵字

Radio frequency identification system ； Real-time indoor positioning ； Image processing ； Fuzzy neural network ； Intelligent Robot

參考文獻

[2] M. Mühlig, M. Gienger, and J. J. Steil, “Human-robot interaction for learning and adaptation of object movements,” IEEE/RSJ 2010 International Conference on Intelligent Robots and Systems, pp. 4901-4907, 2010.

[4] L. Vachhani and K. Sridharan, “Hardware-Efficient Prediction-Correction-Based Generalized-Voronoi-Diagram Construction and FPGA Implementation,” IEEE Transactions on Industrial Electronics, vol. 55, no. 4, pp. 1558-1569, 2008.

[5] Y. Misono, Y. Goto, Y. Tarutoko, K. Kobayashi, and K. Watanabe, “Development of Laser Rangefinder-based SLAM Algorithm for Mobile Robot Navigation,” SICE Annual Conference, pp. 329-396, 2007.

[7] D. I. B. Randeniya, S. Sarkar, and M. Gunaratne, “Vision–IMU Integration Using a Slow-Frame-Rate Monocular Vision System in an Actual Roadway Setting,” IEEE Transactions on Intelligent Transportation Systems, vol.11, no. 2, pp. 256-266, 2010.

[8] D. Bank and T. Kämpke, “High-Resolution Ultrasonic Environment Imaging,” IEEE Transactions on Robotics, vol. 23, no. 2, pp. 370-381, 2007.