使用分割特徵點偵測與二進制方向直方圖特徵之雷射即時定位與建圖演算法

Laser-Based SLAM Using Segmenting Keypoint Detection and B-SHOT Feature

陳定楷(Ting-Kai Chen)

指導教授 ： 連豊力

國立臺灣大學/電機資訊學院/電機工程學研究所/碩士(2018年)

https://doi.org/10.6342/NTU201803951

摘要

同步定位與建圖是自駕車研究中重要的一個課題，環境的資訊會藉由各種感測器收集並運算來找到對自駕車與環境皆一致的狀態。同步地位與建圖的研究使用了多種類型的感測器，包含相機和光達，光達可以量測精準的深度資訊，但資訊密度小於相機影像。
基於光達的同步定位與建圖演算法主要有兩種:直接法以及切割點雲後擬和模型。直接法會先選取感興趣的點來縮小點雲的數量，例如邊緣或角落的點，接著使用迭代最近點演算法或卡爾曼型濾波器來估測幀與幀的轉換。即便直接法可以被使用在任何情境中，其估測的好壞卻很難評價。相對於直接使用原始的點雲，擬和模型法會先將點雲切割成數個子集合，再將子集合擬和成預先設定的模型，最後再使用這些模型完成幀與幀之間的轉換估測。然而當環境中缺乏預先定義模型相關的物體時，擬和模型法就會失效。
在這篇論文中，僅使用光達並且基於特徵的同步定位與建圖演算法被提出，靈感來自 ORB 同步定位與建圖演算法。在此演算法中，諸如地板、阻擋點之不必要的點會先被點雲預處理程序排除，接著關鍵點會基於分割比例被挑選，並且使用二進制方向直方圖特徵描述子描述選取的關鍵點，最後幀到地圖的轉換會藉由特徵描述子的配對、對位以及迭代最近點演算法微調來估測。實驗結果顯示此演算法在較具結構性的工業技術研究院資料集中可以得到一致的估測結果。

關鍵字

同步定位與建圖 ； 關鍵點偵測 ； 特徵描述子 ； 二進制方向直方圖特徵

並列摘要

Simultaneous localization and mapping is a basic and essential part of the autonomous driving research. Environment information gathered from sensors is computed and derives a consistent state of both self-driving car and the environment. Many types of sensor have been utilized in SLAM research, including camera and LiDAR. LiDAR can provide precise depth information, but suffers from the sparsity compared to camera images.
Two main methods have been used in LiDAR-based SLAM: direct method and modeling after segmentation. Direct method first extracts interesting points, such as edge points or corner points, to reduce the point cloud size. ICP or Kalman-based filter are then applied to estimate the transformation from frame to frame. Although this method can be adopted in every scenario, the quality of estimation is hard to evaluate. Instead of directly using original point cloud, model-based method first segment point cloud into subsets, and then models each subset with a defined model. Finally, frame-to-frame transformation is estimated from models. However, the model-based method is prone to the environment which has less defined models.
In this thesis, a feature-based SLAM algorithm, which is inspired from ORB-SLAM, is proposed on only LiDAR data. In the proposed algorithm, unnecessary points, such as ground points and occluded edge points, are removed by point cloud preprocessing module. Next, the keypoints are selected according to their segment ratio and encoded by B-SHOT feature descriptor. Frame-to-local-map transformation is then estimated based on the B-SHOT feature and refined by iterative closest point algorithm. The experimental results show that the estimated result of the proposed algorithm is consistent in the structural scenarios of ITRI dataset.

並列關鍵字

Simultaneous localization and mapping (SLAM) ； keypoint detection ； feature descriptor ； B-SHOT feature

參考文獻

[1: Arun et al. 1987]

K. S. Arun, T. S. Huang and S. D. Blostein, “Least-Squares Fitting of Two 3-D Point Sets,” in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. PAMI-9, no. 5, pp. 698-700, Sept. 1987.

[2: Atia et al. 2015]

M. M. Atia, S. Liu, H. Nematallah, T. B. Karamat and A. Noureldin, “Integrated Indoor Navigation System for Ground Vehicles with Automatic 3-D Alignment and Position Initialization,” in IEEE Transactions on Vehicular Technology, vol. 64, no. 4, pp. 1279-1292, Apr. 2015.

[3: Besl & McKay 1992]