應用於雙眼視覺導航之機器人定位及地圖重建系統

Localization and Mapping for Mobile Robot Navigation with Binocular Vision

涂緒昌(Shiu-Chang Tu)

指導教授 ： 張文中

國立臺北科技大學/電資學院/電機工程系所/碩士(2010年)

摘要

精確的環境地圖在機器人導航的領域中是相當重要的資訊, 因此本論文提出一種雙眼視覺導航系統以完成未知室內環境之即時定位及地圖重建, 利用建立的地圖作為機器人導航之依據。此架構使用兩部攝影機作為感測器, 利用改良式SURF(Speeded Up Robust Features) 特徵點演算法, 使其適合應用於本系統且利於環境偵測以找出現實空間中的特徵點投影在兩影像中的位置,進而藉由兩攝影機之間幾何關係重建出特徵點在現實空間的座標,同時判斷特徵點之間的關係以找出機器人可行走範圍, 並藉由可移動範圍與機器人之間的關係, 設計出導航與避障法則, 使機器人可自主在未知環境中移動並避開障礙物。本論文中所提出之架構已藉一自製輪式機器人系統成功驗證於未知現實環境中。

關鍵字

機器人導航 ； 雙眼視覺 ； SURF ； 移動型機器人

並列摘要

An accurate environment map provides essential information for navigation and control of mobile robots. This thesis presents a binocular visual navigation system to accomplish localization and mapping in an unknown environment. In this
system, two cameras are used as the sensors and the modif ed SURF (Speeded Up Robust Features) interest point detector and descriptor are applied. Interest points observed by the two cameras are detected by SURF to further reconstruct
their coordinates in Cartesian space by the known geometric relation between the two cameras. The reconstructed interest points are then employed to design the navigation rule. The experiments have been successfully implemented using a custom-made wheeled robot equipped with two vedio cameras.

並列關鍵字

Robot Navigation ； Binocular Vision ； SURF ； Mobile Robot

參考文獻

[1] W.-C. Chang. Hybrid force and vision-based contour following of planar robots. Journal of Intelligent and Robotic Systems, 47(3):215–237, November 2006.

[2] W.-C. Chang. Binocular vision-based trajectory following for autonomous robotic manipulation. Robotica, April 2007.

[3] W.-C. Chang. Precise positioning of binocular eye-to-hand robotic manipulators. Journal of Intelligent and Robotic Systems, 49(3):219–236, July 2007.

[4] W.-C. Chang and S.-A. Lee. Autonomous visual servoing with tele-supervision. In Proc. of the 2004 IEEE International Conference on Networking, Sensing and Control,

[5] W.-C. Chang and S.-A. Lee. Autonomous visual-based pose cotrol of mobile robots with tele-supervision. In Proc. of the 2004 IEEE CCA/ISIC/CACSD, Taipei, Taiwan, September 2004.