在智慧物流與倉儲領域中，貨物在各環節中的監控管理非常重要，以往需要以人工進行追蹤與搬運的作法是相當耗時且耗費人力的，因此引入具自主規劃能力的運送系統是當今趨勢，對於倉儲內自動撿貨以及末端物流配送將會有重大影響。若要使機器能在環境中具有可靠的規劃能力，則需要好的環境感知能力來認知當前載具於空間中的位置，而SLAM技術是賦予機器環境感知能力的關鍵。
據此，本研究將會聚焦於已建立環境高精度點雲地圖的場域中，期望可提供載具執行固定場域之常態性任務時的空間定位能力，使得載具能夠穩定執行任務與路徑的規劃。而定位核心演算法將使用基於點雲數據的常態分佈變換演算法(Normal Distribution Transform, NDT)，來實現載具即時空間定位任務，利用常態分佈變換演算法可以離線建立特徵地圖的優勢提高整體演算法的效率。
為改善常態分佈變換演算法的定位精度，本研究將引入高斯混合模型(Gaussian Mixture Model, GMM)來更精確描述點雲分布特徵，並藉由引入分群演算法與最佳分群數選擇方法，高效率完成常態分佈變換演算法與高斯混合模型的建立流程。
在定位流程的後端中，本研究引入了卡爾曼濾波器作為載具位姿的估測，利用物理運動學模型的狀態預測融合常態分佈變換演算法定位結果，來實現當前載具位姿的最佳估測。
於實驗章節中，將在室內外、倉儲、車輛駕駛等場景進行定位演算法驗證，本研究期望開發出可廣泛應用於各場域之定位演算法，用以克服室內GPS無法定位與室外遮擋等問題，提供載具可靠且精確的即時定位結果。

This research intends to develop an accurate and real-time 3D localization algorithm for robots. By using LiDAR to acquire point cloud from surrounding environment, SLAM technique can localize the relative position between two point clouds. This research focuses on the fields that have built the HD point cloud map, using SLAM technique to localize robot by matching current point cloud and HD point cloud map.
Firstly, Normal Distribution Transform (NDT) is applied for point cloud matching problem. Secondly, to improve the performance of NDT, Gaussian Mixture Model (GMM) is applied to build distribution features in NDT. The techniques for data clustering and determining the optimal numbers of clusters are introduced for creating GMM. At last, Kalman Filter (KF) is applied for optimal robot position estimation by fusing the NDT localization result and position states predicted from motion model. Our localization algorithm is experimented and evaluated in variable scenarios like indoor, outdoor and driving environment, the precision of our algorithm is better than other NDT-based algorithm.

[1] C. Ulas and H. Temeltas, "3D Multi-Layered Normal Distribution Transform for Fast and Long Range Scan Matching," Journal of Intelligent & Robotic Systems, vol. 71, 07/01 2013, doi: 10.1007/s10846-012-9780-8.
[2] J. Zhang and S. Singh, "LOAM : Lidar Odometry and Mapping in real-time," Robotics: Science and Systems Conference (RSS), pp. 109-111, 01/01 2014.
[3] E. Rublee, V. Rabaud, K. Konolige, and G. Bradski, "ORB: An efficient alternative to SIFT or SURF," in 2011 International Conference on Computer Vision, 6-13 Nov. 2011 2011, pp. 2564-2571, doi: 10.1109/ICCV.2011.6126544.
[4] R. Mur-Artal, J. M. M. Montiel, and J. D. Tardós, "ORB-SLAM: A Versatile and Accurate Monocular SLAM System," IEEE Transactions on Robotics, vol. 31, no. 5, pp. 1147-1163, 2015, doi: 10.1109/TRO.2015.2463671.
[5] P. J. Besl and N. D. McKay, "A method for registration of 3-D shapes," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 14, no. 2, pp. 239-256, 1992, doi: 10.1109/34.121791.
[6] E. Recherche, E. Automatique, S. Antipolis, and Z. Zhang, "Iterative Point Matching for Registration of Free-Form Curves," Int. J. Comput. Vision, vol. 13, 07/01 1992.
[7] Y. Chen and G. Medioni, "Object modeling by registration of multiple range images," in Proceedings. 1991 IEEE International Conference on Robotics and Automation, 9-11 April 1991 1991, pp. 2724-2729 vol.3, doi: 10.1109/ROBOT.1991.132043.
[8] K.-L. Low, "Linear Least-Squares Optimization for Point-to-Plane ICP Surface Registration," 01/01 2004.
[9] P. Biber and W. Straßer, The Normal Distributions Transform: A New Approach to Laser Scan Matching. 2003, pp. 2743-2748 vol.3.
[10] M. Magnusson, "The Three-Dimensional Normal-Distributions Transform --- an Efficient Representation for Registration, Surface Analysis, and Loop Detection," 2009.
[11] T. Stoyanov, M. Magnusson, H. Andreasson, and A. J. Lilienthal, "Fast and accurate scan registration through minimization of the distance between compact 3D NDT representations," The International Journal of Robotics Research, vol. 31, no. 12, pp. 1377-1393, 2012/10/01 2012, doi: 10.1177/0278364912460895.
[12] S. Z. Ahmed, V. B. Saputra, S. Verma, K. Zhang, and A. H. Adiwahono, "Sparse-3D Lidar Outdoor Map-Based Autonomous Vehicle Localization," in 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 3-8 Nov. 2019 2019, pp. 1614-1619, doi: 10.1109/IROS40897.2019.8967596.
[13] J. Zhao, S. Huang, and L. Zhao, Constrained Gaussian Mixture Models Based Scan Matching Method. 2018.
[14] R. W. Wolcott and R. M. Eustice, "Fast LIDAR localization using multiresolution Gaussian mixture maps," in 2015 IEEE International Conference on Robotics and Automation (ICRA), 26-30 May 2015 2015, pp. 2814-2821, doi: 10.1109/ICRA.2015.7139582.
[15] R. E. KALMAN, "A New Approach to Linear Filtering and Prediction Problems," 1960.
[16] E. S. Masaru Hoshiya "Structural Identification by Extended Kalman Filter ".
[17] S. J. Julier and J. K. Uhlmann, "Unscented filtering and nonlinear estimation," Proceedings of the IEEE, vol. 92, no. 3, pp. 401-422, 2004, doi: 10.1109/JPROC.2003.823141.
[18] Z. Long, X. Zhang, X. Peng, and G. Yang, "An Improved Adaptive Extended Kalman Filter Used for Target Tracking," in 2019 Chinese Automation Congress (CAC), 22-24 Nov. 2019 2019, pp. 1017-1022, doi: 10.1109/CAC48633.2019.8996637.
[19] W. Gao, J. Li, Y. Zhang, G. Wang, and X. Sun, "Improved innovation-based adaptive estimation for measurement noise uncertainty in SINS/GNSS integration system," in 2017 Forum on Cooperative Positioning and Service (CPGPS）, 19-21 May 2017 2017, pp. 22-28, doi: 10.1109/CPGPS.2017.8075091.
[20] O. S. Chernikova, "An Adaptive Unscented Kalman Filter Approach for State Estimation of Nonlinear Continuous-Discrete System," in 2018 XIV International Scientific-Technical Conference on Actual Problems of Electronics Instrument Engineering (APEIE), 2-6 Oct. 2018 2018, pp. 37-40, doi: 10.1109/APEIE.2018.8545564.
[21] P. Biber, S. Fleck, and W. Straßer, A Probabilistic Framework for Robust and Accurate Matching of Point Clouds. 2004, pp. 480-487.
[22] D. Ralph, "Global Convergence of Damped Newton's Method for Nonsmooth Equations via the Path Search," Mathematics of Operations Research, vol. 19, no. 2, pp. 352-389, 1994. [Online]. Available: http://www.jstor.org/stable/3690225.
[23] M. Magnusson, A. Lilienthal, and T. Duckett, "Scan Registration for Autonomous Mining Vehicles Using 3D-NDT," Journal of Field Robotics, vol. 24, pp. 803-827, 10/01 2007, doi: 10.1002/rob.20204.
[24] D. Arthur and S. Vassilvitskii, K-Means++: The Advantages of Careful Seeding. 2007, pp. 1027-1035.
[25] P. Rousseeuw, "Rousseeuw, P.J.: Silhouettes: A Graphical Aid to the Interpretation and Validation of Cluster Analysis. Comput. Appl. Math. 20, 53-65," Journal of Computational and Applied Mathematics, vol. 20, pp. 53-65, 11/01 1987, doi: 10.1016/0377-0427(87)90125-7.
[26] I. Hashlamon and K. Erbatur, "An improved real-time adaptive Kalman -lter with recursive noise covariance updating rules," Turkish Journal of Electrical Engineering and Computer Sciences, 12/01 2013, doi: 10.3906/elk-1309-60.
[27] "Stencil 2--KAARTA." https://www.kaarta.com/products/stencil-2-for-rapid-long-range-mobile-mapping/ (accessed.
[28] Y. Wang, T. Shi, P. Yun, L. Tai, and M. Liu, PointSeg: Real-Time Semantic Segmentation Based on 3D LiDAR Point Cloud. 2018.