利用2D LiDAR進行室內平面圖之建構
|
| 摘要 近年來在不同領域中的研究，對精準定位的需求日漸增加，故同步定位與製圖(Simultaneous Localization and Mapping, SLAM)的應用越來越廣泛，其中視覺型SLAM更是主要應用的技術。與其他昂貴儀器相比，一般攝影機的成本較低，故基於影像的SLAM的應用也相對較廣泛。然而，距離感測器的成本也較其他儀器低，且其效能也較視覺型SLAM高，故本研究擬用成本較低的二維雷射探測與測距(Light Detection and Ranging, LiDAR)儀器，以實行SLAM。本研究的成果顯示，2D LiDAR SLAM的精度在可接受之範圍內，且能完整呈現室內環境的各種特徵，如角落與邊緣等，確立了2D LiDAR SLAM的可行性，以及其應用潛力。
關鍵字：SLAM、LiDAR、點雲處理、測量。 ABSTRACT Due to research in different fields in recent years, the demand for precise positioning has increased day by day, so Simultaneous Localization and Mapping has become more and more widely used, among which visual SLAM is the main applied technology. Compared with other expensive instruments, the cost of general cameras is low, so the application of image-based SLAM is relatively widespread. However, the cost of distance sensors is also lower than other instruments, and its performance is also higher than visual SLAM. Therefore, this study plans to use lower-cost two-dimensional laser detection and LiDAR instruments to implement SLAM. The subsequent results will be Compare with high-precision LiDAR measurement results with higher accuracy and cost to verify the accuracy and feasibility of 2D LiDAR SLAM. The results of this study show that the accuracy of 2D LiDAR SLAM is within an acceptable range and can completely represent various features of the indoor environment, such as corners and edges, etc., establishing the feasibility of 2D LiDAR SLAM and its application potential. KEYWORDS: SLAM; 2D LiDAR; Point Cloud Processing; Measurement. |