摘要： Over recent years, light detection and ranging (lidar) sensor technology has rapidly evolved and miniaturized. The reduced sensor size and weight have opened more doors for lidar sensors to be carried onboard unmanned aircraft systems (UASs). Compared with traditional airborne lidar mapping, UAS platforms offer more flexibility in terms of flight design and data collection, rapid response capabilities, and potentially cost at local mapping scales. UAS-based lidar studies have primarily been focused on monitoring vegetation structure, simultaneous localization and mapping (SLAM) and so forth. A comparison between UAS and terrestrial laser scanning (TLS)-derived plant height for crop monitoring was made in [2]. Descriptive statistics derived from polygon grids were analyzed and a correlation R2 = 0.91 was found in plant height derived from both methods. In [3], a lidar-based perception and guidance system was built on a helicopter to perform obstacle detection and avoidance, terrain following, and close-range inspection, and a high success rate was claimed by the authors. Structure-from-Motion (SfM) / Multi-View stereo (MVS) photogrammetry represents an alternative to airborne lidar to derive 3-D point cloud data. It relies on adequate image overlap to extract key point correspondence and collinearity to reconstruct the 3-D scene. This single return solution is sometimes susceptible to false parallax induced from moving vegetation between overlapping images (for instance due to wind) and to poor feature correspondence in areas where the image texture is highly uniform [4], [5]. One advantage of using lidar is it uses pulsed ranging technique and many lidar systems provide multiple return detection capability. This multi-return capability has enabled lidar to be widely applied to forestry inventory surveys among other applications because it allows for canopy and below canopy measurement. A mini-UAS-borne lidar system was built in [6], and its applicability for fine-scale mapping was validated in terms of tree height estimation, pole detection, road extraction, and digital terrain model refinement. A more recent study in [7] developed a lidar-hyperspectral image fusion method in treated and controlled forests with varying tree density and canopy cover to classify vegetation and measure 3D structure. It was claimed that the fusion method performed better than either data type alone at the study site in the southwestern USA. In this paper, initial results on the testing and evaluation of a single-rotary UAS integrated with a long-range, multi-return lidar sensor is presented. Testing was performed at an airfield in South Texas, USA. The study primarily focuses on: 1) description of the platform and enabling technology (i.e., lidar/IMU/GPS) of the fully integrated UAS solution, 2) sensor calibration and initialization (e.g., boresight calibration and IMU initialization), and 3) description of the geospatial surveying, data processing and analysis.