6.1 結論
本論文以 Kinect 深度攝影機為感測裝置，結合視覺感測裝置與距離感測裝 置，以 SURF 演算法對影像作特徵點擷取，對應深度資訊計算特徵點之三維座標，
取得環境中具獨特性之特徵點並能快速取得對應之距離關係，這些特徵點訊息透 過 EKF 定位系統演算法修正機器人狀態並更新地圖資料庫。在機器人持續運行 之同時利用區域路徑範圍的判斷，將環境區隔為數個範圍較小的區域地圖降低定 位系統的運算複雜度，達成機器人同時定位與地圖建立之即時運算。最後，透過 地圖融合演算法，以地圖間共同特徵之差異整合地圖間資訊並修正區域融合後機 器人狀態，達成具備即時運算且具完整地圖結果之機器人定位與地圖建立系統。
本論文透過實驗證實同時定位與建立地圖結合地圖接合之作法實現於雙獨 立驅動輪式機器人之功效，以本論文所設計之方形路徑實驗，在已建立過之環境 中運行時，耗費運算量最大之區域每次約需 1.2 秒完成所有定位系統之運算，在 不同區域間運算的結果整體平均每次約需 646ms，當機器人應用於長距離之機器 人定位與導航之實驗，機器人於一約 16m X 7m 之環境行走 83m 後，當機器人回 至原點附近時的實際位置與估測狀態平均 X 方向誤差為 7.21 公分，平均 Y 方向 誤差為 7.16 公分而平均角度差則為 1.64 度。
6.2 未來展望
影像以 SURF 取得之特徵點在抗尺度、光源變化影像旋轉以及運算速度等皆 有極佳的表現，然對物體之視角有較大不同時，SURF 特徵點並無法有效的判斷 並比對相同之環境資訊，若以增加感測器之視角以加強感測範圍，則又必頇增加 定位系統所需之運算成本，降低機器人即時運算之效能。這使得機器人在相同的 環境以不偏離取得環境資料的觀測角度太多的周圍移動，大幅降低機器人 SLAM 應用於實際情境之可行性，若能以影像取得在視角變化下之相同環境資訊，則應 能有效強化以視覺感測裝置在 SLAM 於實際環境下之應用。
本論文以深度攝影機 Kinect 結合視覺及距離感測裝置之作法是以視覺感測 器為主取得影像特徵點，距離感測器為輔用以計算影像特徵點之三維座標，若以 距離感測裝置取得環境結構與分佈特性做為第二種環境訊息結合於定位系統，則 定位系統在特徵點稀少視覺系統難以發揮之場所將能有所依據，以兩者相輔相成 可望讓機器人 SLAM 更廣泛應用於各種場所。
參考文獻
[1] Hugh Durrant-Whyte and Tim Bailey, “TUTORIAL : Simultaneous Localization and Mapping : Part 1 ,” IEEE Robotics & Automation Magazine, Vol. 13, No. 2, pp 99-108, 2006
[2] Tim Bailey and Hugh Durrant-Whyte, “TUTORIAL : Simultaneous Localization and Mapping : Part II,” IEEE Robotics & Automation Magazine, Vol. 13, No. 2, pp. 108 -117, 2006
[3] 袁立德,“基於單眼視覺之機器人迴圈原點及同步定位與地圖建立,” 國立交 通大學電機與控制系碩士論文, 2010.
[4] Brian Williams, Paul Smith and Ian Reid,“ Automatic Relocalisation for a Single-Camera Simultaneous Localization and Mapping System,” in Proceedings of IEEE International Conference on Robotics and Automation (ICRA), Roma, Italy, 2007, pp. 2784-2790.
[5] 王兆戊,“全向式移動機器人之同步定位與環境地圖建立,”國立交通大學電機 與控制系碩士論文, 2008.
[6] 黃富盛,“基於權向式影像之機器人同步定位與環境地圖建立,”國立交通大學 電機與控制系碩士論文, 2008.
[7] Olivier Koch, Matthew R. Walter, Albert S. Huang and Seth Teller, “ Ground Robot Navigation using Uncalibrated Cameras,” in Proceedings of IEEE International Conference on Robotics and Automation (ICRA), Anchorage, AK, USA, 2010, pp. 2423-2430.
[8] Olivier Koch and Seth Teller, “ Body-Relative Navigation Guidance Using Uncalibrated Cameras,” in Proceedings of IEEE International Conference on Computer Vision(ICCV), Kyoto, Japan, 2009, pp. 1242 – 1249.
[9] Cristof Schroeter and Horst-Michael Gross, “ A Sensor-Independent Approach to RBPF SLAM –Map Match SLAM applied to Visual Mapping,” in Proceedings of IEEE International Conference on Robotics and Automation (ICRA) , Nice, France, 2008, pp. 2078-2083.
[10] S. Thrun, D. Fox and W. Burgard, “A Probabilistic Approach to Concurrent Mapping and Localization for Mobile Robots,” Machine Learning and Autonomous Robots, vol.31, pp. 29-53, 1998.
[11] J. Tard´os, J. Neira, P. Newman and J. Leonard, “ Robust Mapping and Localization in Indoor Environments using Sonar Data, ” Int. J. Robotics Research, vol.21, pp. 311-330, 2002.
[12] N. Ganganath and H. Leung, “ Mobile Robot Localization Using Odometry And Kinect Sensor, ” in Proceedings of IEEE International Conference on
Emerging Signal Processing Applications (ESPA),Las Vegas, NV, USA, 2012, pp.
91 – 94.
[13] Gert Koostra, Sjoerd de Jong and Lambert R.B Schomaker, “Using Local Symmetry for Landmark Selection,” in Proceedings of International Conference on Computer Vision Systems , Liege, Belgium, 2009, pp. 94-103
[14] Gert Koostra and Lambert R.B. Schomaker, “Using Symmetrical Regions of Interest to Improve Visual SLAM,” in Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), St. Louis, MO, USA, 2009, pp. 930-935.
[15] J. Shi and C. Tomasi, “Good Features to Track,” in Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA, 1994, pp. 593-600.
[16] David G. Lowe, “ Distinctive Image Features from Scale-Invariant Keypoints, ” International Journal of Computer Vision, vol. 60, no. 2, pp. 91-110, Nov. 2004.
[17] H. Bay, T. Tuytelaars and L. V. Gool, “SURF: Speeded Up Robust Features” in Proc. of the 9th European Conf. on Computer Vision, Graz Austria, 2006, pp.
404-417.
[18] L.M. Paz, P. Jensfelt, J.D. Tard´os and J. Neira, “EKF SLAM Updates in O(n) with Divide and Conquer SLAM,” in Proc. of IEEE International Conference on Robotics and Automation (ICRA), Roma, Italy, 2007, pp. 1657-1663.
[19] Carlos Estrada, J. Neira and Juan D. Tardos, “ Hierarchical SLAM: Real-Time Accurate Mapping of Large Environments, ”in IEEE Transactions on Robotics and Automation,Vol.21 No.4, pp. 588-596, 2005.
[20] Brain Williams, Mark Cummins, José Neira, Paul Newman, Ian Reid and Juan Tardos, “An Image-to-map Loop Closing Method for Monocular SLAM,” in Proc. of IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Nice, France, 2008, pp. 2053-2059.
[21] Adrien Angeli, Stѐphane Doncieux, Jean-Arcady Meyer and David Filliat,
“Real-Time Visual Loop-Closure Detection,” in Proc. of IEEE International Conference on Robotics and Automation (ICRA), Pasadena, CA, USA, 2008 ,pp.
1842-1847.
[22] http://www.play.com/Games/Xbox360/4-/10296372/Project-Natal/Product.html [23] http://www.primesense.com/(Primesense)
[24] Christopher Evans , “Note on the OpenSURF Library, ”, 2009.
[25] http://75.98.78.94/(OpenNI)
[26] S. Thrun, D. Fox and W. Burgard, Probabilistic Robotics, Massachusetts Institute of Technology, 2005.
[27] Jose’ Neira, Juan D. Tardos, and Jose A. Castellanos , “Linear Time Vehicle
Relocation in SLAM, ” in Proc. IEEE International Conference on Robotics and Automation(ICRA),Taipei, Taiwan, 2003, pp427-433.
[28] Joan Sola, “ Consistency of the Monocular EKF-SLAM Algorithm for Three Different Landmark Parametrizations, ”in Proc. of IEEE International Conference on Robotics and Automation (ICRA), Anchorage, AK, USA, 2010, pp. 3513-3518.
[29] Chanki Kim and Wan Kyun Chung, “Delayed Resampling in a Rao-Blackwellized Particle Filter SLAM for Consistent Loop Closures,” in Proc.
of IEEE/RSJ International Conference on Intelligent Robots and Systems(IROS), Nice, France, 2008, pp. 2084-2090.
[30] Mark Cummins and Paul Newman, “Probabilistic Appearance Based Navigation and Loop Closing,” in Proceedings of IEEE International Conference on Robotics and Automation (ICRA), Roma, Italy, 2007,pp. 2042 - 2048.
[31] Mark Cummins and Paul Newman,“ Accelerated Appearance-only SLAM, ”in Proc. of IEEE International Conference on Robotics and Automation(ICRA) , Pasadena, CA, USA, 2008, pp.1828-1833.
[32] Jungho Kim, Ouk Choi and In So Kweon, “Efficient Feature Tracking for Scene Recognition using Angular and Scale Constraint,” in Proc. of IEEE/RSJ International Conference on Intelligent Robots and Systems(IROS), Nice, France 2008, pp.4086-4091.