立體視覺相機於建立未知環境地圖時所採用之混合式觀測模型

A New Approach to Stereo Vision SLAM Using a Hybrid Measurement Model

袁維均(Wei-Chun Yuan)

指導教授 ： 傅立成

國立臺灣大學/電機資訊學院/資訊工程學研究所/碩士(2007年)

https://doi.org/10.6342/NTU.2007.00860

摘要

在同步定位及地圖建立的領域中，目前已有很多使用影像系統來完成此任務的方法被提出，如使用單眼相機、立體視覺相機、和全向相機等等。在本論文中，我們提出一個使用立體視覺相機做為輸入，且既簡單又有效率的觀測模型；它結合了影像輸入和環境的三維輪廓，並且用這些資訊來計算出每張影像被拍攝時的相機位置彼此在空間上的相互關係。因為我們在比對環境的三維輪廓之前會先對影像擷取出特徵點，所以我們的方法有著快速且不易被噪訊影響的優點。此外，以一連串的相機位置所組成的序列可被視為是機器人在空間中移動的軌跡，而這個結果在其他解決線上建立地圖的實做中，往往是無法得到的。

關鍵字

同步定位與地圖建立 ； 影像同步定位與地圖建立 ； 立體視覺 ； 機器人學 ； 主動式相機

並列摘要

In visual Simultaneous Localization and Mapping, many models are proposed to make use of various kinds of vision systems including monocular camera, stereo camera, and omni-directional camera. We propose a measurement model suitable for stereo systems having good properties of efficiency and simplicity. Our approach combines visual information with three dimensional point clouds as spatial contour to calculate the relative pose of the spots where the image pairs are taken by the stereo cameras. In our method, feature extraction is done before matching the point clouds, so our model is faster than others and is robust to measurement noise. Furthermore, the sequence of the camera poses yet offers the trajectory of the robot, which is usually unavailable in online SLAM implementation.

並列關鍵字

SLAM ； vSLAM ； Stereo Vision ； Stereo vSLAM ； Robotics ； Active Camera

參考文獻

[2] R. Smith, M. Self, and P. Cheeseman, “Estimating uncertain spatial relationships in robotics,” in IEEE International Conference on Robotics and Automation, vol. 4, 1987, pp. 850–850.

[3] M. Montemerlo, S. Thrun, D. Koller, and B. Wegbreit, “Fastslam: A factored solution to the simultaneous localization and mapping problem,” in Proceedings of the

AAAI National Conference on Artificial Intel ligence, Edmonton, Canada, 2002.

[4] L. Armesto, G. Ippoliti, S. Longhi, and J. Tornero, “Fastslam 2.0: Least-squares approach,” in IEEE/RSJ International Conference on Intel ligent Robots and Systems, 2006, pp. 5013–5018.

[5] C.-C. Wang, C. Thorpe, and S. Thrun, “Online simultaneous localization and mapping with detection and tracking of moving ob jects: Theory and results from a ground vehicle in crowded urban areas,” in Proceedings of the IEEE International Conference on Robotics and Automation, Taipei, Taiwan, September 2003.