機器人定位一直是行動機器人研究領域中的一項重要課題，因為唯有機器人知道自己在目前環境中的位置前提下，才能完成機器人被交代的自主任務。使用視覺感測器做機器人定位除了不需借助額外設備就可以達到定位外，還可以得到目前環境中的地標物之圖像資訊。在利用圖像中特徵點作機器人視覺定位的方法中，較常被使用的是利用連續圖像中相對應特徵點的移動方向及距離來推算出目前拍照機器人的位置，不過因為此方法需對圖像中每個可以辨識的特徵點做運算導致其運算量太大的缺點，所以我們想要結合圖像中之角點分佈及幾何透視投影的概念來做機器人定位，如此可以減輕數學運算負擔且可以得到不差的定位結果。藉由偵測圖像中的角點及其深度資訊，本論文利用影像特徵點的群聚性質在環境中建立垂直狀地標，並利用地標座標比對與影像透視原理完成共三階段之定位，改善以往只利用地標間距離作比對所可能產生的錯誤比對關係。經由室內控制環境定位測試到室外實際環境照相定位之結果，可以用來驗證我們的方法。

Localization is one of the most important issues for mobile robots. Because only a robot knows its position in the environment, it can complete the given task. Visual localization involves detecting feature points in images and building correct correspondences between features in consecutive images. However, the computational costs of processing features points is high. This paper proposes a relatively light-weight localization approach, which preprocesses corner features detected by stereo vision into vertical clusters as natural landmarks in the environment, then localizes the robot by matching locations and perspective geometric properties of landmarks. The experimental results shows that the proposed approach works relatively well in the outdoor environment.

[1] C. Harris and M. Stephens, “A combined corner and edge detection,” in Pro¬ceedings of The Fourth Alvey Vision Conference, pp. 147–151, 1988.
[2] K. Sugihara, “Some location problems for robot navigation using a single cam¬era,” Comput. Vision Graph. Image Process., vol. 42, no. 1, pp. 112–129, 1988.
[3] E. Krotkov, “Mobile robot localization using a single image,” in Proceedings of the IEEE International Conference on Robotics and Automation (ICRA ’89), vol. 2, pp. 978 – 983, May 1989.
[4] G. Caron and E.-M. Mouaddib, “Vertical line matching for omnidirectional stereovision images,” in ICRA ’09. IEEE International Conference on Robotics and Automation, 2009., pp. 2787–2792, May 2009.
[5] Videre-Design, “http://www.videredesign.com,”
[6] Y. Matsumoto and A. Zelinsky, “An algorithm for real-time stereo vision imple¬mentation of head pose and gaze direction measurement,” IEEE International Conference on Automatic Face and Gesture Recognition, vol. 0, p. 499, 2000.
[7] S. N. H. S. Abdullah, M. Khalid, R. Yusof, and K. Omar, “Comparison of feature extractors in license plate recognition,” AMS ’07: Proceedings of the First Asia International Conference on Modelling & Simulation, pp. 502–506, 2007.
[8] G. N. DeSouza and A. C. Kak, “Vision for mobile robot navigation: A survey,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 24, no. 2, pp. 237–267, 2002.
[9] J. Borenstein, H. R. Everett, and L. Feng, Navigating Mobile Robots: Systems and Techniques. Natick, MA, USA: A. K. Peters, Ltd., 1996.
[10] D. Marr and T. Poggio, “Cooperative computation of stereo disparity,” pp. 259– 267, 1988.
[11] A. Cumani, S. Denasi, A. Guiducci, and G. Quaglia, “Robot localisation and mapping with stereo vision,” in ISCGAV’04: Proceedings of the 4th WSEAS International Conference on Signal Processing, Computational Geometry & Artificial Vision, (Stevens Point, Wisconsin, USA), pp. 1–6, World Scientific and Engineering Academy and Society (WSEAS), 2004.
[12] A. Cumani and A. Guiducci, “Mobile robot localisation with stereo vision,” in ISCGAV’05: Proceedings of the 5th WSEAS International Conference on Signal Processing, Computational Geometry & Artificial Vision, (Stevens Point, Wisconsin, USA), pp. 176–181, World Scientific and Engineering Academy and Society (WSEAS), 2005.
[13] S. Barnard and W. Thompson, “Disparity analysis of images,” IEEE Transac¬tions on Pattern Analysis and Machine Intelligence, vol. 2, pp. 333–340, July 1980.
[14] M. Trajkovic and M. Hedley, “Fast corner detection,” Image and Vision Com¬puting, vol. 16, no. 2, pp. 75 – 87, 1998.
[15] H. Moravec, “Towards automatic visual obstacle avoidance,” in Proceedings of the 5th International Joint Conference on Artificial Intelligence, p. 584, August 1977.
[16] H. Moravec, “Visual mapping by a robot rover,” in Proceedings of the 6th International Joint Conference on Artificial Intelligence, pp. 599–601, August 1979.
[17] M. A. Fischler and R. C. Bolles, “Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography,” pp. 726–740, 1987.
[18] R. M. Haralick, Y. H. Chu, L. T. Watson, and L. G. Shapiro, “Matching wire frame objects from their two dimensional perspective projections,” Pattern Recognition, vol. 17, no. 6, pp. 607–619, 1984.
[19] S. Aitya and G. Hager, “Real-time vision-based robot localization,” in Proceed¬ings of IEEE International Conference on Robotics and Automation, pp. 639– 644 vol.1, Apr 1991.
[20] B. D. Lucas and T. Kanade, “An iterative image registration technique with an application to stereo vision,” in Proceedings of the 7th International Joint Conference on Artificial Intelligence (IJCAI ’81), pp. 674–679, April 1981.
[21] K. Konolige, “Small vision systems: hardware and implementation,” in Eighth International Symposium on Robotics Research, p. 111–116, 1997.
[22] I. Carlbom and J. Paciorek, “Planar geometric projections and viewing trans¬formations,” ACM Comput. Surv., vol. 10, no. 4, pp. 465–502, 1978.
[23] M. Betke and L. Gurvits, “Mobile robot localization using landmarks,” in Pro¬ceedings of the IEEE/RSJ/GI International Conference on Intelligent Robots and Systems ’94. ’Advanced Robotic Systems and the Real World’, IROS ’94., vol. 1, pp. 135–142 vol.1, Sep 1994.
[24] J.-H. Haunert and C. Brenner, “Vehicle localization by matching triangulated point patterns,” GIS ’09: Proceedings of the 17th ACM SIGSPATIAL Inter¬national Conference on Advances in Geographic Information Systems, pp. 344– 351, 2009.