Pith. sign in

REVIEW 3 major objections 9 minor 26 references

Anatomy of Uncertainty: Expressive Descriptors of Robotic Manipulator Motion for Non-verbal Communication in Human-Robot Collaboration

T0 review · 3 major / 9 minor · reviewed 2026-08-02 · deepseek-v4-flash

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

Pith's one-line read This paper aims to show that a robot's perceptual uncertainty can be decomposed into two independent axes—Commitment and Vigilance—and translated into a small set of trajectory descriptors that human viewers reliably read back from arm moti

arxiv 2607.13696 v1 pith:OQNCAJRF submitted 2026-07-15 cs.RO

Ridhima Bector , Souravik Dutta , Poornima Ramachandran , Ree Yan Yeoh , Jui Hien Tan , Domenico Campolo , Bernhard Johannes Schmitt This is my paper · ORCID
classification cs.RO
keywords human-robotcollaborationexpressiverobotmotionuncertaintycommunicationLabanMovementAnalysisCommitment-Vigilancestatespacedescriptorsperceptualuserstudynonverbal
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

Robots that move identically regardless of their inner uncertainty can look more confident than they are, which is a problem in shared workspaces. This paper tries to establish that uncertainty can be encoded in the timing and geometry of manipulator motion, and that people can decode those states from movement alone. The authors build a two-dimensional Commitment-Vigilance space, tie five uncertainty states (confidence, curiosity, hesitance, fear, inactivity) to distinct Laban Effort signatures, and parameterize trajectories with eleven computable descriptors. In a video study with 55 valid responses, viewers consistently matched the four tested trajectories to their intended states and, for several descriptors, rated intensity in the predicted directions. If correct, the result gives robots a parametric, perceptually grounded way to signal confidence and doubt without screens or speech.

What carries the argument

The central mechanism is the Commitment-Vigilance (C-V) state space, a two-dimensional latent map built from the four Laban Effort factors (Weight, Time, Space, Flow). Commitment measures how strongly motion is invested toward or against a goal; Vigilance measures how much motion is allocated to scanning, checking, and monitoring. This space bridges abstract uncertainty states and concrete trajectories: each of the five canonical states has a defined Effort signature, and each is implemented as a weighted combination of five primitives whose parameters are eleven scalar descriptors computed from end-effector poses and joint angles. The descriptors are what make the framework computable and t

What would settle it

A controlled study that generates trajectories algorithmically from the C-V equations without animator hand-tuning, showing each clip individually to naive viewers; if recognition falls to chance, the hand-crafted stimuli rather than the proposed mapping would be carrying the effect.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that perceptual uncertainty is not a single scalar to display but a structured behavioral state that can be mapped into motion along two independent axes. The Commitment axis, C = −c1W − c2T, combines strength and urgency of movement; the Vigilance axis, V = v1S − v2F, combines indirect spatial attention and bound flow. These define five canonical states—confident, curious, hesitant, fearful, and inactive—each realized through proportions of five motion primitives (approach, pause, retreat, exploration, oscillation) and measured through eleven scalar descriptors such as approach acceleration, pause count and duration, retreat count and distance, gaz

Load-bearing premise

The load-bearing premise is that the perceptual validation supports the C-V mapping and descriptor parameterization themselves, even though the tested trajectories were hand-designed by animators and presented four at a time rather than generated by the mapping and shown individually.

Editorial extensions

If this is right

  • A robot that estimates its own uncertainty could translate that estimate into motion by selecting descriptor values from the appropriate C-V region, producing a non-verbal, display-free signal during collaboration.
  • Perceived intensity can be scaled along known axes: shorter pauses and higher approach acceleration increase perceived confidence, while longer pauses, more retreats, slower approaches, and larger retreat distances increase perceived hesitance and fear.
  • The eleven descriptors give a common measurement language for expressive motion that is independent of the specific manipulator, since they are defined on end-effector trajectories and joint angles.
  • The state and descriptor structure is directly reusable for autonomous trajectory generation, for example through dynamic movement primitives parameterized by these descriptors, which the paper identifies as its planned next step.
  • The observed perceptual asymmetries imply that expressive uncertainty should be designed as coordinated multi-descriptor changes rather than a single kinematic tweak.

Reading between the lines

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

  • The free weights c1, c2, v1, v2 in the C-V equations are never fitted; an editor's inference is that the perceptual data could calibrate these weights, turning the model from a qualitative map into a predictive generative model—a testable next step the paper does not carry out.
  • Part A presented all four trajectories simultaneously in a continuous loop; an inference is that recognition may be partly relative rather than absolute, so a single-clip test would separate intrinsic legibility from comparison-driven labeling.
  • The fifth state, inactivity, was defined but not evaluated; a natural extension is to test whether a low-commitment, low-vigilance trajectory reads as disengagement rather than confusion or boredom.
  • The descriptors were defined for a goal-directed pre-grasp approach; the editor's inference is that the same C-V mapping may transfer to other joint tasks like handover or inspection, but descriptor meanings such as gaze angle would likely need re-grounding in the new task geometry.
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

3 major / 9 minor

Summary. This paper proposes a Commitment-Vigilance (C-V) behavioral model for expressing robot perceptual uncertainty through manipulator motion. The model projects four Laban Effort factors onto two latent dimensions via Eqs. (1)-(2), defines five canonical states (confidence, curiosity, hesitance, fear, inactivity) with Effort signatures in Table I, and introduces five motion primitives parameterized by eleven kinematic/geometric descriptors (Section III.B). A remote video study (N=55 after exclusions) tested recognition of four animator-designed base trajectories (Part A) and whether one-at-a-time descriptor variations alter perceived intensity (Part B). Part A shows significant non-uniform response distributions with a single dominant intended label for each trajectory; Part B shows significant intensity effects for several descriptors, while tilt velocity, shiver amplitude, and retreat acceleration produced no clear dominant attribution. The paper claims this establishes a perceptually grounded mathematical framework for encoding robot uncertainty in motion.

Significance. If the C-V framework were fully specified and validated as a generative model, it would be a valuable step beyond prior LMA-based affective mappings: it offers a compositional account connecting uncertainty states to measurable trajectory parameters, and the descriptor-level intensity data provide concrete design guidance. The availability of code, videos, questionnaire, and appendices is a strength for reproducibility. However, as presented, the empirical support attaches to the animator-designed trajectories and the descriptor vocabulary, not to the C-V equations as a generative mapping. The mathematical core is under-specified and internally inconsistent, so the current contribution is best described as a useful empirical descriptor study plus a promising but unproven conceptual framework.

major comments (3)
  1. [§III.A, Eqs. (1)-(2) and Table I] The proposed 'mathematical framework' is under-specified. The weighting coefficients c1, c2, v1, v2 are only said to be positive; no values or constraints are given. With arbitrary positive weights, C and V range over [-(c1+c2), c1+c2] and [-(v1+v2), v1+v2], not [-1,1], so the claim C,V∈[-1,1] requires c1+c2=v1+v2=1, which is never stated. Table I's canonical states (e.g., Confident at C=1,V=-1 with all Effort factors at ±1) implicitly impose exactly this normalization. The two dimensions are also asserted to be orthogonal without derivation. More seriously, Table I lists Hesitant as (C≈0,V≈0), while the composition paragraph in §III.B says 'Hesitance (C<0,V<0)'; the region enumeration also repeats '(C<0,V>0)' for both 'high vigilance' and 'low or no vigilance.' These inconsistencies and missing constraints must be fixed before the framework can be evaluated.
  2. [§IV and §VII] The perceptual validation does not test the C-V framework as a generative model. Section IV states that the four base trajectories were 'designed by animators who mapped robot's uncertainty to relevant expressive motion descriptors,' and Table II lists hand-set descriptor values. Neither Eqs. (1)-(2) nor the composition rules in §III.B are used to produce the stimuli; no inversion or optimization from a (C,V) target to descriptor values is given. The Limitations in §VII acknowledge this ('descriptor values and expressive trajectories were designed manually'). Therefore Part A validates the descriptor vocabulary and the specific hand-crafted trajectories, not the claim that the C-V framework is 'perceptually grounded' or that it can autonomously encode uncertainty states. The abstract and conclusions should be scoped accordingly, or the authors should instantiate the mapping (e.g., by spe
  3. [§IV.A, Part A protocol] Part A presented all four expressive videos simultaneously on a continuous loop. This measures the discriminability of the four trajectories when alternatives are visible, not recognition of a single motion in a naturalistic setting. The paper reports only chi-squared statistics and standardized residuals, not the percentage of participants selecting the intended label per trajectory when viewed alone. The hypothesis H1 states that trajectories 'will be perceived as expressing distinct behavioral states ... at rates significantly above chance level,' but the simultaneous-presentation design can inflate such rates through contrast. I request per-expression recognition rates, or a follow-up single-stimulus condition, and at minimum a discussion of how simultaneous presentation may affect the recognition claim.
minor comments (9)
  1. [Eq. (6)] The 'Approach Acceleration' definition differentiates the discrete vector difference g_{i+1}-g_i over an interval; please define it as the continuous-time peak magnitude of the end-effector acceleration during approach episodes, or clarify the discrete approximation.
  2. [Eq. (7)] Counting configurations with all joint velocities zero can overcount one pause depending on the sampling rate; clarify that a pause is a contiguous episode of zero-velocity configurations.
  3. [§III.B, descriptor definitions] Several descriptors (Pause Duration, Retreat Distance, Horizontal/Vertical Gaze) are defined with the assumption that 'all episodes cover equal value' in a given trajectory. State explicitly that this is a design constraint for the stimuli, not a general property of the definition.
  4. [Table II] For each behavior most descriptors are listed as 'N/A'; the selection of which descriptors are relevant to which behavior is stated only qualitatively. Explain the criterion, or at least note that untested descriptors were held at baseline values.
  5. [§III.B] The composition rules ('dominated by', 'secondary', 'tertiary') are qualitative. If the claim is that proportions 'follow directly from its C-V position,' provide the explicit mapping or soften the wording.
  6. [§III.A] The phrase 'C is anchored in Laban Near state, and Vigilance in Laban Remote state' introduces Near/Remote states that are not defined in the paper; define them or remove the reference.
  7. [§V.B] Report the degrees of freedom for the Part B chi-square tests, which are omitted, and state whether any correction for multiple comparisons was applied.
  8. [Table I / §III] Inactivity is included in the model but not evaluated in the user study; add a sentence explaining whether this state is intentionally deferred to future work.
  9. [General] Minor typographical and formatting issues: the duplicate region description '(C<0,V>0)' in §III.A, and the use of 'χ^2(5)' without degrees of freedom in some Part A sentences. These should be corrected in revision.

Circularity Check

1 steps flagged · score 4.0 of 10

C-V to Effort-signature mapping is definitional by Eqs. (1)-(2); perceptual study validates manually animated trajectories, not the generative framework.

  1. self definitional [Section III.A, Eqs. (1)-(2) and Table I; Section IV; Section VII]
    "C=−c1W−c2T; W,T∈[−1,+1] (1); V=v1S−v2F; S,F∈[−1,+1] (2) ... TABLE I: Canonical robot behavioral states along the uncertainty continuum and their Effort signatures derived from Commitment-Vigilance (C-V) space. Confident (C=1,V=−1) Strong(−1) Sudden(−1) Direct(−1) Free(+1) ... Four base trajectories corresponding to the four expressive behaviors were designed by animators ... it does not yet autonomously compute the descriptor values required for a robot’s current perceptual condition."

    The Effort signatures in Table I are not derived from the C-V states; they are selected so that substituting the listed W,T,S,F values into Eqs. (1)-(2) yields exactly the stated C,V coordinates at equal weighting (e.g., Confident gives C=1,V=−1; Curious gives C=1,V=1). Thus the claimed state-to-Effort-signature mapping is the inverse of the defining linear map and is true by construction, carrying no independent empirical content. Moreover, the perceptual validation used trajectories 'designed by animators,' not generated from this mapping, and the Limitations admit the model 'does not yet autonomously compute the descriptor values,' so the human-study results support the hand-designed descriptor vocabulary rather than the C-V generative framework.

full rationale

The paper contains no self-citation chain, no imported uniqueness theorem, and no fitted parameters renamed as predictions. The Part A/B human studies use 55 independent participants and the stimulus variations are not fitted to the model, so the descriptor-level perceptual findings have genuine external content. The nonzero circularity score comes from the paper's central C-V contribution: the mapping from behavioral states to Laban Effort signatures in Table I is constructed from Eqs. (1)-(2) rather than derived from data or from appraisal theory. For every state, the listed Effort factors are chosen to reproduce the stated C,V coordinates by definition. This is a mild but real self-definitional step at the core of the claimed 'mathematical framework.' Additionally, Section IV shows the tested trajectories were manually animated, and Section VII explicitly acknowledges the model does not yet compute descriptor values autonomously; therefore the perceptual validation does not test the C-V framework as a generative model. Internal inconsistencies (Section III.A says Hesitance is (C<0,V<0) while Table I lists (C≈0,V≈0), and one region description repeats (C<0,V>0)) are correctness issues, not circularity. Overall, the empirical descriptor work is independent, but the central C-V-to-Effort mapping reduces by construction, giving a score of 4.

Assumptions & free parameters 2 free parameters · 4 assumptions · 1 invented entities

The C-V framework is the main invented structure: two latent dimensions defined as linear combinations of Laban Effort factors, with unspecified coefficients and a hand-assigned state table. The descriptor values in Table II are manually chosen per trajectory. The empirical perception study is the only independent evidence, and it validates the hand-crafted motions rather than the mathematical mapping.

free parameters (2)
  • c1, c2, v1, v2 (C-V weighting coefficients) = unspecified
    Appear in Eqs. (1)-(2). Table I's state values imply sums c1+c2=1 and v1+v2=1 to give C=±1, V=±1, but no values or fitting are provided.
  • Per-trajectory descriptor values (Table II) = e.g., approach acceleration 0.05-0.1 m/s²; pause counts 1-5; retreat distance 0.06 m; shiver amplitude 6°
    Hand-selected by animators for each of the four base trajectories; not derived from the C-V model and not predicted by it.
assumptions (4)
  • domain assumption Laban Effort factors (Weight, Time, Space, Flow) encode distinct psychological functions relevant to uncertainty
    Basis for the C-V projection in Section III-A; relies on the LMA interpretive literature rather than on measurement or derivation.
  • domain assumption Commitment and Vigilance are orthogonal, independent dimensions
    Eqs. (1)-(2) treat C and V as separate axes with no interaction; no empirical or theoretical justification for orthogonality is given.
  • ad hoc to paper Five canonical states correspond to regions of C-V space and to the Effort signatures in Table I
    Table I assignments (e.g., confident = Strong/Sudden/Direct/Free) are asserted, not derived from Eqs. (1)-(2); with equal weights the C,V values do not uniquely determine these signatures.
  • domain assumption Uniform distribution across six response categories is an appropriate chance baseline for recognition
    Part A chi-square tests assume uniform guessing; no neutral baseline video was shown, so demand characteristics may inflate recognition.
invented entities (1)
  • Commitment-Vigilance (C-V) latent state space
    purpose: Two-dimensional model of robot uncertainty in motion; organizes five behavioral states and maps them to Laban Effort signatures
    No falsifiable prediction outside this paper; weights and state boundaries are unspecified, and trajectories used in validation were designed by hand, not generated from C-V coordinates.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Anatomy of Uncertainty: Expressive Descriptors of Robotic Manipulator Motion for Non-verbal Communication in Human-Robot Collaboration." pith.science (2026). https://pith.science/paper/OQNCAJRF

@misc{pith2026260713696,
  author       = {Pith},
  title        = {Pith review of: Anatomy of Uncertainty: Expressive Descriptors of Robotic Manipulator Motion for Non-verbal Communication in Human-Robot Collaboration},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/OQNCAJRF}},
  note         = {Machine review of arXiv:2607.13696}
}
read the original abstract

Robots operating in human-robot collaboration must communicate not only their intended actions but also uncertainty arising from incomplete or ambiguous perception. This work introduces a mathematical framework for expressing perceptual uncertainty through robotic manipulator motion. Drawing on Laban Movement Analysis, robot behavior is organized in a Commitment-Vigilance state space that maps uncertainty-related states - confidence, curiosity, hesitance, fear, and inactivity - to distinct Laban Effort signatures. Five motion primitives - approach, pause, retreat, exploration, and oscillation - are then parameterized using eleven kinematic and geometric descriptors, including acceleration, pause and retreat characteristics, gaze angles, tilt, and shiver amplitude. A video-based human-subject study evaluated recognition of four expressive trajectories and the influence of individual descriptors on perceived intensity. Participants reliably identified the intended behavioral states, while several descriptors significantly modulated expressiveness. The results establish a perceptually grounded basis for encoding robot uncertainty in motion and support future autonomous trajectory generation using parametric movement representations for collaborative tasks in shared environments. Code, videos, questionnaire and appendices are available at "https://bit.ly/github-aou".

Figures

Figures reproduced from arXiv: 2607.13696 by the authors.

Figure 1
Figure 1. Commitment-Vigilance state space with robot beha [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. A robotic manipulator in a goal-directed task. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Geometric parameters of robot end effector with [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Velocity profiles of the base trajectories of the robotic [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 6
Figure 6. Figure 6: Top 3 words used by participants to describe the [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 7
Figure 7. Figure 7: Behaviors assigned by participants for the descriptor [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 5
Figure 5. Figure 5: Words used by participants to describe the descriptor [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 8
Figure 8. Figure 8: Kinematic variables explored across the four mental [PITH_FULL_IMAGE:figures/full_fig_p009_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

26 extracted references

  1. [1]

    Planning and ac- ting in partially observable stochastic domains,

    L. P. Kaelbling, M. L. Littman, and A. R. Cassandra, “Planning and ac- ting in partially observable stochastic domains,”Artificial Intelligence, vol. 101, no. 1-2, pp. 99–134, May 1998

    1998

  2. [2]

    Active perception,

    R. Bajcsy, “Active perception,”Proceedings of the IEEE, vol. 76, no. 8, pp. 966–1005, 1988

    1988

  3. [3]

    Effects of nonverbal communication on efficiency and robustness in human- robot teamwork,

    C. Breazeal, C. Kidd, A. Thomaz, G. Hoffman, and M. Berlin, “Effects of nonverbal communication on efficiency and robustness in human- robot teamwork,” in2005 IEEE/RSJ International Conference on Intelligent Robots and Systems. IEEE, 2005, pp. 708–713

    2005

  4. [4]

    Expressing thought: improving robot readability with animation principles,

    L. Takayama, D. Dooley, and W. Ju, “Expressing thought: improving robot readability with animation principles,” inProceedings of the 6th international conference on Human-robot interaction, ser. HRI’11. ACM, Mar. 2011, pp. 69–76

    2011

  5. [5]

    Legibility and predictability of robot motion,

    A. D. Dragan, K. C. Lee, and S. S. Srinivasa, “Legibility and predictability of robot motion,” in2013 8th ACM/IEEE International Conference on Human-Robot Interaction (HRI). IEEE, Mar. 2013, pp. 301–308

    2013

  6. [6]

    Designing robots with movement in mind,

    G. Hoffman and W. Ju, “Designing robots with movement in mind,” Journal of Human-Robot Interaction, vol. 3, no. 1, p. 89, Mar. 2014

    2014

  7. [7]

    Body movement: Coping with the environment,

    J. L. Hanna, I. Bartenieff, and D. Lewis, “Body movement: Coping with the environment,”Ethnomusicology, vol. 27, no. 1, p. 124, Jan. 1983

    1983

  8. [8]

    Communicating affect via flight path exploring use of the laban effort system for designing affective locomotion paths,

    M. Sharma, D. Hildebrandt, G. Newman, J. E. Young, and R. Eski- cioglu, “Communicating affect via flight path exploring use of the laban effort system for designing affective locomotion paths,” in2013 8th ACM/IEEE International Conference on Human-Robot Interaction (HRI). IEEE, Mar. 2013, pp. 293–300

    2013

Show all 26 references
  1. [9]

    Expressive motion with x, y and theta: Laban effort features for mobile robots,

    H. Knight and R. Simmons, “Expressive motion with x, y and theta: Laban effort features for mobile robots,” inThe 23rd IEEE Interna- tional Symposium on Robot and Human Interactive Communication. IEEE, Aug. 2014, pp. 267–273

    2014

  2. [10]

    Humans and robotic arm: Laban movement theory to create emotional connection,

    C. La Viola, L. Fiorini, G. Mancioppi, J. Kim, and F. Cavallo, “Humans and robotic arm: Laban movement theory to create emotional connection,” in2022 31st IEEE International Conference on Robot and Human Interactive Communication (RO-MAN). IEEE, Aug. 2022, pp. 566–571

    2022

  3. [11]

    Ex- pressive robot motion timing,

    A. Zhou, D. Hadfield-Menell, A. Nagabandi, and A. D. Dragan, “Ex- pressive robot motion timing,” inProceedings of the 2017 ACM/IEEE International Conference on Human-Robot Interaction, ser. HRI ’17. ACM, Mar. 2017, pp. 22–31

    2017

  4. [12]

    K. R. Scherer,Appraisal Considered as a Process of Multilevel Sequential Checking. Oxford University PressNew York, NY , May 2001, pp. 92–120

    2001

  5. [13]

    The hierarchical model of approach-avoidance mo- tivation,

    A. J. Elliot, “The hierarchical model of approach-avoidance mo- tivation,”Motivation and Emotion, vol. 30, no. 2, pp. 111–116, 2006

    2006

  6. [14]

    Methods for expressing robot intent for human–robot collaboration in shared workspaces,

    G. Lemasurier, G. Bejerano, V . Albanese, J. Parrillo, H. A. Yanco, N. Amerson, R. Hetrick, and E. Phillips, “Methods for expressing robot intent for human–robot collaboration in shared workspaces,” ACM Transactions on Human-Robot Interaction, vol. 10, no. 4, pp. 1–27, 2021

    2021

  7. [15]

    S. J. Burton, A.-A. Samadani, R. Gorbet, and D. Kuli ´c,Laban Movement Analysis and Affective Movement Generation for Robots and Other Near-Living Creatures. Springer International Publishing, Nov. 2015, pp. 25–48

    2015

  8. [16]

    Expressing uncertainty in human- robot interaction,

    C. Bartneck and E. Moltchanova, “Expressing uncertainty in human- robot interaction,”PLOS ONE, vol. 15, no. 7, p. e0235361, 2020

    2020

  9. [17]

    Robot-assisted decision-making: Unveiling the role of uncertainty visualisation and embodiment,

    S. Sch ¨ombs, S. Pareek, J. Goncalves, and W. Johal, “Robot-assisted decision-making: Unveiling the role of uncertainty visualisation and embodiment,” inProceedings of the CHI Conference on Human Factors in Computing Systems, ser. CHI ’24. ACM, May 2024, pp. 1–16

    2024

  10. [18]

    Warning signals for poor performance improve human- robot interaction,

    R. Van den Brule, G. Bijlstra, R. Dotsch, P. Haselager, and D. H. J. Wigboldus, “Warning signals for poor performance improve human- robot interaction,”Journal of Human-Robot Interaction, vol. 5, no. 2, p. 69, 2016

    2016

  11. [19]

    Expressing robot incapability,

    M. Kwon, S. H. Huang, and A. D. Dragan, “Expressing robot incapability,” inProceedings of the 2018 ACM/IEEE International Conference on Human-Robot Interaction, ser. HRI ’18. ACM, Feb. 2018, pp. 87–95

    2018

  12. [20]

    Design and impact of hesitation gestures during human-robot resource conflicts,

    A. Moon, C. A. C. Parker, E. A. Croft, and H. F. M. Van der Loos, “Design and impact of hesitation gestures during human-robot resource conflicts,”Journal of Human-Robot Interaction, vol. 2, no. 3, 2013

    2013

  13. [21]

    Design of hesitation gestures for nonverbal human-robot negotiation of conflicts,

    A. Moon, M. Hashmi, H. F. M. V . D. Loos, E. A. Croft, and A. Billard, “Design of hesitation gestures for nonverbal human-robot negotiation of conflicts,”ACM Transactions on Human-Robot Interaction, vol. 10, no. 3, pp. 1–25, 2021

    2021

  14. [22]

    Employing laban shape for generating emotionally and functionally expressive trajectories in robotic manipulators,

    S. B. Raghu, C. Lohrmann, A. Bakshi, J. Kim, J. C. Herrera, B. Hayes, and A. Roncone, “Employing laban shape for generating emotionally and functionally expressive trajectories in robotic manipulators,” in 2025 34th IEEE International Conference on Robot and Human Interactive ...

    2025

  15. [23]

    The emote model for effort and shape,

    D. Chi, M. Costa, L. Zhao, and N. Badler, “The emote model for effort and shape,” inProceedings of the 27th annual conference on Computer graphics and interactive techniques - SIGGRAPH ’00, ser. SIGGRAPH ’00. ACM Press, 2000, pp. 173–182

    2000

  16. [24]

    How do we recognize emotion from movement? specific motor components contribute to the recognition of each emotion,

    A. Melzer, T. Shafir, and R. P. Tsachor, “How do we recognize emotion from movement? specific motor components contribute to the recognition of each emotion,”Frontiers in Psychology, vol. 10, 2019

    2019

  17. [25]

    It’s not what you do, it’s how you do it: Grounding uncertainty for a simple robot,

    J. Hough and D. Schlangen, “It’s not what you do, it’s how you do it: Grounding uncertainty for a simple robot,” inProceedings of the 2017 ACM/IEEE International Conference on Human-Robot Interaction, ser. HRI ’17. ACM, Mar. 2017, pp. 274–282

    2017

  18. [26]

    How to communicate robot motion intent: A scoping review,

    M. Pascher, U. Gruenefeld, S. Schneegass, and J. Gerken, “How to communicate robot motion intent: A scoping review,” inProceedings of the 2023 CHI Conference on Human Factors in Computing Systems, ser. CHI ’23. ACM, Apr. 2023, pp. 1–17. 11

    2023

Pith tools

Reviewed August 2, 2026 · model on record in the stance chip above.