Pith. sign in

REVIEW 3 major objections 5 minor 45 references

K-VARK: Kernelized Variance-Aware Residual Kalman Filter for Sensorless Force Estimation in Collaborative Robots

T0 review · 3 major / 5 minor · reviewed 2026-08-03 · deepseek-v4-flash

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

Pith's one-line read A Kalman filter that trusts a learned residual-torque model only where data supports it cuts sensorless force-estimation error by more than 20 percent.

desk verdict K-VARK is a genuinely new combination — KMP residual mean/variance into an adaptive Kalman filter — but the headline 20% gain shrinks to single digits against the stronger baseline, and the paper never tests whether a velocity-only free-motion residual model stays valid under load. read the letter →

arxiv 2512.13009 v2 pith:GJ2YKPXL submitted 2025-12-15 cs.RO

Oğuzhan Akb{i}y{i}k , Naseem Alhousani , Fares J. Abu-Dakka This is my paper · ORCID
classification cs.RO
keywords sensorlessforceestimationresidualtorquemodelingKernelizedMovementPrimitivesadaptiveKalmanfiltervariationalBayesheteroscedasticvariancecollaborativerobotsmomentumobserver
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 proposes K-VARK, a sensorless force observer that replaces the usual 'friction model' in a momentum-based Kalman filter with a kernelized probabilistic model of residual torques—everything the nominal dynamics fail to explain. The key move is to feed not just the predicted mean of that residual model into the filter, but also its input-dependent variance, which is added to the measurement noise covariance. In data-rich velocity regions the filter trusts the virtual measurement; in sparse or extrapolated regions the variance inflates and the filter falls back on its prior, while a variational-Bayes step adapts the process noise online. On a 6-DoF collaborative robot, the authors report over 20% lower RMSE than a Gaussian-process baseline for external torque estimation, with real-time per-sample cost, in both loaded-trajectory and F/T-instrumented polishing experiments. The authors explicitly list limitations: single robot, limited near-zero-velocity analysis, and no full ablation of the variance and VB components.

What carries the argument

The engine is the virtual measurement ζ*_k = x_k − x_{k−1} − t_s u_{k−1} + t_s μ*_k (Eq. 28), which converts the momentum residual into a noisy observation of external torque after subtracting the KMP mean. Its noise covariance is Σ_{ν,k} = t_s² Σ*_{k} + Σ_{emp,k}: the KMP predictive covariance—capturing both aleatoric data variability and epistemic distance-to-training effects—plus an innovation-driven empirical term. A variational-Bayes inverse-Wishart update adapts the process noise covariance online, so the filter can respond to contact transitions while the KMP variance decides how much to trust the measurement.

What would settle it

Run the robot along the training trajectories with a known end-effector mass attached, and compare K-VARK's estimates against a wrist F/T sensor. If the error grows systematically with payload size while the KMP variance stays approximately constant, the velocity-only residual model is missing load-dependent torque and the central claim is falsified.

Watch

Extended reading notes

Core claim

K-VARK's central claim is that residual torques—the mismatch between commanded motor torque and the nominal rigid-body model—should be treated as a probability distribution, not a fixed friction curve, and that distribution's variance is as informative as its mean. The paper shows that when the KMP-predicted mean is subtracted from the momentum residual, the leftover is a clean virtual measurement of external torque only if the residual model is right; when the KMP variance is added to the measurement noise, the Kalman gain automatically down-weights that virtual measurement in regions where the residual model is uncertain. On a 6-DoF arm, this yields joint-space and Cartesian force/torque e

Load-bearing premise

The offline residual-torque model, trained only in free motion as a function of joint velocity, is assumed unchanged during loaded and contact motion; if residual torque also depends on payload, configuration, or contact state, the virtual measurement is biased and the force estimate inherits that bias.

Editorial extensions

If this is right

  • Joint-space and Cartesian external wrench estimates improve over GP- and NN-based observers, most clearly against the Gaussian-process baseline, while per-sample compute stays below real-time limits.
  • In data-rich velocity regions the filter trusts the learned residual model; in extrapolated regions it automatically down-weights the virtual measurement, which should make the observer more robust to novel motions.
  • The KMP variance encodes both aleatoric and epistemic uncertainty, so unlike GP-based observers the filter carries a calibrated confidence signal into the estimation loop.
  • Because no force/torque sensor is needed, the approach bears directly on cost-sensitive collaborative applications like polishing, assembly, and surface finishing.
  • The variational-Bayes process-noise adaptation couples to the measurement variance, so the filter can track time-varying disturbances without hand-tuned gain scheduling.

Reading between the lines

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

  • The same variance-gating principle could set safety thresholds in impedance control or collision detection: whenever KMP variance is large, a controller could automatically reduce interaction stiffness or raise a contact alarm.
  • A natural falsifying extension is to carry known payloads during training-free operation; if the velocity-only residual model cannot explain residuals under load—their remaining bias exceeding the KMP variance—the central assumption fails.
  • Because KMP's uncertainty hyperparameters (λ2, σ_f²) can be tuned without changing the predictive mean, transferring the model to a new robot may only require recalibrating variance scaling, not re-fitting the mean—an unverified but plausible consequence.
  • Near-zero velocity is the weak spot the authors themselves flag: stick-slip and temperature-dependent friction are exactly the regimes where a velocity-only heteroscedastic model is most likely to under-represent the true residual distribution.
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.

Referee Report

3 major / 5 minor

Summary. The paper proposes K-VARK, a sensorless external-torque observer for collaborative robots. A per-joint residual-torque model is learned offline with Kernelized Movement Primitives (KMP) from free-motion excitation data, providing a predictive mean and an input-dependent variance. The mean is used to correct a momentum-based residual, giving a virtual measurement of external torque; the variance is added to the measurement-noise covariance, and the process-noise covariance is adapted online via variational Bayes. Experiments on a 6-DoF robot compare the method against GMR-GP, GPADKF, and NN-based observers, reporting lower RMSE in joint-space and Cartesian wrench estimation.

Significance. If the reported accuracy gains hold under the stated operating conditions, integrating KMP-style heteroscedastic uncertainty into an adaptive Kalman filter is a plausible and practically useful contribution to sensorless force estimation. The work addresses a real limitation of GP-based residual models, which typically capture only epistemic uncertainty, and it provides real-robot comparisons against multiple baselines, including the GPADKF approach of [24] and a neural-network variant. The paper is also transparent about several limitations in its conclusion. However, as detailed below, the central quantitative claim is not yet supported by the experiments as written, and there is a formal inconsistency in the virtual-measurement update that must be resolved.

major comments (3)
  1. [Section IV-C, Eqs. (28)-(29) and Algorithm 1, line 18] The virtual measurement is defined as ζ*_k = x_k − x_{k−1} − t_s u_{k−1} + t_s μ*_k = −t_s τ_ext,k−1, but Eq. (29) and the KF recursion (44) treat ζ*_k as a direct measurement of the current state τ_ext,k. This is an unflagged one-step lag. With the random-walk state model (32), the update (44) corrects the current external-torque estimate with a measurement of the previous external torque, which is not the stated equality. Either define ζ*_{k+1} = −t_s τ_ext,k, or set the measurement model to H ω_{k−1}. The same index issue applies to the KMP mean: Eq. (8) requires the residual torque at the previous sample, τ_r,k−1, not μ*_k.
  2. [Section V-A, Eq. (46)] The ground truth in the first experiment is τ_loaded − τ_free. This equals the true external torque only if the residual torque is identical with and without the load. The residual model is trained on free-motion data using only joint velocity as input (Remark 1, Section IV-A), and the ARD analysis in Section VI (Fig. 13) shows that non-velocity features are not wholly irrelevant. Under a payload or during contact, load-dependent changes in friction and other residual effects are not captured by μ*, so they are attributed to τ_ext by Eq. (28). The reported RMSE therefore does not isolate external-torque estimation error. This is the load-bearing assumption for the headline claim and should be validated with varying payloads/contact configurations, or the claim should be weakened.
  3. [Abstract and Section V-C, Figs. 7-8, Table IV] The abstract's 'over 20% reduction in RMSE' is computed only against the GMR-GP baseline in Experiment 1 (1.49 → 1.18). Against GPADKF, the state-of-the-art baseline from [24], the improvement is about 7% (1.27 → 1.18). In Experiment 2, the Cartesian RMSE improvement over GMR-GP is about 4.7%. The quantitative claims should be reported per-domain and against the strongest baseline. In addition, the Cartesian average in Table IV is computed as a Euclidean norm over forces and moments without unit normalization, producing a mixed-unit scalar that is not a meaningful average; separate force and moment metrics should be reported.
minor comments (5)
  1. [Section II and V-A] Several typos: 'methods are often suffer', 'Generalized momentum observerss', and '64GP of RAM'.
  2. [Section III-D, Eq. (21)] The notation l^{-1} in the squared-exponential kernel is ambiguous; l is introduced as a vector in the hyperparameter list. Define whether l is a diagonal matrix or a vector and write the kernel accordingly.
  3. [Figures 5 and 7] The confidence intervals and method colors are hard to distinguish in black-and-white print. Consider using line styles or adding direct labels.
  4. [Table IV] K-VARK is not the lowest in several columns (e.g., τ1, τ6, mz). The claim of highlighting the lowest value in each column is not realized, and pairwise improvement should be stated with per-column numbers rather than only averages.
  5. [Section IV-D, Eq. (35)] The notation E[P_{k|k-1}^{-1}] is unclear; the standard posterior precision update uses P_{k|k-1}^{-1}, not an expectation of it. Please clarify the operation.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular reasoning found: the external-torque estimate is the momentum residual after subtracting an independently trained KMP residual model, not a rearrangement of fitted force labels.

full rationale

The central derivation is self-contained. Equation (28) defines the virtual measurement as ζ*_k = x_k − x_{k−1} − t_s u_{k−1} + t_s μ*_k, which is exactly the discrete momentum balance (8) rearranged: x_k − x_{k−1} − t_s u_{k−1} = −t_s(τ_ext,k−1 + τ_r,k−1). The KMP model is trained offline on free-motion residual torques τ_r = τ_m − τ_EL (Eq. 27), with no external torque present, so μ*_k is an input to the observer rather than a copy of the force estimate. The external torque is whatever remains after subtracting μ*_k; this is a standard residual-observer construction and is not circular. The Kalman update, VB covariance adaptation, and KMP uncertainty injection do not fit any parameter to the external-torque labels. The main validity concern — that the velocity-only free-motion residual model must remain valid under load, as presupposed by the Experiment 1 ground truth τ_loaded − τ_free — is a correctness/generalization issue, not a circularity issue. Similarly, the abstract's 'over 20%' improvement is computed against GMR-GP rather than the GPADKF baseline (~7% in Experiment 1), which is an overstatement but not circular. The only self-citation ([25], co-authored by F. J. Abu-Dakka) supports the general statement that KMP has been used in imitation learning; it is not load-bearing for the force-estimation derivation. The paper's own acknowledged limitations — single-robot evaluation, limited near-zero-velocity analysis, and incomplete ablation of VB and variance components — further indicate honest reporting rather than hidden circularity.

Assumptions & free parameters 9 free parameters · 8 assumptions · 0 invented entities

The method rests on standard robot dynamics and KMP, but the central claim depends on several hand-set parameters (KMP hyperparameters, GMM components, filter covariances) and on the modeling assumption that residual torque is a function of joint velocity alone. No code or data are released.

free parameters (9)
  • KMP length scale l_j (per joint) = [0.0100, 0.0200, 0.0110, 0.0790, 0.1154, 0.1244]
    Set empirically in Section V-B; controls kernel decay and variance growth outside the training region.
  • KMP mean regularization lambda1_j = [0.0676, 0.0020, 0.5032, 0.0729, 0.0300, 0.03418]
    Set empirically; controls mean fitting in Eq. (22).
  • KMP variance regularization lambda2_j = 1e5 * [10,20,20,20,20,20]
    Set empirically; scales the predictive covariance in Eq. (23) by N/lambda2 and directly scales the measurement noise covariance.
  • KMP kernel amplitude sigma_f^2 = 1e4
    Set empirically; affects asymptotic variance and kernel scale.
  • Number of GMM components N = 20
    Chosen for the KMP reference distribution (Section V-B); affects mean/variance smoothness.
  • Empirical noise forgetting factor rho = not specified
    Appears in Eq. (31) for online adaptation of Sigma_emp; no value or adaptation schedule is given.
  • VB iterations M = not specified
    Algorithm 1 input; number of variational iterations affects adaptation speed.
  • Initial covariances Sigma_d, Sigma_emp = not specified
    Algorithm 1 initialization; values are not reported.
  • Empirical noise bound/weight in Remark 2 = not specified
    Sigma_emp is 'bounded and appropriately weighted' but no bound or weight is given; needed to avoid double-counting.
assumptions (8)
  • standard math Rigid-body Euler-Lagrange dynamics (Eq. 1) with skew-symmetric Mdot - 2C
    Standard robot model; used for the momentum formulation.
  • standard math Forward Euler discretization of momentum dynamics (Eq. 8) at sampling period t_s
    Discrete state-space model underpinning the Kalman filter.
  • standard math KMP predictive mean/covariance formulas (Eqs. 22-23) from [12]
    Adopted without derivation; assumes a positive-definite kernel and GMR reference distribution.
  • domain assumption Residual torque depends only on joint velocity
    Section IV-A Remark 1 and Section VI: input restricted to velocity; if false, the virtual measurement is biased.
  • domain assumption External torque follows a first-order random walk (Eq. 32)
    Assumes slowly varying external force; process model for the filter.
  • domain assumption Measurement noise zero-mean Gaussian with covariance t_s^2 Sigma_* + Sigma_emp (Eq. 30)
    Assumes KMP residual prediction errors are zero-mean and independent of sensor noise.
  • domain assumption Inverse-Wishart prior and conjugate update for process covariance (Eqs. 33-40)
    Variational Bayesian adaptation from [13]; the paper does not re-derive it.
  • domain assumption Ground truth external torque in first experiment equals loaded minus free motor torques (Eq. 46)
    Assumes identical dynamics/friction across the two runs except for the added load.

how reviews work

0 comments
Cite this review

Pith. "Pith review of K-VARK: Kernelized Variance-Aware Residual Kalman Filter for Sensorless Force Estimation in Collaborative Robots." pith.science (2026). https://pith.science/paper/GJ2YKPXL

@misc{pith2026251213009,
  author       = {Pith},
  title        = {Pith review of: K-VARK: Kernelized Variance-Aware Residual Kalman Filter for Sensorless Force Estimation in Collaborative Robots},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GJ2YKPXL}},
  note         = {Machine review of arXiv:2512.13009}
}
read the original abstract

Reliable estimation of contact forces is crucial for ensuring safe and precise interaction of robots with unstructured environments. However, accurate sensorless force estimation remains challenging due to inherent modeling errors and complex residual dynamics and friction. To address this challenge, in this paper, we propose K-VARK (Kernelized Variance-Aware Residual Kalman filter), a novel approach that integrates a kernelized, probabilistic model of joint residual torques into an adaptive Kalman filter framework. Through Kernelized Movement Primitives trained on optimized excitation trajectories, K-VARK captures both the predictive mean and input-dependent heteroscedastic variance of residual torques, reflecting data variability and distance-to-training effects. These statistics inform a variance-aware virtual measurement update by augmenting the measurement noise covariance, while the process noise covariance adapts online via variational Bayesian optimization to handle dynamic disturbances. Experimental validation on a 6-DoF collaborative manipulator demonstrates that K-VARK achieves over 20% reduction in RMSE compared to state-of-the-art sensorless force estimation methods, yielding robust and accurate external force/torque estimation suitable for advanced tasks such as polishing and assembly.

Figures

Figures reproduced from arXiv: 2512.13009 by the authors.

Figure 1
Figure 1. Comparison between KF behaviors using GP or KMP for process [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 3
Figure 3. GMR reference trajectories for residual torque modeling across six [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figure 2
Figure 2. Experimental setup using Orion5, F/T sensor and custom made [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗
Figures from the paper (6 more)
Figure 6
Figure 6. Figure 6: Average RMSE of residual torque modelling for 4 different methods. [PITH_FULL_IMAGE:figures/full_fig_p010_6.png]
Figure 7
Figure 7. Figure 7: Time histories of external torque estimates for all six joints: [PITH_FULL_IMAGE:figures/full_fig_p010_7.png]
Figure 10
Figure 10. Figure 10: Comparison of total computation time (left axis) and average [PITH_FULL_IMAGE:figures/full_fig_p011_10.png]
Figure 9
Figure 9. Figure 9: External Force Estimates for 6 Cartesian Directions [PITH_FULL_IMAGE:figures/full_fig_p011_9.png]
Figure 12
Figure 12. Figure 12: Residual torque plots agains velocity for all joints [PITH_FULL_IMAGE:figures/full_fig_p012_12.png]
Figure 13
Figure 13. Figure 13: ARD Kernel heatmap for the features VII. CONCLUSION We introduced K-VARK, a sensorless force observer that couples a probabilistic residual-torque model with a variance￾aware disturbance Kalman filter. By (i) learning residual means and input-dependent variance from e…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

45 extracted references · 3 linked inside Pith

  1. [24]

    Contact force estimation of robot manipulators with imperfect dynamic model: on gaussian process adaptive disturbance kalman filter,

    Y . Wei, S. Lyu, W. Li, X. Yu, Z. Wang, and L. Guo, “Contact force estimation of robot manipulators with imperfect dynamic model: on gaussian process adaptive disturbance kalman filter,”IEEE Transactions on Automation Science and Engineering, 2023

    2023

  2. [1]

    Industry 4.0: The future of productivity and growth in manufacturing industries,

    M. R ¨ußmann, M. Lorenz, P. Gerbert, M. Waldner, J. Justus, P. Engel, and M. Harnisch, “Industry 4.0: The future of productivity and growth in manufacturing industries,”Boston consulting group, vol. 9, no. 1, pp. 54–89, 2015

    2015

  3. [2]

    Collaborative robots in manufacturing and assembly systems: literature review and future research agenda,

    A. Keshvarparast, D. Battini, O. Battaia, and A. Pirayesh, “Collaborative robots in manufacturing and assembly systems: literature review and future research agenda,”Journal of Intelligent Manufacturing, vol. 35, no. 5, pp. 2065–2118, 2024

  4. [3]

    Review on robot-assisted polishing: Status and future trends,

    X. Ke, Y . Yu, K. Li, T. Wang, B. Zhong, Z. Wang, L. Kong, J. Guo, L. Huang, M. Idiret al., “Review on robot-assisted polishing: Status and future trends,”Robotics and Computer-integrated manufacturing, vol. 80, p. 102482, 2023

    2023

  5. [4]

    Sensorless force estimation for industrial robots using disturbance observer and neural learning of fric- tion approximation,

    S. Liu, L. Wang, and X. V . Wang, “Sensorless force estimation for industrial robots using disturbance observer and neural learning of fric- tion approximation,”Robotics and Computer-Integrated Manufacturing, vol. 71, p. 102168, 2021

    2021

  6. [5]

    Collision detection and external force estimation for robot manipulators using a composite momentum observer,

    B. Ibari, M. Hebali, B. Rezali, and M. Bennaoum, “Collision detection and external force estimation for robot manipulators using a composite momentum observer,”AIMS Electronics and Electrical Engineering, vol. 8, no. 2, pp. 237–254, 2024

    2024

  7. [6]

    Toward sensorless interaction force estimation for industrial robots using high-order finite- time observers,

    L. Han, J. Mao, P. Cao, Y . Gan, and S. Li, “Toward sensorless interaction force estimation for industrial robots using high-order finite- time observers,”IEEE Transactions on Industrial Electronics, vol. 69, no. 7, pp. 7275–7284, 2021

    2021

  8. [7]

    Comprehensive modeling and identification of nonlinear joint dynamics for collaborative industrial robot manipulators,

    E. Madsen, O. S. Rosenlund, D. Brandt, and X. Zhang, “Comprehensive modeling and identification of nonlinear joint dynamics for collaborative industrial robot manipulators,”Control Engineering Practice, vol. 101, p. 104462, 2020

    2020

Show all 45 references
  1. [8]

    Friction identification and compensation in robotic manipulators,

    M. R. Kermani, R. V . Patel, and M. Moallem, “Friction identification and compensation in robotic manipulators,”IEEE Transactions on Instrumentation and Measurement, vol. 56, no. 6, pp. 2346–2353, 2007

    2007

  2. [9]

    Adaptive control based friction estimation for tracking control of robot manipulators,

    J. Huang, D. Tateo, P. Liu, and J. Peters, “Adaptive control based friction estimation for tracking control of robot manipulators,”IEEE Robotics and Automation Letters, 2025

    2025

  3. [10]

    Residual-based stiffness estimation in robots with flexible transmissions,

    F. Flacco and A. De Luca, “Residual-based stiffness estimation in robots with flexible transmissions,” in2011 IEEE international conference on robotics and automation. IEEE, 2011, pp. 5541–5547

    2011

  4. [11]

    A black- box physics-informed estimator based on gaussian process regression for robot inverse dynamics identification,

    G. Giacomuzzo, R. Carli, D. Romeres, and A. Dalla Libera, “A black- box physics-informed estimator based on gaussian process regression for robot inverse dynamics identification,”IEEE Transactions on Robotics, 2024

    2024

  5. [12]

    Kernelized movement primitives,

    Y . Huang, L. Rozo, J. Silv ´erio, and D. G. Caldwell, “Kernelized movement primitives,”The International Journal of Robotics Research, vol. 38, no. 7, pp. 833–852, 2019

    2019

  6. [13]

    Recursive noise adaptive kalman filtering by variational bayesian approximations,

    S. Sarkka and A. Nummenmaa, “Recursive noise adaptive kalman filtering by variational bayesian approximations,”IEEE Transactions on Automatic control, vol. 54, no. 3, pp. 596–600, 2009

    2009

  7. [14]

    Sensorless robot collision detection and hybrid force/motion control,

    A. De Luca and R. Mattone, “Sensorless robot collision detection and hybrid force/motion control,” inProceedings of the 2005 IEEE international conference on robotics and automation. IEEE, 2005, pp. 999–1004

    2005

  8. [15]

    Collision detection and safe reaction with the dlr-iii lightweight manipulator arm,

    A. De Luca, A. Albu-Schaffer, S. Haddadin, and G. Hirzinger, “Collision detection and safe reaction with the dlr-iii lightweight manipulator arm,” in2006 IEEE/RSJ international conference on intelligent robots and systems. IEEE, 2006, pp. 1623–1630

    2006

  9. [16]

    Review and comparison of dry friction force models,

    E. Pennestr `ı, V . Rossi, P. Salvini, and P. P. Valentini, “Review and comparison of dry friction force models,”Nonlinear dynamics, vol. 83, pp. 1785–1801, 2016

    2016

  10. [17]

    Revisiting the lugre friction model,

    K. Johanastrom and C. Canudas-De-Wit, “Revisiting the lugre friction model,”IEEE Control systems magazine, vol. 28, no. 6, pp. 101–114, 2008

    2008

  11. [18]

    Probabilistic latent variable modeling for dynamic friction identification and estimation,

    V . Vantilborgh, S. De Witte, F. Ostyn, T. Lefebvre, and G. Crevecoeur, “Probabilistic latent variable modeling for dynamic friction identification and estimation,”arXiv preprint arXiv:2412.15756, 2024

    2024 arXiv

  12. [19]

    Learning-based adaption of robotic friction models,

    P. Scholl, M. Iskandar, S. Wolf, J. Lee, A. Bacho, A. Dietrich, A. Albu- Sch¨affer, and G. Kutyniok, “Learning-based adaption of robotic friction models,”Robotics and Computer-Integrated Manufacturing, vol. 89, p. 102780, 2024

    2024

  13. [20]

    Modeling the static friction in a robot joint by genetically optimized bp neural network,

    X. Tu, Y . Zhou, P. Zhao, and X. Cheng, “Modeling the static friction in a robot joint by genetically optimized bp neural network,”Journal of Intelligent & Robotic Systems, vol. 94, pp. 29–41, 2019

    2019

  14. [21]

    Neural-learning-based force sensorless admittance control for robots with input deadzone,

    G. Peng, C. P. Chen, W. He, and C. Yang, “Neural-learning-based force sensorless admittance control for robots with input deadzone,”IEEE Transactions on Industrial Electronics, vol. 68, no. 6, pp. 5184–5196, 2020

    2020

  15. [22]

    Contact force estimation for robot manipulator using semiparametric model and disturbance kalman filter,

    J. Hu and R. Xiong, “Contact force estimation for robot manipulator using semiparametric model and disturbance kalman filter,”IEEE Trans- actions on Industrial Electronics, vol. 65, no. 4, pp. 3365–3375, 2017

    2017

  16. [23]

    Semi-parametric gaussian process for robot system identification,

    T. Wu and J. Movellan, “Semi-parametric gaussian process for robot system identification,” in2012 IEEE/RSJ International Conference on Intelligent Robots and Systems. IEEE, 2012, pp. 725–731

    2012

  17. [25]

    Uncertainty-aware imitation learning using kernelized movement prim- itives,

    J. Silv ´erio, Y . Huang, F. J. Abu-Dakka, L. Rozo, and D. G. Caldwell, “Uncertainty-aware imitation learning using kernelized movement prim- itives,” in2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2019, pp. 90–97

    2019

  18. [26]

    State-and context-dependent robotic manipulation and grasping via uncertainty-aware imitation learning,

    T. R. Winter, A. M. Sundaram, W. Friedl, M. A. Roa, F. Stulp, and J. Silv ´erio, “State-and context-dependent robotic manipulation and grasping via uncertainty-aware imitation learning,”arXiv preprint arXiv:2410.24035, 2024

    2024 arXiv

  19. [27]

    Hierarchical kernelized movement prim- itives for learning human-robot collaborative trajectories in referred object handover,

    K. Qian, Z. Yue, and J. Bai, “Hierarchical kernelized movement prim- itives for learning human-robot collaborative trajectories in referred object handover,”Applied Intelligence, vol. 55, no. 1, pp. 1–15, 2025

    2025

  20. [28]

    A kmp-based interactive learning approach for robot trajectory adaptation with obstacle avoid- ance,

    S. Xiao, X. Chen, Y . Lu, J. Ye, and H. Wu, “A kmp-based interactive learning approach for robot trajectory adaptation with obstacle avoid- ance,”Industrial Robot: the international journal of robotics research and application, vol. 51, no. 2, pp. 326–339, 2024

    2024

  21. [29]

    Adaptive sliding mode disturbance observer-based composite control with prescribed performance of space manipulators for target capturing,

    Y . Zhu, J. Qiao, and L. Guo, “Adaptive sliding mode disturbance observer-based composite control with prescribed performance of space manipulators for target capturing,”IEEE Transactions on Industrial Electronics, vol. 66, no. 3, pp. 1973–1983, 2018

    1973

  22. [30]

    A nonlinear disturbance observer for robotic manipulators,

    W.-H. Chen, D. J. Ballance, P. J. Gawthrop, and J. O’Reilly, “A nonlinear disturbance observer for robotic manipulators,”IEEE Transactions on industrial Electronics, vol. 47, no. 4, pp. 932–938, 2000

    2000

  23. [31]

    Collision detection, isolation and identification for humanoids,

    J. V orndamme, M. Schappler, and S. Haddadin, “Collision detection, isolation and identification for humanoids,” in2017 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2017, pp. 4754–4761

    2017

  24. [32]

    6d virtual sensor for wrench estimation in robotized interaction tasks exploiting extended kalman filter,

    L. Roveda, A. Bussolan, F. Braghin, and D. Piga, “6d virtual sensor for wrench estimation in robotized interaction tasks exploiting extended kalman filter,”Machines, vol. 8, no. 4, p. 67, 2020

    2020

  25. [33]

    M. W. Spong and M. Vidyasagar,Robot dynamics and control. John Wiley & Sons, 2008

    2008

  26. [34]

    Robot collisions: A survey on detection, isolation, and identification,

    S. Haddadin, A. De Luca, and A. Albu-Sch ¨affer, “Robot collisions: A survey on detection, isolation, and identification,”IEEE Transactions on Robotics, vol. 33, no. 6, pp. 1292–1312, 2017

    2017

  27. [35]

    Actuator failure detection and isolation using generalized momenta,

    A. De Luca and R. Mattone, “Actuator failure detection and isolation using generalized momenta,” in2003 IEEE international conference on robotics and automation (cat. No. 03CH37422), vol. 1. IEEE, 2003, pp. 634–639

    2003

  28. [36]

    A review on kalman filter models,

    M. Khodarahmi and V . Maihami, “A review on kalman filter models,” Archives of Computational Methods in Engineering, vol. 30, no. 1, pp. 727–747, 2023

    2023

  29. [37]

    Adaptive kalman filtering,

    S. C. Rutan, “Adaptive kalman filtering,”Analytical Chemistry, vol. 63, no. 22, pp. 1103A–1109A, 1991

    1991

  30. [38]

    Adaptive kalman filtering for ins/gps,

    A. Mohamed and K. Schwarz, “Adaptive kalman filtering for ins/gps,” Journal of geodesy, vol. 73, no. 4, pp. 193–203, 1999. 14

    1999

  31. [39]

    Dynamic identification of industrial robot based on nonlinear friction model and ls-sos algorithm,

    J. Dong, J. Xu, Q. Zhou, J. Zhu, and L. Yu, “Dynamic identification of industrial robot based on nonlinear friction model and ls-sos algorithm,” IEEE Transactions on Instrumentation and Measurement, vol. 70, pp. 1–12, 2021

    2021

  32. [40]

    O’Hagan and J

    A. O’Hagan and J. J. Forster,Kendall’s advanced theory of statistics, volume 2B: Bayesian inference. Arnold, 2004, vol. 2

    2004

  33. [41]

    Bayesian approach to extended object and cluster tracking using random matrices,

    J. W. Koch, “Bayesian approach to extended object and cluster tracking using random matrices,”IEEE Transactions on Aerospace and Electronic Systems, vol. 44, no. 3, pp. 1042–1059, 2008

    2008

  34. [42]

    A c++ implementation of a carte- sian impedance controller for robotic manipulators,

    M. Mayr and J. M. Salt-Ducaju, “A c++ implementation of a carte- sian impedance controller for robotic manipulators,”arXiv preprint arXiv:2212.11215, 2022

    2022 arXiv

  35. [43]

    End-effector force estimation for flexible-joint robots with global friction approximation using neural networks,

    X. Liu, F. Zhao, S. S. Ge, Y . Wu, and X. Mei, “End-effector force estimation for flexible-joint robots with global friction approximation using neural networks,”IEEE Transactions on Industrial Informatics, vol. 15, no. 3, pp. 1730–1741, 2018

    2018

  36. [44]

    A survey of feature selection methods for gaussian mixture models and hidden markov models,

    S. Adams and P. A. Beling, “A survey of feature selection methods for gaussian mixture models and hidden markov models,”Artificial Intelligence Review, vol. 52, pp. 1739–1779, 2019

    2019

  37. [45]

    Bayesian sparse gaussian mixture model for clustering in high dimensions,

    D. Yao, F. Xie, and Y . Xu, “Bayesian sparse gaussian mixture model for clustering in high dimensions,”Journal of Machine Learning Research, vol. 26, no. 21, pp. 1–50, 2025

    2025

Pith tools

Reviewed August 3, 2026 · model on record in the stance chip above.