SLAM-Aided Stem Mapping for Forest Inventory with Small-Footprint Mobile LiDAR

Abstract

Accurately retrieving tree stem location distributions is a basic requirement for biomass estimation of forest inventory. Combining Inertial Measurement Units (IMU) with Global Navigation Satellite Systems (GNSS) is a commonly used positioning strategy in most Mobile Laser Scanning (MLS) systems for accurate forest mapping. Coupled with a tactical or consumer grade IMU, GNSS offers a satisfactory solution in open forest environments, for which positioning accuracy better than one decimeter can be achieved. However, for such MLS systems, positioning in a mature and dense forest is still a challenging task because of the loss of GNSS signals attenuated by thick canopy. Most often laser scanning sensors in MLS systems are used for mapping and modelling rather than positioning. In this paper, we investigate a Simultaneous Localization and Mapping (SLAM)-aided positioning solution with point clouds collected by a small-footprint LiDAR. Based on the field test data, we evaluate the potential of SLAM positioning and mapping in forest inventories. The results show that the positioning accuracy in the selected test field is improved by 38% compared to that of a traditional tactical grade IMU + GNSS positioning system in a mature forest environment and, as a result, we are able to produce a unambiguous tree distribution map.

An Integrated GNSS/INS/LiDAR-SLAM Positioning Method for Highly Accurate Forest Stem Mapping

“…There are rich and obvious features like lines and corners in indoor environments, and SLAM can acquire satisfactory solutions using feature matching methods. The features in forests are not continuous or stable, due to the density changes and the uneven terrain found in forests; moreover, GNSS satellites can be observed occasionally, particularly in open forest areas where SLAM cannot succeed in scan matching, leading to drift and gross errors [16]. The main cause of these errors is incorrect estimation of the heading angle, because there are not enough features for scan matching.…”

“…The detailed process and performance of IMLE-SLAM can be found in [33]. Moreover, some premises that are the same as in [16] have been taken into account, which are that multiple segmental laser scans from 3D space are projected onto 2D stem profiles to generate the stem map according to the accurate roll and pitch estimates from GNSS/INS by assuming that most mature stems have straight and circular stems. Because the main type of tree in the field test is upright pine trees, the 2D projection assumption is reasonable in our experiment.…”

“…One excellent grid-map based method is Maximum Likelihood Estimation (MLE) [33]. Tang et al [16] employ the Improved Maximum Likelihood Estimation (IMLE) method for positioning in forests and obtain promising results in mature forests. In applying this method, a grid map is built based on probabilistic and feature uncertainty models, and then a brute-force search matching method is applied for scan-to-map matching to find the optimal rigid-body transformation.…”

“…2017, 9, 3 3 of 16 The remainder of this paper is organized as follows: Section 2 describes the materials and methods utilized in this research; Section 3 introduces the field tests; the experimental results are discussed in Section 4; and Section 5 draws conclusions.…”

“…Most studies investigating SLAM are mainly for indoor environments, where GNSS is denied and which are full of features. Similarly, the application of SLAM to forests has been investigated in several studies [16][17][18][19][20][21][22][23][24]. In addition, many of them have studied the feasibility of using SLAM technology for MLS in forests [17][18][19].…”