Tree-SLAM: semantic object SLAM for efficient mapping of individual trees in orchards

David Rapado-Rincon; Gert Kootstra

REVIEW 3 major objections 5 minor 18 references

Tree-SLAM: semantic object SLAM for efficient mapping of individual trees in orchards

T0 review · 3 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Tree-SLAM is a semantic SLAM system that maps individual orchard trees with a geo-localization error as low as 18 cm by fusing odometry, GPS, and trunk observations in a factor graph.

desk verdict A useful orchard-tree SLAM system with a genuinely novel cascade association step, but the evaluation section needs to state what was matched against what before the 18 cm claim is citable. read the letter →

arxiv 2507.12093 v1 pith:VDZBE2RG submitted 2025-07-16 cs.RO

David Rapado-Rincon , Gert Kootstra This is my paper · ORCID

classification cs.RO

keywords semanticSLAMorchardmappingfactorgraphtreetrunkdetectioninstancesegmentationdataassociationGPS-deniedlocalizationprecisionagriculture

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper sets out to show that a semantic SLAM system built around tree trunks as landmarks can produce maps of individual trees in orchards even when GPS is unreliable under the canopy. The key claim is that fusing wheel odometry, noisy GPS, and repeated RGB-D observations of trunks in a factor graph yields geo-localization errors as low as 18 cm, less than 20% of the tree planting distance. If true, this gives an agricultural robot the tree-level map it needs for targeted spraying, monitoring, or harvesting without relying on clear GPS sky view. The authors test the system on apple and pear orchards in leafless and leafed seasons and compare it against a baseline that clusters raw trunk detections.

What carries the argument

The central mechanism is a factor graph whose variables are the robot poses and the 2D positions of individual tree landmarks. It is fed by three kinds of measurements: odometry factors linking consecutive poses, GPS pose factors when available, range-bearing factors from each trunk observation, and distance factors between every pair of trunks seen in the same frame. The distance factors are the unusual piece: because the RGB-D camera can measure the relative distance between two visible trunks accurately, these constraints help the graph resist GPS noise. Two supporting mechanisms carry the landmark quality: PCA on the masked trunk point cloud estimates the trunk center by correcting the visible-surface centroid outward by half the estimated width, and a cascade-graph association propagates identities outward from high-confidence matches using the spatial layout of the row.

What would settle it

Survey the true ground-truth trunk centers of apple trees with protective nets and tall grass, then compare them with the PCA-corrected estimates the pipeline produces; if the corrected centroids are systematically displaced toward the camera or away from the row line, the correction step is biased in exactly the conditions where the paper reports its weakest results.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that an orchard's repetitive structure, normally a liability for feature-based SLAM, can be turned into an asset: tree trunks are detected by an instance segmentation model, their centers estimated from partial point clouds, and their identities maintained by a cascade association that uses the known row layout. Once trunks are recognized as stable landmarks, a factor graph jointly optimizes robot poses and landmark positions using odometry, GPS, and range-bearing trunk observations together with inter-tree distance constraints. The paper reports that in leafless pear rows, 99.0% of ground-truth trees fall within half the planting distance with a mean error of 0.18 m, and that the factor-graph approach raises recall for young apple trees under noisy GPS from 44.6% to 84.6% compared with the clustering baseline. The paper is careful to report where the method does not win: for young apple trees in leafed conditions, the simpler baseline yields a lower mean error, and the paper attributes this to degraded trunk detection from grass occlusion and protective nets.

Load-bearing premise

The whole map accuracy rests on the assumption that a one-sided point cloud of a trunk is a partial cylinder, so that shifting the measured surface centroid outward by half the estimated width gives the true trunk center; when trunks are irregular, netted, or partly hidden by grass, that shift is biased and the bias flows into every landmark position.

Editorial extensions

If this is right

Orchard robots can get tree-level maps with mean geo-localization error as low as 0.18 m, below 20% of the planting distance, under canopy where GPS alone is unreliable.
The factor-graph approach recovers trees that a detection-clustering baseline misses when GPS is noisy: recall for young apple trees in leafless conditions rises from 44.6% to 84.6%.
The cascade-graph association is load-bearing for identity maintenance: removing it cuts recall for leafed apple trees from 0.75 to 0.42 and raises mean error from 0.34 m to 0.47 m.
PCA-based trunk center estimation is load-bearing for mature trunks: replacing it with simple averaging drops pear leafless precision from 0.99 to 0.79.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

An extension the paper leaves implicit is to estimate the planting distance online from the detected trunk spacing instead of taking it as known input; that would let the association radius and clustering epsilon adapt to irregular or mixed orchards.
The distance factors between co-visible trunks act as an implicit row-structure prior; a testable variant would strengthen or weaken those factors based on detection confidence, which could recover some of the leafed-apple cases where the baseline won.
Since the method is designed for GPS degradation rather than total loss, an informative extreme test would run the factor graph with GPS completely disabled and measure how far the odometry-plus-landmark constraints drift over a full row loop.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Quick take: this is a solid systems contribution. What is actually new is the cascade-graph data association that propagates matches from high-confidence SORT associations outward through neighboring tree positions, plus a distance factor between trees seen in the same frame. That combination is not in the orchard SLAM papers I know, and the ablations show both pieces matter: dropping the graph stage drops apple-leafed recall from 0.75 to 0.42, and dropping PCA hurts pear precision. There is also a real dataset, more than 8,000 annotated images across seasons. That deserves credit.

The evaluation is where I hesitate. The headline 18 cm / less than 20% of planting distance is only the pear-leafless number; the same table has apple-leafless at 0.40 m and pear-leafed at 0.28 m. Leading with the best case is fine, but the abstract and conclusion state the <20% claim as if it were general. The evaluation is also thin: one apple row, one pear row per season, no repeated runs, no uncertainty intervals. For a system whose whole point is robustness under GPS degradation, I would want to see variance across runs.

The bigger issue is the ground-truth denominator. Table 1 says 32 of 135 pear trees were measured with RTK, yet Table 3 reports pear recall near 99%. If precision and recall are computed only against the 32 measured trees, and predictions outside that subset are ignored, the numbers are valid for that subset, but that restriction is never stated in Section 3.2. If the unmeasured predictions are counted as false positives, precision near 0.99 is impossible with 135 predicted trees. The sentence "our algorithm does not predict trees that do not actually exist" suggests the evaluation is restricted to a measured subset, but the manuscript should say so. Until this is clarified, every quantitative result sits under a cloud, and this is more load-bearing than the PCA bias worry.

Other soft spots are minor by comparison: factor graph covariances and the cascade radius r are not reported, and the DBSCAN epsilon was tuned on the same data. Neither sinks the method.

Who this is for: people building orchard mapping or row-crop semantic SLAM systems, and precision-agriculture groups wanting a practical baseline. It deserves a serious referee, but the referee should send it back for a rewritten evaluation section with a clear matching protocol, repeated runs, and error bars.

Referee Report

3 major / 5 minor

Summary. The paper presents Tree-SLAM, a semantic SLAM system for orchard tree mapping. It detects trunks with YOLOv8 instance segmentation, estimates trunk centers from masked RGB-D point clouds via PCA, re-identifies trunks with a cascade-graph data association method, and fuses odometry, GPS, and range-bearing trunk observations in a GTSAM factor graph. The system also introduces an inter-tree distance factor between co-observed trunks. It is evaluated on four orchard rows (pear and apple, leafless and leafed) against a DBSCAN-clustering baseline, with RTK-GPS ground truth for a subset of trees. The authors report a mean true-positive localization error as low as 0.18 m for leafless pear trees, which is less than 20% of the 1.1 m planting distance, and ablation studies showing the contribution of PCA-based center estimation and graph-based data association.

Significance. If the reported accuracy holds, Tree-SLAM is a useful practical contribution: it addresses the repetitive-scene and GPS-degradation problems that limit feature-based SLAM in orchards, and it provides a large annotated trunk segmentation dataset plus a comparison against a clustering baseline. The use of independent RTK-GPS ground truth is a strength, and the ablation studies give evidence that the proposed components matter. However, the central accuracy and precision/recall claims are currently under-specified because the ground-truth tree subset used for matching is ambiguous, and the evaluation is thin (one row per condition, no repeated runs, no uncertainty intervals). The headline '<20% of planting distance' is only demonstrated in one of four row/season conditions.

major comments (3)

[Section 3.2, Tables 1 and 3] Table 1 reports that each pear row has only 32 of 135 trees measured with RTK ground truth, yet Table 3 reports pear-leafless recall of 99.0% and Table 4 reports pear-leafless precision of 0.99. If Tree-SLAM produces predictions for the full 135-tree row, precision cannot reach 0.99 when only 32 ground-truth trees are available, because any prediction not matched to a measured tree would be a false positive under the definition given in Section 3.2. If, instead, predictions are restricted to the measured subset before matching, that restriction is never stated. The sentence in Section 3.2 that 'Tree-SLAM only predicts the trees that were actually measured' appears to presuppose such a restriction, but the algorithm has no access to the measured subset. Please specify exactly how the predicted set and ground-truth set are constructed for each row, how unmatched predictions outside the measured subset are counted, and how the 103 unmeasured pear trees per row are treated; re-report the metrics if the current numbers use a different denominator than stated.
[Abstract; Section 3.2, Table 3] The headline claim that the map error is 'less than 20% of the planting distance' is only demonstrated in the pear-leafless condition (0.18 m / 1.1 m ≈ 16%). The other conditions exceed 20%: pear leafed 0.28 m ≈ 25%, apple leafless 0.40 m ≈ 33%, and apple leafed 0.32 m ≈ 27% of the respective planting distances. The abstract and conclusion should either state that the 20% figure is a best-case result or explain why the target is not met in the other three conditions; the current wording overstates the empirical support.
[Section 3.2, Table 3] Each row/season condition is evaluated from a single recording, and all metrics are point estimates without uncertainty intervals. Given that GPS quality, detection noise, and association outcomes vary between runs, the abstract's 'robustness' claim and the Discussion's general conclusions would require repeated runs or a quantitative uncertainty analysis (e.g., multiple traversals, bootstrap resampling, or per-run variance). Please add such evidence or qualify the robustness statements accordingly.

minor comments (5)

[Equation (8)] Equation (8) is typeset incorrectly in the version I reviewed: the bearing-residual term and fraction formatting are garbled, and the Greek letters do not render properly. Please regenerate the equation and verify all subscripts, superscripts, and denominators.
[Table 4] The caption and table body do not label which component is removed in each row. A single check mark under the 'PCA' or 'Graph' column is ambiguous; please add explicit row labels such as 'w/o PCA' and 'w/o graph association'.
[Section 2.3.2, Eq. (2)] The radius r used in the cascade-graph neighborhood search is never given a numerical value or a selection procedure. Please report how this radius was chosen, since it is a free parameter in the data association stage.
[Section 2.2] The sensor name 'Topcom' appears to be a typo for 'Topcon'; please correct it.
[General] The manuscript does not include a data or code availability statement. Given the emphasis on a new large dataset, please add a statement or explain any restrictions on sharing the data and implementation.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the central mapping claim is evaluated against independent RTK-GPS ground truth, and no prediction reduces by construction to a fitted input.

full rationale

Tree-SLAM is an empirical systems paper. Its central claim—that the factor graph produces tree maps with an 18 cm true-positive mean error—is evaluated against RTK-GPS tree positions measured independently (Table 1) and not used as factors in the optimization. The cost function (Eq. 8) contains odometry, GPS pose, range-bearing, and inter-tree distance factors, but no ground-truth tree-position factors; the ground truth is only used after mapping to compute precision, recall, and error. The PCA-based trunk-center refinement is an algorithmic choice whose effect is tested in ablation (Table 4) against the same external ground truth, not against the quantity it estimates. The DBSCAN epsilon is set from the known planting distance (60%) and the evaluation gate is PD/2; both are known geometry rather than fitted outputs. The self-citations ([1], [2], [4], [18]) appear in background and future-work statements and do not carry the load-bearing accuracy claim. The acknowledged limitation of degraded performance for young apple trees in leafed conditions is a performance statement, not a circular derivation. The skeptic concern about the pear GT subset (32 of 135) is an evaluation-denominator or validity question about how unmatched predictions were handled, but it does not exhibit any equation or construction in which a predicted quantity equals its input; therefore, per the hard rules, it does not count as circularity. No step in the paper's derivation chain fits the seven enumerated patterns.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The method introduces no new physical entities; the free parameters are all engineering choices in the SLAM pipeline. The most consequential are the factor graph covariances, which are not reported at all.

free parameters (5)

DBSCAN epsilon (trunk filtering) = 60% of known planting distance
Used in Section 2.3.1 to cluster detections; described as determined empirically to separate trees while filtering noise. Not justified by theory or sensitivity analysis.
YOLO confidence threshold = 0.1
Low threshold adopted to maximize recall (Section 2.3.1); affects false positive rate and downstream association.
Factor graph noise covariances (Sigma_odom, Sigma_GPS, sigma_r, sigma_phi, sigma_delta) = not specified
No values are reported; these weights determine the balance between sensors and directly affect optimized positions. A practitioner cannot reproduce without choosing them.
Association radius r (cascade graph, Eq. 2) = not specified
Radius for neighbor-based association in Section 2.3.2 is not given; presumably tied to planting distance but not stated.
DBSCAN epsilon for baseline = 0.5 m
Tuned on the same evaluation data (Figure 6) to maximize F1; this is a comparison baseline, but it is a fitted parameter.

assumptions (4)

domain assumption The orchard ground is flat enough to justify a 2D top-down representation.
Section 2.3.3: 'This is a reasonable assumption because the ground is mostly flat and the trees are planted in rows.' If the ground undulates, discarding Z biases landmark positions.
domain assumption Tree trunks are static and their 2D positions are fixed over the data collection.
Tracking assumes each track's 2D world location remains static (Section 2.3.2). If trunks move (e.g., flexible young trees), landmarks are inconsistent.
domain assumption The planting distance is known and uniform for the row being mapped.
DBSCAN epsilon is set to 60% of planting distance (Section 2.3.1) and the cascade graph uses spatial neighborhood relations (Eq. 2). Variable spacing would require re-tuning.
standard math Measurement noises are Gaussian and independent (standard factor graph assumption).
The cost function (Eq. 8) uses squared Mahalanobis terms; if GPS errors under canopy have heavy tails or bias, the optimization will be suboptimal.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Tree-SLAM: semantic object SLAM for efficient mapping of individual trees in orchards." pith.science (2026). https://pith.science/paper/VDZBE2RG

@misc{pith2026250712093,
  author       = {Pith},
  title        = {Pith review of: Tree-SLAM: semantic object SLAM for efficient mapping of individual trees in orchards},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VDZBE2RG}},
  note         = {Machine review of arXiv:2507.12093}
}

read the original abstract

Accurate mapping of individual trees is an important component for precision agriculture in orchards, as it allows autonomous robots to perform tasks like targeted operations or individual tree monitoring. However, creating these maps is challenging because GPS signals are often unreliable under dense tree canopies. Furthermore, standard Simultaneous Localization and Mapping (SLAM) approaches struggle in orchards because the repetitive appearance of trees can confuse the system, leading to mapping errors. To address this, we introduce Tree-SLAM, a semantic SLAM approach tailored for creating maps of individual trees in orchards. Utilizing RGB-D images, our method detects tree trunks with an instance segmentation model, estimates their location and re-identifies them using a cascade-graph-based data association algorithm. These re-identified trunks serve as landmarks in a factor graph framework that integrates noisy GPS signals, odometry, and trunk observations. The system produces maps of individual trees with a geo-localization error as low as 18 cm, which is less than 20\% of the planting distance. The proposed method was validated on diverse datasets from apple and pear orchards across different seasons, demonstrating high mapping accuracy and robustness in scenarios with unreliable GPS signals.

Figures

Figures reproduced from arXiv: 2507.12093 by the authors.

**Figure 2.** Examples of the pear (top) and apple (bottom) trees of the mapping dataset [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗

**Figure 3.** Overview of the Tree-SLAM pipeline, illustrating the main processing stages [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗

Figures from the paper (5 more)

**Figure 4.** Figure 4: The trunk association and tracking algorithm. First a SORT-like association is [PITH_FULL_IMAGE:figures/full_fig_p010_4.png]

**Figure 5.** Figure 5: An illustration of the factor graph used in Tree-SLAM. The graph represents the [PITH_FULL_IMAGE:figures/full_fig_p012_5.png]

**Figure 6.** Figure 6: Selection of the optimal epsilon parameter for the DBSCAN-based baseline [PITH_FULL_IMAGE:figures/full_fig_p016_6.png]

**Figure 7.** Figure 7: Comparison of mapping performance between Tree-SLAM and the baseline [PITH_FULL_IMAGE:figures/full_fig_p017_7.png]

**Figure 8.** Figure 8: Satellite view containing ground truth and predicted tree positions for the pear [PITH_FULL_IMAGE:figures/full_fig_p018_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

18 extracted references · 13 canonical work pages

[1]

Kootstra, A

G. Kootstra, A. Bender, T. Perez, E. J. van Henten, Robotics in Agriculture, in: M. H. Ang, O. Khatib, B. Siciliano (Eds.), Ency- clopedia of Robotics, Springer, Berlin, Heidelberg, 2020, pp. 1–19. doi:10.1007/978-3-642-41610-1_43-1. URLhttps://doi.org/10.1007/978-3-642-41610-1_43-1

work page doi:10.1007/978-3-642-41610-1_43-1 2020
[2]

Kootstra, X

G. Kootstra, X. Wang, P. M. Blok, J. Hemming, E. van Henten, Se- lective Harvesting Robotics: Current Research, Trends, and Future Di- rections, Current Robotics Reports 2 (1) (2021) 95–104.doi:10.1007/ s43154-020-00034-1. URLhttps://doi.org/10.1007/s43154-020-00034-1

work page doi:10.1007/s43154-020-00034-1 2021
[3]

Sparrow, M

R. Sparrow, M. Howard, Robots in agriculture: prospects, impacts, ethics, and policy, Precision Agriculture 22 (3) (2021) 818–833.doi: 10.1007/s11119-020-09757-9. URLhttps://doi.org/10.1007/s11119-020-09757-9

work page doi:10.1007/s11119-020-09757-9 2021
[4]

Kootstra, Advances in visual perception for agricultural robotics, in: Advances in agri-food robotics, Burleigh Dodds Science Publishing, 2024, pp

G. Kootstra, Advances in visual perception for agricultural robotics, in: Advances in agri-food robotics, Burleigh Dodds Science Publishing, 2024, pp. 3–36

work page 2024
[5]

W. Dong, P. Roy, V. Isler, Semantic mapping for or- chard environments by merging two-sides reconstructions of tree rows, Journal of Field Robotics 37 (1) (2020) 97–121, eprint: https://onlinelibrary.wiley.com/doi/pdf/10.1002/rob.21876. doi:10.1002/rob.21876. URLhttps://onlinelibrary.wiley.com/doi/abs/10.1002/rob. 21876

work page doi:10.1002/rob.21876 2020
[6]

Tiozzo Fasiolo, L

D. Tiozzo Fasiolo, L. Scalera, E. Maset, A. Gasparetto, Towards au- tonomous mapping in agriculture: A review of supportive technologies for ground robotics, Robotics and Autonomous Systems 169 (2023) 104514.doi:10.1016/j.robot.2023.104514. 23 URLhttps://www.sciencedirect.com/science/article/pii/ S0921889023001537

arXiv 2023
[7]

Shalal, T

N. Shalal, T. Low, C. McCarthy, N. Hancock, A review of autonomous navigation systems in agricultural environments, SEAg 2013: Innovative agricultural technologies for a sustainable future (2013)

work page 2013
[8]

Hroob, R

I. Hroob, R. Polvara, S. Molina, G. Cielniak, M. Hanheide, Bench- mark of Visual and 3D Lidar SLAM Systems in Simulation Envi- ronment for Vineyards, in: C. Fox, J. Gao, A. Ghalamzan Esfa- hani, M. Saaj, M. Hanheide, S. Parsons (Eds.), Towards Autonomous Robotic Systems, Springer International Publishing, 2021, pp. 168–177. doi:10.1007/978-3-030-89177-0_17

work page doi:10.1007/978-3-030-89177-0_17 2021

Show all 18 references

[9]

P. Shao, F. Mo, Y. Chen, N. Ding, R. Huang, Monocular Object SLAM using Quadrics and Landmark Reference Map for Outdoor UA V Appli- cations, in: 2021 IEEE International Conference on Real-time Com- puting and Robotics (RCAR), 2021, pp. 1195–1201.doi:10.1109/ RCAR52367.2021.951...

2021
[10]

Nicholson, M

L. Nicholson, M. Milford, N. S¨ underhauf, QuadricSLAM: Dual Quadrics From Object Detections as Landmarks in Object-Oriented SLAM, IEEE Robotics and Automation Letters 4 (1) (2019) 1–8, conference Name: IEEE Robotics and Automation Letters.doi:10.1109/LRA.2018. 2866205

2019 doi
[11]

Macenski, T

S. Macenski, T. Foote, B. Gerkey, C. Lalancette, W. Woodall, Robot Operating System 2: Design, architecture, and uses in the wild, Science Robotics 7 (66) (2022) eabm6074, publisher: American Association for the Advancement of Science.doi:10.1126/scirobotics.abm6074. URLhttps:...

2022 doi
[12]

Rehder, J

J. Rehder, J. Nikolic, T. Schneider, T. Hinzmann, R. Siegwart, Extend- ing kalibr: Calibrating the extrinsics of multiple IMUs and of individual axes, in: 2016 IEEE International Conference on Robotics and Automa- tion (ICRA), 2016, pp. 4304–4311.doi:10.1109/ICRA.2016.7487628....

2016
[13]

Jocher, J

G. Jocher, J. Qiu, A. Chaurasia, Ultralytics YOLO (2023). URLhttps://github.com/ultralytics/ultralytics

2023
[14]

Bewley, Z

A. Bewley, Z. Ge, L. Ott, F. Ramos, B. Upcroft, Simple Online and Real- time Tracking, 2016 IEEE International Conference on Image Processing (ICIP) (2016) 3464–3468ArXiv: 1602.00763.doi:10.1109/ICIP.2016. 7533003. URLhttp://arxiv.org/abs/1602.00763

2016 arXiv
[15]

Dellaert, G

F. Dellaert, G. Contributors, borglab/gtsam (May 2022).doi:10.5281/ zenodo.5794541. URLhttps://github.com/borglab/gtsam)

2022
[16]

Brown, A

J. Brown, A. Paudel, D. Biehler, A. Thompson, M. Karkee, C. Grimm, J. R. Davidson, Tree detection and in-row localization for autonomous precision orchard management, Computers and Electronics in Agricul- ture 227 (2024) 109454.doi:10.1016/j.compag.2024.109454. URLhttps://www....

2024
[17]

Huang, P

P. Huang, P. Huang, Z. Wang, X. Wu, J. Liu, L. Zhu, Deep-Learning- Based Trunk Perception with Depth Estimation and DW A for Robust Navigation of Robotics in Orchards, Agronomy 13 (4) (2023) 1084, number: 4 Publisher: Multidisciplinary Digital Publishing Institute. doi:10.3390...

2023 doi
[18]

A. K. Burusa, J. Scholten, X. Wang, D. Rapado-Rincon, E. J. van Hen- ten, G. Kootstra, Semantics-aware next-best-view planning for efficient search and detection of task-relevant plant parts, Biosystems Engineer- ing 248 (2024) 1–14.doi:10.1016/j.biosystemseng.2024.09.018. 25

2024 doi

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.

[1] [1]

Kootstra, A

G. Kootstra, A. Bender, T. Perez, E. J. van Henten, Robotics in Agriculture, in: M. H. Ang, O. Khatib, B. Siciliano (Eds.), Ency- clopedia of Robotics, Springer, Berlin, Heidelberg, 2020, pp. 1–19. doi:10.1007/978-3-642-41610-1_43-1. URLhttps://doi.org/10.1007/978-3-642-41610-1_43-1

work page doi:10.1007/978-3-642-41610-1_43-1 2020

[2] [2]

Kootstra, X

G. Kootstra, X. Wang, P. M. Blok, J. Hemming, E. van Henten, Se- lective Harvesting Robotics: Current Research, Trends, and Future Di- rections, Current Robotics Reports 2 (1) (2021) 95–104.doi:10.1007/ s43154-020-00034-1. URLhttps://doi.org/10.1007/s43154-020-00034-1

work page doi:10.1007/s43154-020-00034-1 2021

[3] [3]

Sparrow, M

R. Sparrow, M. Howard, Robots in agriculture: prospects, impacts, ethics, and policy, Precision Agriculture 22 (3) (2021) 818–833.doi: 10.1007/s11119-020-09757-9. URLhttps://doi.org/10.1007/s11119-020-09757-9

work page doi:10.1007/s11119-020-09757-9 2021

[4] [4]

Kootstra, Advances in visual perception for agricultural robotics, in: Advances in agri-food robotics, Burleigh Dodds Science Publishing, 2024, pp

G. Kootstra, Advances in visual perception for agricultural robotics, in: Advances in agri-food robotics, Burleigh Dodds Science Publishing, 2024, pp. 3–36

work page 2024

[5] [5]

W. Dong, P. Roy, V. Isler, Semantic mapping for or- chard environments by merging two-sides reconstructions of tree rows, Journal of Field Robotics 37 (1) (2020) 97–121, eprint: https://onlinelibrary.wiley.com/doi/pdf/10.1002/rob.21876. doi:10.1002/rob.21876. URLhttps://onlinelibrary.wiley.com/doi/abs/10.1002/rob. 21876

work page doi:10.1002/rob.21876 2020

[6] [6]

Tiozzo Fasiolo, L

D. Tiozzo Fasiolo, L. Scalera, E. Maset, A. Gasparetto, Towards au- tonomous mapping in agriculture: A review of supportive technologies for ground robotics, Robotics and Autonomous Systems 169 (2023) 104514.doi:10.1016/j.robot.2023.104514. 23 URLhttps://www.sciencedirect.com/science/article/pii/ S0921889023001537

arXiv 2023

[7] [7]

Shalal, T

N. Shalal, T. Low, C. McCarthy, N. Hancock, A review of autonomous navigation systems in agricultural environments, SEAg 2013: Innovative agricultural technologies for a sustainable future (2013)

work page 2013

[8] [8]

Hroob, R

I. Hroob, R. Polvara, S. Molina, G. Cielniak, M. Hanheide, Bench- mark of Visual and 3D Lidar SLAM Systems in Simulation Envi- ronment for Vineyards, in: C. Fox, J. Gao, A. Ghalamzan Esfa- hani, M. Saaj, M. Hanheide, S. Parsons (Eds.), Towards Autonomous Robotic Systems, Springer International Publishing, 2021, pp. 168–177. doi:10.1007/978-3-030-89177-0_17

work page doi:10.1007/978-3-030-89177-0_17 2021