Pith. sign in

REVIEW 4 major objections 5 minor 42 references

Learning Rock Pushability on Rough Planetary Terrain

T0 review · 4 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

Pith's one-line read One exploratory push tells a robot whether a rock on rough terrain is pushable, by predicting the peak force it will feel and comparing that to its own strength.

desk verdict A real problem and a real dataset, but the central pushability claim is unvalidated: movement is never measured, the force threshold is fit on the same data, and the split leaks. read the letter →

arxiv 2505.09833 v2 pith:A5IHNA6M submitted 2025-05-14 cs.RO cs.LG

classification cs.ROcs.LG
keywords rockpushabilityBayesianlinearregressionaffordancelearningforcefeedbackplanetaryterrainpointcloudsegmentationnonprehensilepushingmobilemanipulation
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

This paper proposes replacing obstacle avoidance with obstacle removal on rough planetary terrain. It argues that a robot can decide whether a rock is pushable by predicting, from visual features alone, the peak force its arm will experience during a push, and then checking that predicted force against the robot's physical capabilities. The claim is that a Bayesian linear model trained on a modest dataset of simulated and real push experiments can make this prediction accurately enough to classify rocks as pushable, too large to move, or too small to bother with. If true, fleets of robots could clear paths once instead of detouring around the same rocks repeatedly, improving long-term navigation efficiency in places like lunar or Martian surfaces.

What carries the argument

The central mechanism is a Bayesian linear regressor (implemented with BayesianRidge) that predicts the maximum force-feedback magnitude $f^j_{\mathrm{max}}$ of a push action, conditioned on features $R_i = (V_i, E_i, N_i, g_i)$: obstacle volume, ellipsoid-fit shape error, surface normal set, and a visual likelihood score. The training phase initializes a prior $\mathbf{w} := \mathcal{N}(0, \lambda^{-1} I)$ and updates it with Bayes' rule on observed force feedback; the prediction phase computes the posterior expected peak force and applies the decision rule $\hat{F}_{\mathrm{peak},j} \le \delta \Rightarrow$ Pushable. This probabilistic treatment provides uncertainty quantification and lets the visual preliminary estimate $g_j$ serve as a prior belief, which the proprioceptive interaction then refines.

What would settle it

Run a push experiment where the robot records the force signal and also tracks the rock's displacement (e.g., by differencing the point cloud before and after the push). If a substantial fraction of rocks classified as not pushable (predicted peak force above $\delta$) still move, or rocks classified as pushable fail to move, the peak-force proxy fails. Concretely, across at least 100 pushes on varied slopes and rock sizes, compare the predicted peak force against measured displacement and check whether the decision rule's accuracy is significantly below chance.

Watch

Extended reading notes

Core claim

The paper's central claim is that pushability affordance—whether an obstacle will move when pushed—can be estimated as the probability distribution of the maximum force feedback during a push, conditioned on visually observable features: the rock's volume, its shape (quantified by ellipsoid-fitting residual error), and the surface normals of the ground it rests on. The authors build a pipeline that segments rocks from point clouds using surface-normal cosine similarity and DBSCAN clustering, fits 3D ellipsoids to score shape, estimates surface normals relative to the robot, and feeds these features into a Bayesian linear regressor trained on force-feedback labels from 90 simulated and real push experiments. The regressor predicts the peak force magnitude $\hat{F}_{\mathrm{peak},j}$, and Algorithm 1 classifies the rock as pushable when that prediction is at or below the robot's capability $\delta$. The paper reports that the model's predictions are strongly driven by rock volume and surface normal, and insensitive to the rock's position in the point cloud, implying the robot can assess pushability at any timestep before committing to a push.

Load-bearing premise

The system never measures whether a rock actually moves; it assumes the peak force felt during a push is a reliable indicator of pushability, so a rock that moves despite high resistance, or resists lightly yet stays put, would break the classification.

Editorial extensions

If this is right

  • If the prediction is accurate, a robot can clear pushable rocks from a path instead of avoiding them, reducing the detour time for itself and for all subsequent agents traversing the same route.
  • The visual preliminary filter can pre-sort obstacles into Static (too large), Pushable, or Override (too small to bother), so the robot only wastes a push on rocks that have a real chance of moving.
  • Because the model is Bayesian, the robot gets a distribution over peak force rather than a single number, enabling risk-aware decisions: it can decline to push when the uncertainty crosses a safety threshold.
  • The finding that surface normals and volume dominate the prediction, while rock position does not, suggests the robot can evaluate pushability from any viewpoint, which could simplify the approach strategy.
  • The framework is intended to generalize across simulated and real rough terrain, with the real-world validation showing accurate volume and slope estimation even when the test ramp slope (8 degrees) differs from the model's estimate (5 degrees).

Reading between the lines

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

  • A natural testable extension is to replace the peak-force proxy with actual rock displacement measured by differencing point clouds before and after the push; this would provide the ground-truth label the current framework never measures and could retrain the model to be robust to rocks that move despite high resistance.
  • The same Bayesian conditioning could be extended to predict not just the peak force but the full force-time profile, which would allow the robot to distinguish rocks that yield after a sticking point from those that never move, a distinction the peak value alone cannot capture.
  • The paper's own lessons-learned note that contact with rough concave/convex rock surfaces causes rotation, slipping, or no movement depending on local contact pose; a more detailed local contact-surface feature at the push point could improve prediction and is an implicit next step.
  • The framework's logic could be transferred to other actuators, such as the quadruped's legs for kicking or jumping, by learning a mapping from the same visual features to the different force profiles those actions produce, which the authors list as future work.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 5 minor

Summary. The paper proposes a two-stage framework for estimating whether rocks on rough planetary terrain can be pushed by a mobile manipulator. The visual stage segments the scene point cloud using cosine-similarity-based surface normals and DBSCAN clustering, extracts volume, an ellipsoid-fit shape error, and surface normals, and filters obstacles through a hand-designed likelihood rule. The second stage uses a Bayesian linear regressor to predict the maximum force-feedback magnitude during a push and classifies an obstacle as pushable when the predicted maximum is below the robot's capability δ (Algorithm 1). The authors report simulation experiments (90 runs, 857 recorded data points) and real-world validation of the visual preprocessing module; no quantitative evaluation of the force prediction or pushability decision is reported, and no movement ground truth is measured.

Significance. The application is timely and relevant: clearing easily movable rocks instead of avoiding them could improve long-term route efficiency for repeated traversals on lunar or Martian terrain. The idea of combining visual features of the rock and supporting surface with proprioceptive force feedback is sensible, and the authors state that code and data will be released. However, the evaluation does not support the central claim. There are no prediction metrics (RMSE, R², classification accuracy, confusion matrix), no validation of the force-based proxy against actual rock movement, and a data-splitting procedure that leaks information across timesteps of the same push run. The real-world experiments validate only volume and surface-normal estimation, not pushability decisions. As presented, the paper is a promising preliminary framework rather than a demonstrated system.

major comments (4)
  1. [Section III-A and Algorithm 1] Pushability is defined as the probability that the obstacle moves, but the target variable learned and thresholded is the predicted maximum force magnitude f_max, and rock movement is never measured as ground truth. The decision rule in Algorithm 1 compares the predicted f_max to δ, and Section IV states a 20 N threshold that was 'empirically observed' without being derived from any measured movement outcome. A rock that moves despite a high transient force, or one that does not move despite a low force, would be misclassified, and the reported data cannot distinguish these cases. This is the load-bearing validation gap for the paper's central claim.
  2. [Section IV, dataset construction and split] The dataset construction and train/test split invalidate any quantitative claim that might be drawn from the learning experiment. The paper states that 'each timestep in the point cloud ROS bag is added to the dataset along with the resulting force signal' and that 'the maximum force-feedback magnitude for each run is recorded as the label for all data points in that run'; the dataset is then 'shuffled and split into training and test sets.' Consequently, multiple timesteps from the same push run, which share one force label, fall into both training and test sets, producing optimistic estimates. No accuracy, RMSE, R², or classification metrics are reported anywhere, so the actual effect of this leakage cannot be assessed.
  3. [Section IV, real-world experiments] The real-world experiments only validate the visual preliminary prediction module: estimated rock volumes (Figure 8) and average surface normals (Figure 3 and surrounding text). There is no real-world force-feedback prediction and no real-world pushability decision. The conclusion's claims about pushability therefore rest entirely on the simulation experiments, which are subject to the proxy and leakage concerns above.
  4. [Section V-B] The 'Lessons Learned' section lists possible contact outcomes—translational movement, rotational movement, slipping of the end effector, or no movement—and states that these depend on contact pose, orientation, local rock surface shape, and normal vector. This directly undermines the assumption that a single scalar f_max is a sufficient statistic for pushability, and no experimental evidence is provided to quantify how often each outcome occurs or how well f_max discriminates them. The central mapping from predicted force to pushability is therefore unverified even in the simulation domain.
minor comments (5)
  1. [Section IV, real-world surface normal text] The text says '0.1π which is roughly equal to 5 degrees,' but 0.1π radians is approximately 18 degrees, not 5 degrees. This numerical error should be corrected, and the reported slope angle should be reconciled with the stated 8-degree test ramp.
  2. [Algorithm 1] The pseudocode's Bayesian update is not written consistently: the evidence term p(F_peak,i | R_i) is not the proper normalizer for a standard posterior update, and the procedure does not match the statement that Scikit-learn's BayesianRidge is used for fitting. The pseudocode should be corrected or replaced with the actual inference procedure.
  3. [Section III-A, surface definition] The definition of the supporting surface S_j uses the condition z_min ≤ p_z, which appears to include all ground points above the lower z-bound of the bounding box rather than only points inside the scaled 3D box. The geometric condition should be made precise.
  4. [General presentation] There are several typographical and formatting issues, such as 'Figure 2.' with a stray period before the colon, broken equation text ('p i ∈R 3.wheren is'), and inconsistent notation for the DBSCAN output (e.g., the set notation with 'u' is unclear). These should be cleaned up in revision.
  5. [Section III-B.1, VPP likelihood] The heuristic likelihood g_j = (δ/(V_j E_j))(1 - θ/π) contains several free design choices, and no sensitivity analysis or ablation is provided for the thresholds T_cs, T_low, T_high, or the DBSCAN parameters. Since the downstream regressor uses features derived from this stage, the robustness of the full pipeline to these choices is unclear.

Circularity Check

2 steps flagged · score 6.0 of 10

Pushability is defined through the maximum force-feedback variable, so the central classification reduces to the fitted regression target.

  1. self definitional [Section III-A (Problem Definition) and Algorithm 1, lines 9-14]
    "The pushability affordance of an obstacle O_j is defined as the probability of it moving by predicting the force feedback conditioned on the size V_j and shape E_j of the object and the normal surface N_j such that A_j = (f_j max | V_j, E_j, N_j)."

    This sentence defines the affordance (probability of movement) as a conditional statement about the maximum force magnitude, which is precisely the variable the Bayesian regressor is trained to output. Algorithm 1 then makes the pushable/not-pushable decision by thresholding the predicted peak force against delta. Because the dataset records only peak force as the label and never a movement outcome, the derivation of pushability from force feedback is true by definition rather than by measurement; predicting pushability is the same operation as predicting the fitted force target.

  2. fitted input called prediction [Section IV (Experiments), simulation threshold sentence; Algorithm 1]
    "In the simulation experiments, it was empirically observed that if the magnitude of the sensed force feedback exceeds 20 N, the robot is unable to move the rock. ... The maximum force-feedback magnitude for each run is recorded as the label for all data points in that run."

    The 20 N cutoff is an empirical observation made on the same force-feedback data that constitute the training labels, and the decision rule compares the predicted peak force to this cutoff. No independent movement label enters the loop, so classifying a rock as pushable when predicted force is below the cutoff restates the calibration of the fitted variable rather than testing the stated affordance, namely whether the rock moves. The threshold may be physically meaningful, but as presented it is read off the predicted quantity.

full rationale

The BayesianRidge regression from point-cloud features to maximum force is internally self-contained and, considered purely as a force predictor, could be evaluated on held-out force labels without circularity. The circularity is narrower but central: the paper renames this regression target pushability and verbally defines pushability as the probability that the rock moves, yet movement is never measured as a ground-truth label. Consequently, the claimed pushability classification reduces by construction to a thresholded prediction of the same force-feedback quantity used to fit the model. The self-citations in the paper, notably [34]-[36] and [39], are background or related-work references and are not load-bearing in the derivation; there is no imported uniqueness theorem and no ansatz smuggled in via citation. The core concern is therefore not a general lack of independent support but a specific definitional collapse of the affordance into the fitted variable.

Assumptions & free parameters 6 free parameters · 6 assumptions · 0 invented entities

The framework rests on several hand-set thresholds and modeling assumptions. The most consequential is the assumption that peak push force is a sufficient proxy for actual rock movement, which is never validated against ground truth; this assumption, together with the fitted thresholds, carries the central claim.

free parameters (6)
  • Cosine similarity threshold (T_cs) = 0.85
    Table I; separates ground from obstacle points during segmentation; no sensitivity analysis is reported.
  • DBSCAN epsilon and min points = epsilon=0.5 m, min_points=5
    Table I; controls rock clustering; values are chosen without empirical justification.
  • Bounding box scale factor (s) = 1.5
    Table I; defines the surface patch around each rock and hence the surface normal features.
  • Visual likelihood thresholds (T_low, T_high) = 0.3, 0.8
    Table I and Eq. 1; hand-set to classify rocks as Static, Pushable, or Override.
  • Pushability force threshold (simulation) = 20 N
    Section IV; empirically observed threshold above which the robot cannot move the rock, used as a proxy for pushability without measuring actual movement.
  • BayesianRidge prior parameters = scikit-learn defaults (unreported)
    Model is fit using BayesianRidge from Scikit-learn; the specific prior hyperparameters are not listed in the paper.
assumptions (6)
  • domain assumption No slipping between the end effector and the obstacle surface during pushing.
    Stated in Section V-A; if slipping occurs, the force feedback pattern would not reflect the resistance of the rock, invalidating the predicted-force-to-pushability mapping.
  • domain assumption Rocks are detached from the ground and are not outcrops.
    Stated in Section V-A; outcrops or embedded rocks would require much higher forces and break the volume/shape-based prediction.
  • domain assumption The median surface normal of the scene point cloud represents the ground normal.
    Section III-B1; on rough terrain with large rocks, the median normal may not correspond to the ground, causing incorrect segmentation.
  • domain assumption After ground filtering, remaining points form spatially distinct clusters, one per obstacle, so DBSCAN separates rocks reliably.
    Section III-B1; touching or overlapping rocks would be merged into a single cluster, corrupting volume/shape features.
  • ad hoc to paper The maximum force magnitude over a fixed-length push is a sufficient statistic for pushability.
    Section III-A defines pushability as a function of f_max; this is a modeling choice that is never validated against actual rock movement. It is the central proxy that makes the whole pipeline work.
  • ad hoc to paper The empirically observed 20 N threshold in simulation separates movable from immovable rocks.
    Section IV states this was empirically observed; the threshold is derived from the same simulation data and is not justified from physics or independently measured movement labels.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Learning Rock Pushability on Rough Planetary Terrain." pith.science (2026). https://pith.science/paper/A5IHNA6M

@misc{pith2026250509833,
  author       = {Pith},
  title        = {Pith review of: Learning Rock Pushability on Rough Planetary Terrain},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/A5IHNA6M}},
  note         = {Machine review of arXiv:2505.09833}
}
read the original abstract

In the context of mobile navigation in unstructured environments, the predominant approach entails the avoidance of obstacles. The prevailing path planning algorithms are contingent upon deviating from the intended path for an indefinite duration and returning to the closest point on the route after the obstacle is left behind spatially. However, avoiding an obstacle on a path that will be used repeatedly by multiple agents can hinder long-term efficiency and lead to a lasting reliance on an active path planning system. In this study, we propose an alternative approach to mobile navigation in unstructured environments by leveraging the manipulation capabilities of a robotic manipulator mounted on top of a mobile robot. Our proposed framework integrates exteroceptive and proprioceptive feedback to assess the push affordance of obstacles, facilitating their repositioning rather than avoidance. While our preliminary visual estimation takes into account the characteristics of both the obstacle and the surface it relies on, the push affordance estimation module exploits the force feedback obtained by interacting with the obstacle via a robotic manipulator as the guidance signal. The objective of our navigation approach is to enhance the efficiency of routes utilized by multiple agents over extended periods by reducing the overall time spent by a fleet in environments where autonomous infrastructure development is imperative, such as lunar or Martian surfaces.

Figures

Figures reproduced from arXiv: 2505.09833 by the authors.

Figure 1
Figure 1. A depiction of rock pushing in rough terrains is [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. The architecture of our proposed pushability learning probabilistic model, which incorporates both visual and [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. VPP can determine the normal of surface the obstacle [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: A rock-pushing experiment in the simulation environ [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 5
Figure 5. Figure 5: VPP distinguishes the slope of the surface in the [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]
Figure 6
Figure 6. Figure 6: The mean and standard deviation of force feedback [PITH_FULL_IMAGE:figures/full_fig_p006_6.png]
Figure 8
Figure 8. Figure 8: VPP accurately distinguishes the obstacles by their [PITH_FULL_IMAGE:figures/full_fig_p006_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

42 extracted references · 42 canonical work pages

  1. [1]

    Rafai, A. N. A., Adzhar, N., & Jaini, N. I. (2022). A review on path planning and obstacle avoidance algorithms for autonomous mobile robots. *Journal of Robotics, 2022*(1), 2538220

  2. [2]

    Guo, B., Guo, N., & Cen, Z. (2022). Obstacle avoidance with dynamic avoidance risk region for mobile robots in dynamic environments. *IEEE Robotics and Automation Letters, 7*(3), 5850-5857

  3. [3]

    Choi, J., Lee, G., & Lee, C. (2021). Reinforcement learning-based dy- namic obstacle avoidance and integration of path planning. *Intelligent Service Robotics, 14*, 663-677

  4. [4]

    & Tsuda, Y

    Morota, T., Sugita, S., Cho, Y ., Kanamaru, M., Tatsumi, E., Sakatani, N., ... & Tsuda, Y . (2020). Sample collection from asteroid (162173) Ryugu by Hayabusa2: Implications for surface evolution. *Science, 368*(6491), 654-659

  5. [5]

    S., Adam, C

    Lauretta, D. S., Adam, C. D., Allen, A. J., Ballouz, R. L., Barnouin, O. S., Becker, K. J., ... & Yumoto, K. (2022). Spacecraft sample collection and subsurface excavation of asteroid (101955) Bennu. *Science, 377*(6603), 285-291

  6. [6]

    Cruz Ulloa, C., Prieto S ´anchez, G., Barrientos, A., & Del Cerro, J. (2021). Autonomous thermal vision robotic system for victims recognition in search and rescue missions. *Sensors, 21*(21), 7346

  7. [7]

    Groves, K., Hernandez, E., West, A., Wright, T., & Lennox, B. (2021). Robotic exploration of an unknown nuclear environment using radiation informed autonomous navigation. *Robotics, 10*(2), 78

  8. [8]

    Di, K., Yue, Z., Liu, Z., & Wang, S. (2013). Automated rock detection and shape analysis from mars rover imagery and 3D point cloud data. Journal of Earth Science, 24(1), 125-135

Show all 42 references
  1. [9]

    D., Zou, T., & Zeleke, D

    Teji, M. D., Zou, T., & Zeleke, D. S. (2023). A survey of off- road mobile robots: Slippage estimation, robot control, and sensing technology. Journal of Intelligent & Robotic Systems, 109(2), 38

  2. [10]

    Suomalainen, M., Karayiannidis, Y ., & Kyrki, V . (2022). A survey of robot manipulation in contact. Robotics and Autonomous Systems, 156, 104224

  3. [11]

    Ozdamar, I., Sirintuna, D., Arbaud, R., & Ajoudani, A. (2024). Pushing in the Dark: A Reactive Pushing Strategy for Mobile Robots Using Tactile Feedback. IEEE Robotics and Automation Letters

  4. [12]

    Heins, A., & Schoellig, A. P. (2024). Force Push: Robust Single-Point Pushing With Force Feedback. IEEE Robotics and Automation Letters

  5. [13]

    A., & Lepora, N

    Yang, M., Lin, Y ., Church, A., Lloyd, J., Zhang, D., Barton, D. A., & Lepora, N. F. (2023). Sim-to-real model-based and model-free deep reinforcement learning for tactile pushing. IEEE Robotics and Automation Letters

  6. [14]

    R., Ugur, E., & Ucoluk, G

    Sahin, E., Cakmak, M., Dogar, M. R., Ugur, E., & Ucoluk, G. (2007). To afford or not to afford: A new formalization of affordances toward affordance-based robot control. Adaptive Behavior, 15(4), 447-472

  7. [15]

    Y ., Piater, J., Oztop, E., & Ugur, E

    Ahmetoglu, A., Seker, M. Y ., Piater, J., Oztop, E., & Ugur, E. (2022). Deepsym: Deep symbol generation and rule learning for planning from unsupervised robot interaction. Journal of Artificial Intelligence Research, 75, 709-745

  8. [16]

    Lloyd, J., & Lepora, N. F. (2021). Goal-driven robotic pushing using tactile and proprioceptive feedback. IEEE Transactions on Robotics, 38(2), 1201-1212

  9. [17]

    Jamone, L., Ugur, E., Cangelosi, A., Fadiga, L., Bernardino, A., Piater, J., & Santos-Victor, J. (2016). Affordances in psychology, neuro- science, and robotics: A survey. *IEEE Transactions on Cognitive and Developmental Systems, 10*(1), 4-25

  10. [18]

    Arvidson, R. E. (2016). Aqueous history of Mars as inferred from landed mission measurements of rocks, soils, and water ice. *Journal of Geophysical Research: Planets, 121*(9), 1602-1626. Presented at the 2025 IEEE ICRA Workshop on Field Robotics

  11. [19]

    W., Gellert, R., VanBommel, S., Arvidson, R

    Mittlefehldt, D. W., Gellert, R., VanBommel, S., Arvidson, R. E., Ashley, J. W., Clark, B. C., ... & Schroeder, C. (2021). Geology and geochemistry of Noachian bedrock and alteration events, Meridiani Planum, Mars: MER Opportunity observations. *Journal of Geophys- ical Resear...

  12. [20]

    J., Valyear, K

    Seminara, L., Gastaldo, P., Watt, S. J., Valyear, K. F., Zuher, F., & Mastrogiovanni, F. (2019). Active haptic perception in robots: a review. *Frontiers in Neurorobotics, 13*, 467142

  13. [21]

    S., Zenati, A., & Aouf, N

    Bahraini, M. S., Zenati, A., & Aouf, N. (2021, May). Autonomous cooperative visual navigation for planetary exploration robots. In 2021 IEEE International Conference on Robotics and Automation (ICRA) (pp. 9653-9658). IEEE

  14. [22]

    G., Demmel, N., Wedler, A., Triebel, R., Cremers, D., & St ¨urzl, W

    Wudenka, M., M ¨uller, M. G., Demmel, N., Wedler, A., Triebel, R., Cremers, D., & St ¨urzl, W. (2021, September). Towards robust monocular visual odometry for flying robots on planetary missions. In 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS...

  15. [23]

    M., Morrell, B., Delaune, J., & Scaramuzza, D

    Mahlknecht, F., Gehrig, D., Nash, J., Rockenbauer, F. M., Morrell, B., Delaune, J., & Scaramuzza, D. (2022). Exploring event camera- based odometry for planetary robots. IEEE Robotics and Automation Letters, 7(4), 8651-8658

  16. [24]

    Zhou, B., Luo, S., & Zhang, S. (2022). An Innovative Pose Deter- mination Algorithm for Planetary Rover Onboard Visual Odometry. Aerospace, 9(7), 391

  17. [25]

    Lu, C., Yu, M., Li, H., & Cui, H. (2025). Landmark-aware autonomous odometry correction and map pruning for planetary rovers. Acta Astronautica, 226, 86-96

  18. [26]

    M., Choi, M., & Shin, H

    Hong, S., Bangunharcana, A., Park, J. M., Choi, M., & Shin, H. S. (2021). Visual SLAM-based robotic mapping method for planetary construction. Sensors, 21(22), 7715

  19. [27]

    Romero-Bautista, V ., Altamirano-Robles, L., D ´ıaz-Hern´andez, R., Zapotecas-Mart´ınez, S., & Sanchez-Medel, N. (2024). Evaluation of visual SLAM algorithms in unstructured planetary-like and agricul- tural environments. Pattern Recognition Letters, 186, 106-112

  20. [28]

    A., & Zhao, Y

    Kuang, B., Wisniewski, M., Rana, Z. A., & Zhao, Y . (2021). Rock segmentation in the navigation vision of the planetary rovers. Mathe- matics, 9(23), 3048

  21. [29]

    A., Zhao, Y ., Sun, S., & Nnabuife, S

    Kuang, B., Gu, C., Rana, Z. A., Zhao, Y ., Sun, S., & Nnabuife, S. G. (2022). Semantic terrain segmentation in the navigation vision of planetary rovers—A systematic literature review. Sensors, 22(21), 8393

  22. [30]

    (2022, March)

    Ebadi, K., Coble, K., Kogan, D., Atha, D., Schwartz, R., Padgett, C., & Vander Hook, J. (2022, March). Semantic mapping in unstructured environments: Toward autonomous localization of planetary robotic explorers. In IEEE Aerospace Conference

  23. [31]

    A., & Zhao, Y

    Kuang, B., Rana, Z. A., & Zhao, Y . (2021). Sky and ground segmenta- tion in the navigation visions of the planetary rovers. Sensors, 21(21), 6996

  24. [32]

    Gonzalez, R., Apostolopoulos, D., & Iagnemma, K. (2018). Slippage and immobilization detection for planetary exploration rovers via ma- chine learning and proprioceptive sensing. Journal of Field Robotics, 35(2), 231-247

  25. [33]

    C., & Olivares-Mendez, M

    Coloma, S., Martinez, C., Yalc ¸ın, B. C., & Olivares-Mendez, M. A. (2022). Enhancing rover teleoperation on the moon with proprioceptive sensors and machine learning techniques. IEEE Robotics and Automa- tion Letters, 7(4), 11434-11441

  26. [34]

    Kilic, C., Gu, Y ., & Gross, J. N. (2022). Proprioceptive slip detection for planetary rovers in perceptually degraded extraterrestrial environ- ments. in Field Robotics, vol. 2, pp. 1754-1778, Aug. 2022, doi: 10.55417/fr.2022054

  27. [35]

    Kilic, C., Ohi, N., Gu, Y ., & Gross, J. N. (2021). Slip-based au- tonomous ZUPT through Gaussian process to improve planetary rover localization. IEEE robotics and automation letters, 6(3), 4782-4789

  28. [36]

    N., Ohi, N., Watson, R., Strader, J., Swiger, T.,

    Kilic, C., Gross, J. N., Ohi, N., Watson, R., Strader, J., Swiger, T., ... & Gu, Y . (2019, November). Improved planetary rover inertial navigation and wheel odometry performance through periodic use of zero-type constraints. In 2019 IEEE/RSJ International Conference on Intell...

  29. [37]

    & Deng, Z

    Feng, W., Ding, L., Zhou, R., Xu, C., Yang, H., Gao, H., ... & Deng, Z. (2023). Learning-based end-to-end navigation for planetary rovers considering non-geometric hazards. IEEE Robotics and Automation Letters, 8(7), 4084-4091

  30. [38]

    Dimastrogiovanni, M., Cordes, F., & Reina, G. (2020). Terrain es- timation for planetary exploration robots. Applied Sciences, 10(17), 6044

  31. [39]

    Girgin, T., & U ˘gur, E. (2024). Multi-Object Graph Affordance Net- work: Goal-Oriented Planning through Learned Compound Object Affordances. IEEE Transactions on Cognitive and Developmental Systems

  32. [40]

    & Wang, H

    Zhang, J., Gireesh, N., Wang, J., Fang, X., Xu, C., Chen, W., ... & Wang, H. (2024, May). Gamma: Graspability-aware mobile manipulation policy learning based on online grasping pose fusion. In 2024 IEEE International Conference on Robotics and Automation (ICRA) (pp. 1399-1405). IEEE

  33. [41]

    E., Haustein, J

    King, J. E., Haustein, J. A., Srinivasa, S. S., & Asfour, T. (2015, May). Nonprehensile whole arm rearrangement planning on physics manifolds. In 2015 IEEE International Conference on Robotics and Automation (ICRA) (pp. 2508-2515). IEEE

  34. [42]

    E., Erdem, A., Erdem, E., Asfour, T., & Ugur, E

    Tekden, A. E., Erdem, A., Erdem, E., Asfour, T., & Ugur, E. (2024). Object and relation centric representations for push effect prediction. Robotics and Autonomous Systems, 174, 104632. Presented at the 2025 IEEE ICRA Workshop on Field Robotics

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.