Navigation and Mapping in Forest Environment Using Sparse Point Clouds

Abstract

Odometry during forest operations is demanding, involving limited field of vision (FOV), back-and-forth work cycle movements, and occasional close obstacles, which create problems for state-of-the-art systems. We propose a two-phase on-board process, where tree stem registration produces a sparse point cloud (PC) which is then used for simultaneous location and mapping (SLAM). A field test was carried out using a harvester with a laser scanner and a global navigation satellite system (GNSS) performing forest thinning over a 520 m strip route. Two SLAM methods are used: The proposed sparse SLAM (sSLAM) and a standard method, LeGO-LOAM (LLOAM). A generic SLAM post-processing method is presented, which improves the odometric accuracy with a small additional processing cost. The sSLAM method uses only tree stem centers, reducing the allocated memory to approximately 1% of the total PC size. Odometry and mapping comparisons between sSLAM and LLOAM are presented. Both methods show 85% agreement in registration within 15 m of the strip road and odometric accuracy of 0.5 m per 100 m. Accuracy is evaluated by comparing the harvester location derived through odometry to locations collected by a GNSS receiver mounted on the harvester.

Unimodal and Multimodal Perception for Forest Management: Review and Dataset

“…They concluded that the addition of LiDAR contributed to an improvement of 38% compared to the traditional approach of only using GNSS+IMU. In [79], the authors proposed a SLAM method called sparse SLAM (sSLAM) whose main application is in forests and for sparse point clouds. They tested their method on the field with a LiDAR and a GNSS-mounted on a harvester and compared their method with LeGO-LOAM.…”

3D Point Clouds in Forest Remote Sensing

“…Among these, only one paper exclusively uses passive RS data [21], while 29 papers use at least one LiDAR dataset in the analysis [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19][20][22][23][24][25][26][27][28][29][30]. Ten papers exclusively use airborne laser scanning (ALS) data [4,6,7,10,11,13,18,23,26,27], nine papers exclusively use terrestrial laser scanning (TLS) data in the analysis [3,9,15,16,20,22,24,25,30], two papers exclusively use mobile laser scanning (MLS) data …”

“…Finally, five papers use combined active and passive remote sensing data sets [2,14,17,19,28]. Regarding the scale of the analysis, 18 of the studies perform individual tree level (ITL) analysis [1][2][3][8][9][10][11][12][14][15][16]19,20,[23][24][25][26]30], eight papers report stand level (SL) analysis [6,7,17,18,21,22,27,29] and four report a combination of ITL and SL [4,5,13,28]. Tree position, diameter at breast height (DBH) and individual tree height (h) are the most common variables of interest, analyzed in nine, six and six papers, respectively, while the most commonly used methods are 3D reconstruction, point filtering and statistical modelling, which are used in eight, five and five papers, respectively (see Table 1).…”

“…The SM include 30 published manuscripts (see Table 1): 28 research papers [1,2,[4][5][6][7][8][9][10][11][12][13][14][15][17][18][19][20][21][22][23][24][25][26][27][28][29][30], one letter [16] and one technical note [3]. Among these, only one paper exclusively uses passive RS data [21], while 29 papers use at least one LiDAR dataset in the analysis [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19]…”

“…Cont. SM include 30 published manuscripts (see Table1): 28 research papers[1,2,[4][5][6][7][8][9][10][11][12][13][14][15][17][18][19][20][21][22][23][24][25][26][27][28][29][30], one letter…”