pith. sign in

arxiv: 2604.07038 · v1 · submitted 2026-04-08 · 💻 cs.RO · q-bio.NC

Exploring the proprioceptive potential of joint receptors using a biomimetic robotic joint

Akihiro Miki , Shun Hasegawa , Sota Yuzaki , Yuta Sahara , Yoshimoto Ribayashi , Kento Kawaharazuka , Kei Okada This is my paper

Pith reviewed 2026-05-10 18:38 UTC · model grok-4.3

classification 💻 cs.RO q-bio.NC
keywords joint receptorsproprioceptionbiomimetic robotType I receptorssensory neuroscienceroboticsproprioceptive sensing
0
0 comments X

The pith

Type I joint receptors alone enable proprioceptive sensing with less than 2 degrees average error.

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

Traditional neuroscience holds that joint receptors act mainly as limit detectors at extreme angles, leaving muscle spindles as the primary source of joint position information. The authors built a biomimetic robotic joint to test whether Type I receptors, which fire during slow sustained movements, can provide usable position data on their own. Their experiments found average sensing errors below 2 degrees for both bending and twisting, implying that joint receptors may carry more of the proprioceptive load than textbooks suggest and that their weighting relative to muscle spindles could vary with development or species.

Core claim

We developed a biomimetic robotic joint replicating Type I joint receptors and found that these sensors alone supported proprioceptive estimation of joint angle with an average error of less than 2 degrees across bending and twisting motions. The work indicates that joint receptors contribute more substantially to continuous position sensing than the conventional limit-detector model allows and that the balance between receptor types in neural processing may differ across contexts.

What carries the argument

The biomimetic robotic joint that embeds sensors with response properties matched to biological Type I joint receptors.

If this is right

  • Joint receptors are likely involved in ongoing angle sensing rather than only boundary detection.
  • The balance of inputs from joint receptors versus muscle spindles may be tuned differently during development and across species.
  • Proprioceptive differences between joints such as the elbow and knee in certain neuropathies may trace to receptor distributions.
  • Biomimetic hardware can serve as a platform for testing sensory hypotheses that are difficult to isolate in living tissue.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Robotic limbs could incorporate joint-receptor-like sensors to improve natural feel and accuracy.
  • The method offers a way to explore how proprioceptive systems adapt after injury or during learning.
  • Similar receptor roles might be examined in other joints or animal models to test generality.

Load-bearing premise

The biomimetic joint's sensor responses and placements match those of actual Type I receptors in a biological joint capsule.

What would settle it

Recording Type I receptor activity from a living joint, decoding angle from those signals alone, and checking whether the error remains below 2 degrees.

Figures

Figures reproduced from arXiv: 2604.07038 by Akihiro Miki, Kei Okada, Kento Kawaharazuka, Shun Hasegawa, Sota Yuzaki, Yoshimoto Ribayashi, Yuta Sahara.

Figure 1
Figure 1. Figure 1: Overview of the conceptual framework and methodology of this study. This figure illustrates how joint receptors were anatomically and functionally modeled, implemented in a biomimetic joint with embedded strain gauge sensors, and evaluated using a neural network inspired by the dorsal column–medial lemniscal pathway. a, An anatomical diagram of the human body model and joint capsule is shown (adapted from2… view at source ↗
Figure 2
Figure 2. Figure 2: Design and implementation of the biomimetic joint used in this study. This figure illustrates how the biomimetic joint was designed with reference to anatomical structures, fabricated with embedded strain gauge sensors, and tested under different motion conditions (bending, twisting, and push–pull). a, Anatomical diagram of a biological joint. The joint capsule surrounds the joint formed by bone and cartil… view at source ↗
Figure 3
Figure 3. Figure 3: Data acquisition and modeling framework for proprioception estimation using the biomimetic joint. a, The process of moving the biomimetic joint to acquire data from 60 strain gauge sensors embedded in the joint capsule, functioning as joint receptors, and joint position and orientation data. b, Joint movement data is obtained using SLAM Camera (Intel RealSense T265), providing relative coordinates and orie… view at source ↗
Figure 4
Figure 4. Figure 4: Performance of proprioception estimation using the trained deep learning model. a-f, This graph shows the estimation of the joint’s coordinates and orientation generated using deep learning based on sensory receptor data collected from 60 strain gauge sensors embedded in the joint capsule. Due to the open-type ball joint loosely connected by soft tissues, the graph visualizes not only orientation(roll(a), … view at source ↗
Figure 5
Figure 5. Figure 5: Redundancy analysis of proprioception estimation under reduced sensor input with ten independent trials. This figure shows the mean and standard deviation of angular errors across ten independent trials for different sensor reduction ratios. Panels (a) and (b) correspond to pitch and roll, respectively. Error bars represent the standard deviation across trials for each reduction ratio, magnified tenfold fo… view at source ↗
Figure 6
Figure 6. Figure 6: Identification of key joint receptors for proprioception estimation. This figure highlights the sensory receptors that the deep learning model identified as most important during different joint movements, together with visualizations of their spatial distribution. Graphs showing key joint receptors important for the deep learning model during twisting (a), bending (b), and push-pull (c) movements near the… view at source ↗
Figure 7
Figure 7. Figure 7: Identification of disconnected strain gauge sensors. a, Plot of strain gauge sensors, including those with disconnections. For clarity, a selected subset of strain gauge sensors is plotted. b, Enlarged view of a portion of (a). The strain gauge sensor highlighted in orange within the dashed region is observed to be disconnected. 19/26 [PITH_FULL_IMAGE:figures/full_fig_p019_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Validation of deep learning–based proprioception estimation with an additional dataset. This figure shows the performance of the trained model when tested on a dataset collected independently but with the same procedure as in the Performance Verification of Joint Receptors Using Deep Learning section. The biomimetic joint is an open-type ball joint loosely connected by soft tissues, which permits not only … view at source ↗
Figure 9
Figure 9. Figure 9: Training process for redundancy analysis. This graph illustrates the training process used to examine the redundancy of proprioception acquired through joint receptors, as discussed in the Analysis of Redundancy in Joint Receptors section. The horizontal axis represents the number of training iterations (epochs), while the vertical axis represents the loss function. The reduction ratio is represented by a … view at source ↗
Figure 10
Figure 10. Figure 10: Redundancy analysis of proprioception estimation under reduced sensor input from a single trial. This figure illustrates the angular errors for pitch (a) and roll (b) in one trial. Error bars represent the within-trial standard deviation of errors without scaling. Because this is a single trial, the results are subject to stochastic variation inherent in deep learning, and no statistical analysis can be p… view at source ↗
Figure 11
Figure 11. Figure 11: Distribution of critical joint receptors in near-limit regions where twisting(roll) angles exceeded 20◦ . a-e, Bar graphs showing the results of five trials conducted in regions where twisting(roll) angles exceeded 20◦ in the Distribution of Critical Sensory Receptors Near Range of Motion Limits section. For each trial, SHAP values representing the contributions of all motion components (x, y, z, roll, pi… view at source ↗
Figure 12
Figure 12. Figure 12: Distribution of critical joint receptors in near-limit regions where bending(pitch or yaw) angles exceeded 70◦ . a–e, Bar graphs showing the results of five trials conducted in regions where bending(pitch or yaw) angles exceeded 70◦ in the Distribution of Critical Sensory Receptors Near Range of Motion Limits section. For each trial, SHAP values representing the contributions of all motion components (x, … view at source ↗
Figure 13
Figure 13. Figure 13: Distribution of critical joint receptors in near-limit regions where the absolute value of x-translation exceeded 2 mm. a–e, Bar graphs showing the results of five trials conducted in regions where the absolute value of x-translation exceeded 2 mm in the Distribution of Critical Sensory Receptors Near Range of Motion Limits section. For each trial, SHAP values representing the contributions of all motion … view at source ↗
read the original abstract

In neuroscience, joint receptors have traditionally been viewed as limit detectors, providing positional information only at extreme joint angles, while muscle spindles are considered the primary sensors of joint angle position. However, joint receptors are widely distributed throughout the joint capsule, and their full role in proprioception remains unclear. In this study, we specifically focused on mimicking Type I joint receptors, which respond to slow and sustained movements, and quantified their proprioceptive potential using a biomimetic joint developed with robotics technology. Results showed that Type I-like joint receptors alone enabled proprioceptive sensing with an average error of less than 2 degrees in both bending and twisting motions. These findings suggest that joint receptors may play a greater role in proprioception than previously recognized and that the relative contributions of muscle spindles and joint receptors are differentially weighted within neural networks during development and evolution. Furthermore, this work may prompt new discussions on the differential proprioceptive deficits observed between the elbows and knees in patients with hereditary sensory and autonomic neuropathy type III. Together, these findings highlight the potential of biomimetics-based robotic approaches for advancing interdisciplinary research bridging neuroscience, medicine, and robotics.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript describes the construction of a biomimetic robotic joint that replicates the slow/sustained response properties and capsule distribution of Type I joint receptors. Experiments on this hardware demonstrate that these receptors alone support proprioceptive estimation of joint angle with an average error below 2 degrees during controlled bending and twisting motions. The work concludes that joint receptors may contribute more substantially to proprioception than the traditional limit-detector view allows, with implications for neural weighting of sensory inputs and clinical observations in hereditary sensory neuropathy.

Significance. If the biomimetic model is shown to faithfully capture the relevant biological response characteristics, the result would provide direct hardware-based evidence that isolated Type I-like signals can yield usable proprioceptive information at low error. The robotic validation approach itself constitutes a strength, offering a controllable platform for testing receptor hypotheses that is independent of fitted parameters or self-referential definitions. The reported <2° error supplies a concrete, falsifiable metric that could stimulate further cross-disciplinary work at the neuroscience-robotics interface.

major comments (2)
  1. [Methods] Methods section: the abstract and results present a quantitative claim of average proprioceptive error <2° yet supply no information on sensor calibration procedures, motion-control protocols, statistical analysis methods, or controls for mechanical confounding factors (e.g., friction, hysteresis). These omissions prevent evaluation of whether the reported error is robust or artifactual.
  2. [Results] Results / Discussion: the central claim that Type I-like receptors alone suffice for proprioception rests on the untested assumption that the robotic joint accurately reproduces the response dynamics and spatial distribution of biological Type I receptors. No quantitative validation data (e.g., firing-rate curves, adaptation time constants, or capsule-location comparisons) are provided to support this replication fidelity.
minor comments (2)
  1. [Abstract] Abstract: the term 'proprioceptive sensing' is used without an explicit operational definition (e.g., whether it refers to absolute angle estimation or relative change detection).
  2. [Figures] Figure captions and text: ensure that all reported error values are accompanied by the number of trials, standard deviation, and motion range tested.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments, which identify key areas where additional detail will strengthen the manuscript. We respond to each major comment below and indicate planned revisions.

read point-by-point responses

  1. Referee: [Methods] Methods section: the abstract and results present a quantitative claim of average proprioceptive error <2° yet supply no information on sensor calibration procedures, motion-control protocols, statistical analysis methods, or controls for mechanical confounding factors (e.g., friction, hysteresis). These omissions prevent evaluation of whether the reported error is robust or artifactual.

    Authors: We agree that the Methods section requires substantially more detail to support evaluation of the reported error. In the revised manuscript we will expand this section to describe the sensor calibration procedure (reference angle measurements and repeatability testing), the motion-control protocols (specific angular velocities, ranges, and trial counts for bending and twisting), the statistical methods used to compute average error, and the controls applied for mechanical factors including friction reduction and bidirectional hysteresis assessment. These additions will allow readers to assess whether the <2° error is robust. revision: yes

  2. Referee: [Results] Results / Discussion: the central claim that Type I-like receptors alone suffice for proprioception rests on the untested assumption that the robotic joint accurately reproduces the response dynamics and spatial distribution of biological Type I receptors. No quantitative validation data (e.g., firing-rate curves, adaptation time constants, or capsule-location comparisons) are provided to support this replication fidelity.

    Authors: The biomimetic joint was constructed to match the slow/sustained response properties and capsule distribution of Type I receptors as characterized in the existing neuroscience literature. We acknowledge that the original submission did not include quantitative comparisons such as firing-rate curves or adaptation time constants. In the revision we will add explicit description in the Methods of how sensor response thresholds and adaptation profiles were tuned to literature values, together with a clearer statement of the limitations of the biomimetic model and the absence of direct quantitative biological validation data. This will make the basis for the replication assumption transparent while noting the need for future cross-validation. revision: partial

Circularity Check

0 steps flagged

No significant circularity; experimental results from hardware measurements

full rationale

The paper describes construction of a biomimetic robotic joint to replicate Type I receptor response properties and distribution, followed by direct experimental quantification of proprioceptive error (<2° average) during controlled bending and twisting motions. The central claim is an empirical outcome of sensor outputs under physical testing, with no equations, parameter fits, self-definitions, or derivations that reduce to inputs by construction. No self-citations are invoked as load-bearing uniqueness theorems, and the methodology does not rename known results or smuggle ansatzes. This is a self-contained hardware validation study whose logic chain terminates in measured data rather than circular reduction.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim depends on the untested premise that the robotic hardware faithfully reproduces biological receptor dynamics; this is a domain assumption rather than a derived result.

axioms (1)
  • domain assumption The biomimetic robotic joint accurately replicates the response properties of biological Type I joint receptors to slow and sustained movements.
    This assumption is required for the experimental results to inform biological proprioception; it is stated implicitly by the study design.

pith-pipeline@v0.9.0 · 5524 in / 1197 out tokens · 42861 ms · 2026-05-10T18:38:09.407572+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

50 extracted references · 50 canonical work pages · 1 internal anchor

  1. [1]

    M., Pincivero, D

    Lephart, S. M., Pincivero, D. M., Giraido, J. L. & Fu, F. H. The role of proprioception in the management and rehabilitation of athletic injuries.The Am. journal sports medicine25, 130–137 (1997). 3.Kandel, E. R.et al. Principles of neural science, vol. 4 (McGraw-hill New York, 2000)

    work page 1997

  2. [2]

    Riemann, B. L. & Lephart, S. M. The sensorimotor system, part i: the physiologic basis of functional joint stability.J. athletic training37, 71 (2002). 5.Ralphs, J. & Benjamin, M. The joint capsule: structure, composition, ageing and disease.J. anatomy184, 503 (1994)

    work page 2002

  3. [3]

    Williams, W. J. A systems-oriented evaluation of the role of joint receptors and other afferents in position and motion sense.Critical Rev. Biomed. Eng.7, 23–77 (1981). 7.Grigg, P., Finerman, G. A. & Riley, L. H. Joint-position sense after total hip replacement.JBJS55, 1016–1025 (1973)

    work page 1981

  4. [4]

    & Larson, G

    Clark, F., Horch, K., Bach, S. & Larson, G. Contributions of cutaneous and joint receptors to static knee-position sense in man.J. Neurophysiol.42, 877–888 (1979)

    work page 1979

  5. [5]

    & Noordenbos, W

    Wall, P. & Noordenbos, W. Sensory functions which remain in man after complete transection of dorsal columns.Brain 100, 641–653 (1977). 15/26

    work page 1977

  6. [6]

    D., Kirkpatrick, J

    Ross, E. D., Kirkpatrick, J. B. & Lastimosa, A. C. Position and vibration sensations: functions of the dorsal spinocerebellar tracts?Annals Neurol. Off. J. Am. Neurol. Assoc. Child Neurol. Soc.5, 171–176 (1979)

    work page 1979

  7. [7]

    & Gandevia, S

    Proske, U. & Gandevia, S. C. The proprioceptive senses: Their roles in signaling body shape, body position and movement, and muscle force.Physiol. Rev.92, 1651–1697 (2012)

    work page 2012

  8. [8]

    M., McCloskey, D

    Goodwin, G. M., McCloskey, D. & Matthews, P. The contribution of muscle afferents to keslesthesia shown by vibration induced illusionsof movement and by the effects of paralysing joint afferents.Brain95, 705–748 (1972)

    work page 1972

  9. [9]

    Position sense and state of contraction; the effects of vibration.J

    Eklund, G. Position sense and state of contraction; the effects of vibration.J. Neurol. Neurosurg. & Psychiatry35, 606–611 (1972). 14.Proske, U. & Gandevia, S. C. Kinesthetic senses.Compr. Physiol.8, 1157–1183 (2011)

    work page 1972

  10. [10]

    Bewick, G. S. & Banks, R. W. Mechanotransduction in the muscle spindle.Pflügers Arch. J. Physiol.467, 175–190 (2015)

    work page 2015

  11. [11]

    Golgi tendon organs in mammalian skeletal muscle: functional properties and central actions.Physiol

    Jami, L. Golgi tendon organs in mammalian skeletal muscle: functional properties and central actions.Physiol. reviews72, 623–666 (1992)

    work page 1992

  12. [12]

    Johansson, R. S. & Vallbo, A. B. Tactile sensibility in the human hand: relative and absolute densities of four types of mechanoreceptive units in glabrous skin.The J. physiology286, 283–300 (1979)

    work page 1979

  13. [13]

    Johnson, K. O. The roles and functions of cutaneous mechanoreceptors.Curr. opinion neurobiology11, 455–461 (2001)

    work page 2001

  14. [14]

    W., Smirnova, I

    Witherspoon, J. W., Smirnova, I. V . & McIff, T. E. Neuroanatomical distribution of mechanoreceptors in the human cadaveric shoulder capsule and labrum.J. anatomy225, 337–345 (2014)

    work page 2014

  15. [15]

    & Elb.11, 300–304 (2019)

    Kholinne, E.et al.Neuroanatomical distribution of sensory receptors in the human elbow joint capsule.Should. & Elb.11, 300–304 (2019). 21.Proske, U. A reassessment of the role of joint receptors in human position sense.Exp. Brain Res.241, 943–949 (2023)

    work page 2019

  16. [16]

    Axelrod, F. B. & Gold-von Simson, G. Hereditary sensory and autonomic neuropathies: types ii, iii, and iv.Orphanet journal rare diseases2, 1–12 (2007)

    work page 2007

  17. [17]

    Macefield, V . G. The roles of mechanoreceptors in muscle and skin in human proprioception.Curr. Opin. Physiol.21, 48–56 (2021)

    work page 2021

  18. [18]

    G., Smith, L

    Macefield, V . G., Smith, L. J., Norcliffe-Kaufmann, L., Palma, J.-A. & Kaufmann, H. Sensorimotor control in the congenital absence of functional muscle spindles.Exp. Physiol.109, 27–34 (2024)

    work page 2024

  19. [19]

    Betts, J. G.et al. Anatomy and Physiology(OpenStax, Houston, Texas, 2013). Accessed: 2025-07-08. See also section 12.1: https://openstax.org/books/anatomy-and-physiology/pages/12-1-basic-structure-and-function-of-the-nervous-system

    work page 2013

  20. [20]

    Betts, J. G.et al. Anatomy and Physiology(OpenStax, Houston, Texas, 2013). Accessed: 2025-07-08. See also section 9.6: https://openstax.org/books/anatomy-and-physiology/pages/9-6-anatomy-of-selected-synovial-joints

    work page 2013

  21. [21]

    Betts, J. G.et al. Anatomy and Physiology(OpenStax, Houston, Texas, 2013). Accessed: 2025-07-08. See also section 14.2: https://openstax.org/books/anatomy-and-physiology/pages/14-2-central-processing

    work page 2013

  22. [22]

    Kholinne, E.et al.Topography of sensory receptors within the human glenohumeral joint capsule.J. Should. Elb. Surg.30, 779–786 (2021)

    work page 2021

  23. [23]

    Betts, J. G.et al. Anatomy and Physiology(OpenStax, Houston, Texas, 2013). Accessed: 2025-07-08. See also section 9.4: https://openstax.org/books/anatomy-and-physiology/pages/9-4-synovial-joints

    work page 2013

  24. [24]

    & Rendón-Mancha, J

    Fuentes-Pacheco, J., Ruiz-Ascencio, J. & Rendón-Mancha, J. M. Visual simultaneous localization and mapping: a survey. Artif. intelligence review43, 55–81 (2015)

    work page 2015

  25. [25]

    & Ahmed, A

    Schmidt, P., Scaife, J., Harville, M., Liman, S. & Ahmed, A. Intel® realsense™ tracking camera t265 and intel® realsense™ depth camera d435-tracking and depth.Real Sense(2019)

    work page 2019

  26. [26]

    G., Jeffers, J

    Ng, K. G., Jeffers, J. R. & Beaulé, P. E. Hip joint capsular anatomy, mechanics, and surgical management.JBJS101, 2141–2151 (2019). 33.Kawai, M., Nie, M., Oda, H., Morimoto, Y . & Takeuchi, S. Living skin on a robot.Matter5, 2190–2208 (2022)

    work page 2019

  27. [27]

    & Shi, Y

    Zhang, L., Song, B., Choi, S.-K., Yao, Y . & Shi, Y . Anisotropy-inspired, simulation-guided design and 3d printing of microlattice metamaterials with tailored mechanical-transport performances.Compos. Part B: Eng.236, 109837 (2022)

    work page 2022

  28. [28]

    & Liao, W

    Bodaghi, M., Damanpack, A., Hu, G. & Liao, W. Large deformations of soft metamaterials fabricated by 3d printing. Mater. & Des.131, 81–91 (2017). 16/26

    work page 2017

  29. [29]

    J.et al.Multi-material 3d printed mechanical metamaterials: Rational design of elastic properties through spatial distribution of hard and soft phases.Appl

    Mirzaali, M. J.et al.Multi-material 3d printed mechanical metamaterials: Rational design of elastic properties through spatial distribution of hard and soft phases.Appl. Phys. Lett.113(2018)

    work page 2018

  30. [30]

    Goble, D. J. Proprioceptive acuity assessment via joint position matching: from basic science to general practice.Phys. therapy90, 1176–1184 (2010)

    work page 2010

  31. [31]

    G.et al.Increasing cutaneous afferent feedback improves proprioceptive accuracy at the knee in patients with sensory ataxia.J

    Macefield, V . G.et al.Increasing cutaneous afferent feedback improves proprioceptive accuracy at the knee in patients with sensory ataxia.J. Neurophysiol.115, 711–716 (2016)

    work page 2016

  32. [32]

    & Macefield, V

    Smith, L., Norcliffe-Kaufmann, L., Palma, J.-A., Kaufmann, H. & Macefield, V . G. Elbow proprioception is normal in patients with a congenital absence of functional muscle spindles.The J. Physiol.598, 3521–3529 (2020). 40.Hochreiter, S. & Schmidhuber, J. Long short-term memory.Neural computation9, 1735–1780 (1997)

    work page 2020

  33. [33]

    In Guyon, I.et al.(eds.)Advances in Neural Information Processing Systems, vol

    Vaswani, A.et al.Attention is all you need. In Guyon, I.et al.(eds.)Advances in Neural Information Processing Systems, vol. 30 (Curran Associates, Inc., 2017)

    work page 2017

  34. [34]

    B., Tan, R.-S., Ciaccio, E

    Yildirim, O., Baloglu, U. B., Tan, R.-S., Ciaccio, E. J. & Acharya, U. R. A new approach for arrhythmia classification using deep coded features and lstm networks.Comput. methods programs biomedicine176, 121–133 (2019)

    work page 2019

  35. [35]

    & Zhou, H

    Sun, J., Xie, J. & Zhou, H. Eeg classification with transformer-based models. In2021 IEEE 3rd Global Conference on Life Sciences and Technologies (LifeTech), 92–93 (2021)

    work page 2021

  36. [36]

    G., Norcliffe-Kaufmann, L., Gutierrez, J., Axelrod, F

    Macefield, V . G., Norcliffe-Kaufmann, L., Gutierrez, J., Axelrod, F. B. & Kaufmann, H. Can loss of muscle spindle afferents explain the ataxic gait in riley–day syndrome?Brain134, 3198–3208 (2011)

    work page 2011

  37. [37]

    G., Norcliffe-Kaufmann, L

    Macefield, V . G., Norcliffe-Kaufmann, L. J., Axelrod, F. B. & Kaufmann, H. Relationship between proprioception at the knee joint and gait ataxia in hsan iii.Mov. Disord.28, 823–827 (2013)

    work page 2013

  38. [38]

    J., Norcliffe-Kaufmann, L., Palma, J.-A., Kaufmann, H

    Smith, L. J., Norcliffe-Kaufmann, L., Palma, J.-A., Kaufmann, H. & Macefield, V . G. Impaired sensorimotor control of the hand in congenital absence of functional muscle spindles.J. Neurophysiol.120, 2788–2795 (2018)

    work page 2018

  39. [39]

    & Schomburg, E

    Lundberg, A., Malmgren, K. & Schomburg, E. Role of joint afferents in motor control exemplified by effects on reflex pathways from ib afferents.The J. physiology284, 327–343 (1978)

    work page 1978

  40. [40]

    Zimny, M. L. & Wink, C. S. Neuroreceptors in the tissues of the knee joint.J. Electromyogr. Kinesiol.1, 148–157 (1991)

    work page 1991

  41. [41]

    Riemann, B. L. & Lephart, S. M. The sensorimotor system, part ii: the role of proprioception in motor control and functional joint stability.J. athletic training37, 80 (2002). 50.Arduino, S. A. Arduino.Arduino LLC372(2015). 51.Banzi, M. & Shiloh, M.Getting started with Arduino(Maker Media, Inc., 2022)

    work page 2002

  42. [42]

    & Faigl, J

    Bayer, J. & Faigl, J. On autonomous spatial exploration with small hexapod walking robot using tracking camera intel realsense t265. In2019 European Conference on Mobile Robots (ECMR), 1–6 (2019)

    work page 2019

  43. [43]

    InICRA workshop on open source software, vol

    Quigley, M.et al.Ros: an open-source robot operating system. InICRA workshop on open source software, vol. 3, 5 (Kobe, Japan, 2009)

    work page 2009

  44. [44]

    & Kheddar, A

    Chappellet, K., Caron, G., Kanehiro, F., Sakurada, K. & Kheddar, A. Benchmarking cameras for open vslam indoors. In 2020 25th International Conference on Pattern Recognition (ICPR), 4857–4864 (2021)

    work page 2020

  45. [45]

    B., Daumer, M

    Hausamann, P., Sinnott, C. B., Daumer, M. & MacNeilage, P. R. Evaluation of the intel realsense t265 for tracking natural human head motion.Sci. reports11, 12486 (2021)

    work page 2021

  46. [46]

    Adam: A Method for Stochastic Optimization

    Paszke, A.et al.Pytorch: An imperative style, high-performance deep learning library.Adv. neural information processing systems32(2019). 57.Kingma, D. P. Adam: A method for stochastic optimization.arXiv preprint arXiv:1412.6980(2014). 58.Breiman, L. Random forests.Mach. learning45, 5–32 (2001)

    work page internal anchor Pith review Pith/arXiv arXiv 2019

  47. [47]

    & Lengauer, T

    Altmann, A., Tolo¸ si, L., Sander, O. & Lengauer, T. Permutation importance: a corrected feature importance measure. Bioinformatics26, 1340–1347 (2010)

    work page 2010

  48. [48]

    & Dominici, F

    Fisher, A., Rudin, C. & Dominici, F. All models are wrong, but many are useful: Learning a variable’s importance by studying an entire class of prediction models simultaneously.J. Mach. Learn. Res.20, 1–81 (2019). 61.Pedregosa, F.et al.Scikit-learn: Machine learning in python.J. machine Learn. research12, 2825–2830 (2011)

    work page 2019

  49. [49]

    Lundberg, S. M. & Lee, S.-I. A unified approach to interpreting model predictions. InAdvances in Neural Information Processing Systems, vol. 30, 4765–4774 (2017). 17/26

    work page 2017

  50. [50]

    why should i trust you?

    Ribeiro, M. T., Singh, S. & Guestrin, C. " why should i trust you?" explaining the predictions of any classifier. In Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining, 1135–1144 (2016). Supplementary information Supplementary Methods: Feature Analysis Methods of Deep Learning Models Several methods have bee...

    work page 2016