Pith. sign in

REVIEW 4 major objections 6 minor 36 references

Robustifying Long-term Human-Robot Collaboration through a Multimodal and Hierarchical Framework

T0 review · 4 major / 6 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read The paper claims that integrating vision and speech with hierarchical plan prediction over a shared task graph makes long-term human-robot collaboration more successful, faster, and more satisfying.

desk verdict A useful system-integration paper with a real robot study, but the headline numbers are misreported and the statistics need work before this is publishable as-is. read the letter →

arxiv 2411.15711 v2 pith:DN567AHV submitted 2024-11-24 cs.RO

Peiqi Yu , Abulikemu Abuduweili , Ruixuan Liu , Changliu Liu This is my paper · ORCID
classification cs.RO
keywords long-termhuman-robotcollaborationmultimodalperceptionhierarchicalplanpredictiontaskgraphhumanposedetectionspeechrecognitionassemblyonlineadaptation
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper is trying to establish that long-term human-robot collaboration can be made substantially more reliable by combining multimodal perception (video plus speech) with hierarchical plan prediction over a structured task graph. The authors report that this integrated framework reaches a 91.79% task success rate in a real-world toy-car assembly user study, cuts average completion time by 15.85% relative to ablation groups without both components, and earns an overall user satisfaction score of 84%. A sympathetic reader should care because long-horizon collaboration is where robots currently fail: uncertainty compounds over many subtasks, and the paper's claim is that two-level plan prediction plus speech correction prevents those errors from cascading.

What carries the argument

The central object is the hierarchical task graph: a directed And-Or graph whose nodes are subtasks and whose edges encode which orders are valid, with each subtask represented as a small temporal plan graph of human and robot actions. The load-bearing mechanism is the plan-prediction module, which uses dynamic time warping to align the recorded task progress to the closest reference node sequence in that graph (Eq. 4) and then maps the current detected action onto the most likely next action within that sequence (Eq. 5). Around this sit the supporting mechanisms: multimodal fusion of visual pose and speech commands, hierarchical pose detection that crops to the nearest human, sensitivity-aware action classification that converts dangerous misclassification errors into safer ones, and an online adaptation rule that tunes the trajectory predictor to each user.

What would settle it

Run the same toy-car assembly with a participant instructed to follow an assembly order that is deliberately absent from the task graph, or place the tubes in a new location, and check whether plan predictions and task success stay at the reported 91.79% level; a significant drop would confirm that the manual-graph and fixed-object assumptions carry the stated result.

Watch

Extended reading notes

Core claim

The paper's central claim is that robustness in long-term human-robot collaboration comes from structuring the whole task as a hierarchical task graph and giving the robot two complementary ways to stay aligned with the human: multimodal perception that lets a spoken command confirm or override a visual interpretation, and hierarchical plan prediction that first matches observed task progress to a reference node sequence in the graph using dynamic time warping (Eq. 4) and then refines the predicted action to the most probable next action in that sequence (Eq. 5). In the physical assembly experiments, the full system outperforms every ablation on success rate, completion time, and user satisfaction; the vision-only condition degrades sharply when users may choose among plans, and adding hierarchical plan prediction to multimodal perception raises success by 14.22%. The authors also show that hierarchical pose detection, which focuses on the closest human and ignores others, lowers keypoint deviation in multi-human scenes and raises action-prediction accuracy from about 51% without hierarchy to about 96% with it.

Load-bearing premise

The whole prediction chain assumes a complete, hand-written task graph that contains every plan the human might follow, along with fixed, known object locations; if a user follows a route the graph does not include, the robot has no reference sequence that matches and the prediction has no correct answer to find.

Editorial extensions

If this is right

  • Speech commands give the robot a way to recover from a wrong visual guess mid-task, which is what keeps user plan satisfaction high when plans are flexible.
  • Proactive, prediction-driven assistance translates directly into time savings, with the full system completing tasks on average 15.85% faster than groups lacking both components.
  • Hierarchical pose detection by proximity is what makes action prediction accurate in multi-human scenes, jumping from roughly 51% to 96% accuracy.
  • Online adaptation lowers trajectory-prediction error by about 24% for new users, so the framework does not require retraining per person.
  • Users report that the system lets them focus on the assembly task itself and makes them more willing to keep collaborating for extended periods.

Reading between the lines

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

  • Because the task graph is hand-built per scenario, the framework's scalability bottleneck is graph construction; automating that construction from demonstrations or natural-language instructions is the natural next step.
  • The DTW-based plan alignment should transfer to any long-horizon collaborative domain (cooking, kitting, maintenance) where a task graph exists, which means the method's generality is bounded by graph availability rather than by assembly specifically.
  • The fixed-object assumption means the robustness claim has not yet been tested against clutter or displacement; an object-tracking extension would be the direct stress test of the framework's real-world applicability.
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

4 major / 6 minor

Summary. The paper proposes a multimodal and hierarchical framework for long-term human-robot collaboration. The task is modeled as a hierarchical task graph known to both human and robot; the framework integrates RGB-D vision with speech commands, a hierarchical pose detection module, a hierarchical plan predictor based on DTW alignment to reference sequences, and an online adaptation mechanism for physical action prediction. The system is deployed on a KINOVA Gen3 arm in a toy-car assembly task, and a user study with 10 participants, 6 conditions, and 720 trials is used to evaluate success rate, completion time, adaptability, and user satisfaction. The authors report that the full system achieves a 91.79% average success rate, reduces task completion time by 15.9%, improves pose MSE by 24%, and yields higher user satisfaction, concluding that the framework robustly improves long-term HRC.

Significance. If the reported results are correct, this work is a useful integration of multimodal perception, hierarchical plan prediction, and online adaptation in a physically deployed long-horizon assembly task. The authors provide a real-robot user study with an ablation structure across perception modality and plan prediction, as well as open-source code and a demo video, which are valuable for reproducibility. However, the quantitative evidence is undermined by a systematic confusion between absolute percentage-point differences and relative percentage improvements, and by the complete absence of significance tests or confidence intervals. Because the headline numbers in the abstract and conclusion rest on these statistics, the current form does not support the strength of the claims. The central direction of the effect—full system outperforming ablations—is plausible, but the magnitude and robustness need re-analysis.

major comments (4)
  1. [Sec. V-C3, V-C4 and V-C1] The reported improvements are absolute differences presented as relative percentages. In Sec. V-C3, the text states an 'average reduction of 15.85%' for the best group (PP=1, PM=2) versus groups lacking both components (PP=0, PM=0&1). Using Table IV, the baseline mean is (78.80+86.80+53.19)/3 for PP=0/PM=0 and (80.05+79.34+62.33)/3 for PP=0/PM=1, giving a pooled baseline of 73.42; the best group mean is (67.42+65.83+39.45)/3 = 57.57. The relative reduction is (73.42-57.57)/73.42 = 21.6%, not 15.85%, which is exactly the absolute percentage-point gap. Similarly, in Sec. V-C4, the pose MSE reduction from 1.2574 to 1.017 is a relative decrease of 19.1%, not the reported 24%; the absolute difference 0.2404 was rounded to 24%. The '14.22%' success-rate increase in Sec. V-C1 for PP=1/PM=2 versus PP=0/PM=2 is also an absolute difference of 14.2 percentage points, not a relative improvement, which would be about 18.3%. All such comparisons must be recomputed and explicitly labeled as relative percentages or percentage points.
  2. [Sec. V-C1 through V-C5] The manuscript repeatedly claims 'significant' improvements (e.g., 'significantly reduces task completion time', 'significantly improve action prediction accuracy'), but no significance tests, confidence intervals, or effect sizes are reported for any user-study comparison. Tables II and IV provide means and standard deviations over 40 trials per cell, yet these trials are nested within 10 participants and are not independent; an appropriate repeated-measures or mixed-effects analysis is required. Many pairwise differences are small relative to the reported standard deviations (for example, Task 1 success rates: 74.45±17.15 vs 70.63±18.65), so without statistical evidence the claim that the integrated framework robustly improves long-term HRC is not verifiable. In addition, the multiple comparisons across six groups and three tasks need correction for multiplicity.
  3. [Sec. V-B3] The user study instructs participants to perform each task 'using any plan' and does not report the distribution of plans chosen in each condition. Because the framework's success depends on predicting plans over the manually designed task graph, systematic differences in plan selection across conditions (for instance, if participants in the full-system condition happened to choose easier plans) would confound all group comparisons. The paper should report plan frequencies per condition and, ideally, analyze outcomes per plan or include plan as a covariate in the statistical model.
  4. [Sec. IV-E and Sec. V-C] The central novelty of the paper is hierarchical plan prediction (Eq. 4-5), but no direct metric for plan prediction accuracy is reported. The offline evaluation in Sec. V-C4 measures physical action prediction (intention accuracy and pose MSE), and the user study reports only downstream task success, completion time, and satisfaction. Because the plan prediction module is the key contribution, an evaluation of how often the predicted reference sequence R* matches the human's actual plan, and of how DTW alignment errors affect downstream performance, is necessary to support the claim that hierarchical plan prediction significantly enhances the system.
minor comments (6)
  1. [Appendix A] The proof of the mutual-information inequality contains an algebraic error: H(g,OV)-H(g,OV,OA)+H(OA|OV) is not equal to H(g,OV|OA)+H(OA|OV). The inequality itself is true because conditioning on additional variables cannot increase entropy, but the derivation should be corrected.
  2. [Sec. IV-A] Typo: 'multual information' should be 'mutual information'.
  3. [Fig. 7] The text reports a variance of 0.013 for the Kalman-filtered naive detection but does not report the corresponding variance for the proposed hierarchical method; please add a numerical comparison and error bars for all methods.
  4. [Sec. V-C5] The claims that plan prediction improves user plan satisfaction by 47.5% and multimodal perception by 27.9% should clarify whether these are relative increases or percentage-point differences; the same absolute-vs-relative ambiguity as in Sec. V-C3 appears here.
  5. [Sec. V-B3] For clarity, specify the order of sessions and whether participants had prior exposure to the full system; also state how the 'any plan' choices were elicited and recorded.
  6. [Sec. VI] Typos: 'task progree' should be 'task progress'; 'in future research' should be 'in future work'.

Circularity Check

0 steps flagged · score 2.0 of 10

No load-bearing circularity: core claims are empirical comparisons against baseline user-study groups; self-citations supply component methods that are independently tested, and the DTW plan-prediction equations are model-based template matching rather than a self-referential derivation.

full rationale

The paper's central claims (task success, completion time, user satisfaction) are evaluated in a six-group user study (Sec. V-B3, Tables II and IV, Figs. 9-10) rather than derived from fitted parameters or from cited prior work. The hierarchical plan predictor (Eqs. 4-5) selects the closest manually enumerated reference sequence in the task graph via DTW; this is a hand-designed modeling choice that constrains generalizability (acknowledged in Sec. VI), but it is not circular because the predicted plan is not the quantity being optimized and the target metrics are measured against ground-truth user behavior and baseline groups. Self-citations [2],[3],[4],[11] provide component-level algorithms (online adaptation, plan recognition, intention prediction) that are peer-reviewed prior work and are re-tested here with new data (Tables III and V), so they are not unverified assumptions entailing the headline numbers. Two reporting issues are correctness concerns rather than circularity: the 15.85% time reduction is the absolute percentage-point gap (Table IV), and the '24% MSE reduction' is the absolute drop (Table V); neither indicates that the result was constructed from its inputs. The appendix mutual-information proof is mathematically informal, but the inequality itself is a standard information-theoretic fact and is not used to fit the experimental outcomes.

Assumptions & free parameters 5 free parameters · 5 assumptions · 1 invented entities

The framework relies on a hand-built task graph, the assumption of known object positions, and a set of hand-tuned thresholds and sensitivity scores. These are the main free parameters and domain assumptions, alongside standard information-theoretic facts and pretrained perception models.

free parameters (5)
  • confidence threshold = 0.6
    Used in confidence-based restriction for action prediction (Sec. V-C2); no cross-validation reported, appears chosen to improve error rates on the offline test data.
  • human detection range = 2 meters
    Distance threshold in hierarchical pose detection (Sec. IV-B) that decides when to activate downstream modules; set by hand.
  • sensitivity scores = high/medium/low per Table I
    Hand-assigned cost weights for misclassification types in action prediction (Sec. IV-C); scores are not learned or justified quantitatively.
  • DTW distance metric = unspecified variant
    The alignment in Eq. (4) uses dynamic time warping but the exact parameters (window, cost) are not given (Sec. IV-E).
  • planning horizon n = e.g., 2 seconds
    Objective function in Eq. (2) includes horizon n; the actual value used during experiments is not specified (Sec. III).
assumptions (5)
  • domain assumption Task graph is known to both human and robot and fully covers all valid plans
    Stated in Sec. I and III; the plan predictor Eq. (4)-(5) searches only within this graph.
  • domain assumption Object positions are fixed and exactly known; no uncertainty in object perception
    Stated in Sec. III and V-A; also acknowledged as a limitation in Sec. VI.
  • standard math Joint mutual information is at least the maximum marginal mutual information
    The paper attempts a proof in Appendix A, but the proof is flawed; the inequality itself is a standard information-theoretic result (chain rule / data processing inequality).
  • domain assumption Pretrained MobileNet, BlazePose, and DeepSpeech models provide reliable visual and speech features
    Used in perception modules (Sec. V-B1); their errors are not modeled.
  • domain assumption Supervisor-subordinate paradigm: human leads, robot assists
    Adopted in Sec. III from the literature; defines the collaboration structure.
invented entities (1)
  • No new physical entities introduced
    purpose: None
    The paper introduces no particles, forces, mediators, dimensions, or conserved quantities. The hierarchical task graph is a modeling construct, not a new entity.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Robustifying Long-term Human-Robot Collaboration through a Multimodal and Hierarchical Framework." pith.science (2026). https://pith.science/paper/DN567AHV

@misc{pith2026241115711,
  author       = {Pith},
  title        = {Pith review of: Robustifying Long-term Human-Robot Collaboration through a Multimodal and Hierarchical Framework},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DN567AHV}},
  note         = {Machine review of arXiv:2411.15711}
}
read the original abstract

Long-term Human-Robot Collaboration (HRC) is crucial for enabling flexible manufacturing systems and integrating companion robots into daily human environments over extended periods. This paper identifies several key challenges for such collaborations, such as accurate recognition of human plan, robustness to disturbances, operational efficiency, adaptability to diverse user behaviors, and sustained human satisfaction. To address these challenges, we model the long-term HRC task through a hierarchical task graph and presents a novel multimodal and hierarchical framework to enable robots to better assist humans to advance on the task graph. In particular, the proposed multimodal framework integrates visual observations with speech commands to facilitate intuitive and flexible human-robot interactions. Additionally, our hierarchical designs for both human pose detection and plan prediction allow better understanding of human behaviors and significantly enhance system accuracy, robustness and flexibility. Moreover, an online adaptation mechanism enables real-time adjustment to diverse user behaviors. We deploy the proposed framework to KINOVA GEN3 robot and conduct extensive user studies on real-world long-term HRC assembly scenarios. Experimental results show that our approaches reduce task completion time by 15.9%, achieves an average task success rate of 91.8% and an overall user satisfaction score of 84% in long-term HRC tasks, showcasing its applicability in enhancing real-world long-term HRC.

Figures

Figures reproduced from arXiv: 2411.15711 by the authors.

Figure 1
Figure 1. Overview of the hierarchical task graph for the long-term toy car assembly HRC task. Each node in the long-term graph (left) corresponds to a [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. The architecture of the proposed HRC framework. [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 4
Figure 4. Environment Setting The environment setup is shown in [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figures from the paper (9 more)
Figure 3
Figure 3. Figure 3: Final Assembly Task Each stage is designed to require collaborative efforts, in which the robot’s role is to support the human collaborator by delivering necessary objects and performing specific assistive actions, such as rotating or lifting objects. The actions and m…
Figure 5
Figure 5. Figure 5: Misdetection in naive human pose detection model. [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: Correct detection using the hierarchical pose detection framework. [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 7
Figure 7. Figure 7: Keypoint deviation with and without hierarchical human detection. [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 8
Figure 8. Figure 8: Action prediction error rate under different restriction strategies [PITH_FULL_IMAGE:figures/full_fig_p009_8.png]
Figure 9
Figure 9. Figure 9: The user plan satisfaction rate, as calculated based on 120 trials [PITH_FULL_IMAGE:figures/full_fig_p009_9.png]
Figure 10
Figure 10. Figure 10: Quantified results of the user survey, with scores calculated as [PITH_FULL_IMAGE:figures/full_fig_p009_10.png]
Figure 11
Figure 11. Figure 11: Confusion matrix for intention prediction with and without confidence-based restriction [PITH_FULL_IMAGE:figures/full_fig_p012_11.png]
Figure 12
Figure 12. Figure 12: Car assembly task demonstration. Click the image to watch the video. [PITH_FULL_IMAGE:figures/full_fig_p013_12.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

36 extracted references · 29 canonical work pages

  1. [1]

    Robust on- line model adaptation by extended kalman filter with exponential moving average and dynamic multi-epoch strategy

    Abulikemu Abuduweili and Changliu Liu. Robust on- line model adaptation by extended kalman filter with exponential moving average and dynamic multi-epoch strategy. In Learning for Dynamics and Control , pages 65–74. PMLR, 2020

    work page 2020

  2. [2]

    Robust non- linear adaptation algorithms for multitask prediction net- works

    Abulikemu Abuduweili and Changliu Liu. Robust non- linear adaptation algorithms for multitask prediction net- works. International Journal of Adaptive Control and Signal Processing, 35(3):314–341, 2021

    work page 2021

  3. [3]

    Online model adaptation with feedforward compensation

    Abulikemu Abuduweili and Changliu Liu. Online model adaptation with feedforward compensation. In Confer- ence on Robot Learning, pages 3687–3709. PMLR, 2023

    work page 2023

  4. [4]

    Adaptable Human Intention and Trajectory Prediction for Human-Robot Collaboration

    Abulikemu Abuduweili, Siyan Li, and Changliu Liu. Adaptable human intention and trajectory predic- tion for human-robot collaboration. arXiv preprint arXiv:1909.05089, 2019

    work page Pith review arXiv 1909

  5. [5]

    Assis- tive technology design and development for acceptable robotics companions for ageing years

    Farshid Amirabdollahian, Rieks op den Akker, Sandra Bedaf, Richard Bormann, Heather Draper, Vanessa Ev- ers, Jorge Gallego P ´erez, Gert Jan Gelderblom, Car- olina Gutierrez Ruiz, David Hewson, et al. Assis- tive technology design and development for acceptable robotics companions for ageing years. Paladyn, Journal of Behavioral Robotics , 4(2):94–112, 2013

    work page 2013

  6. [6]

    10 years of human-nao interaction research: A scoping review

    Aida Amirova, Nazerke Rakhymbayeva, Elmira Yadol- lahi, Anara Sandygulova, and Wafa Johal. 10 years of human-nao interaction research: A scoping review. Frontiers in Robotics and AI , 8:744526, 2021

    work page 2021

  7. [7]

    Deep speech 2: End-to-end speech recognition in english and mandarin

    Dario Amodei, Sundaram Ananthanarayanan, Rishita Anubhai, Jingliang Bai, Eric Battenberg, Carl Case, Jared Casper, Bryan Catanzaro, Qiang Cheng, Guoliang Chen, et al. Deep speech 2: End-to-end speech recognition in english and mandarin. In International conference on machine learning, pages 173–182. PMLR, 2016

    work page 2016

  8. [8]

    Blazepose: On-device real-time body pose tracking

    Valentin Bazarevsky, Ivan Grishchenko, Karthik Raveen- dran, Tyler Zhu, Fan Zhang, and Matthias Grundmann. Blazepose: On-device real-time body pose tracking. arXiv preprint arXiv:2006.10204 , 2020

    arXiv 2006

Show all 36 references
  1. [9]

    Realtime multi-person 2d pose estimation using part affinity fields

    Zhe Cao, Tomas Simon, Shih-En Wei, and Yaser Sheikh. Realtime multi-person 2d pose estimation using part affinity fields. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 7291– 7299, 2017

    2017

  2. [10]

    History aware multimodal transformer for vision-and-language navigation

    Shizhe Chen, Pierre-Louis Guhur, Cordelia Schmid, and Ivan Laptev. History aware multimodal transformer for vision-and-language navigation. Advances in neural information processing systems , 34:5834–5847, 2021

    2021

  3. [11]

    Towards efficient human-robot collaboration with robust plan recognition and trajectory prediction

    Yujiao Cheng, Liting Sun, Changliu Liu, and Masayoshi Tomizuka. Towards efficient human-robot collaboration with robust plan recognition and trajectory prediction. IEEE Robotics and Automation Letters , 5(2):2602–2609, 2020

    2020

  4. [12]

    Vision-based holistic scene understanding towards proactive human– robot collaboration

    Junming Fan, Pai Zheng, and Shufei Li. Vision-based holistic scene understanding towards proactive human– robot collaboration. Robotics and Computer-Integrated Manufacturing, 75:102304, 2022

    2022

  5. [13]

    A real-time rescheduling algorithm for multi-robot plan execution

    Ying Feng, Adittyo Paul, Zhe Chen, and Jiaoyang Li. A real-time rescheduling algorithm for multi-robot plan execution. In Proceedings of the International Confer- ence on Automated Planning and Scheduling, volume 34, pages 201–209, 2024

    2024

  6. [14]

    Relay policy learning: Solv- ing long-horizon tasks via imitation and reinforcement learning

    Abhishek Gupta, Vikash Kumar, Corey Lynch, Sergey Levine, and Karol Hausman. Relay policy learning: Solv- ing long-horizon tasks via imitation and reinforcement learning. arXiv preprint arXiv:1910.11956 , 2019

    1910 arXiv

  7. [15]

    Homem de Mello and A.C

    L.S. Homem de Mello and A.C. Sanderson. And/or graph representation of assembly plans. IEEE Transactions on Robotics and Automation , 6(2):188–199, 1990. doi: 10. 1109/70.54734

    1990

  8. [16]

    A multimodal emo- tional human–robot interaction architecture for social robots engaged in bidirectional communication

    Alexander Hong, Nolan Lunscher, Tianhao Hu, Yuma Tsuboi, Xinyi Zhang, Silas Franco dos Reis Alves, Goldie Nejat, and Beno Benhabib. A multimodal emo- tional human–robot interaction architecture for social robots engaged in bidirectional communication. IEEE transactions on cybe...

    2020

  9. [17]

    Multi-agent path finding with kinematic constraints

    Wolfgang H ¨onig, TK Kumar, Liron Cohen, Hang Ma, Hong Xu, Nora Ayanian, and Sven Koenig. Multi-agent path finding with kinematic constraints. In Proceedings of the International Conference on Automated Planning and Scheduling, volume 26, pages 477–485, 2016

    2016

  10. [18]

    Slaves no longer: review on role assignment for human–robot joint motor action

    Nathanael Jarrasse, Vittorio Sanguineti, and Etienne Bur- det. Slaves no longer: review on role assignment for human–robot joint motor action. Adaptive Behavior, 22 (1):70–82, 2014

    2014

  11. [19]

    Towards robust human-robot collaborative man- ufacturing: Multimodal fusion

    Hongyi Liu, Tongtong Fang, Tianyu Zhou, and Lihui Wang. Towards robust human-robot collaborative man- ufacturing: Multimodal fusion. IEEE Access , 6:74762– 74771, 2018

    2018

  12. [20]

    Human motion prediction using adaptable recurrent neural networks and inverse kinematics

    Ruixuan Liu and Changliu Liu. Human motion prediction using adaptable recurrent neural networks and inverse kinematics. IEEE Control Systems Letters , 5(5):1651– 1656, 2021. doi: 10.1109/LCSYS.2020.3042609

    2021

  13. [21]

    Task- agnostic adaptation for safe human-robot handover

    Ruixuan Liu, Rui Chen, and Changliu Liu. Task- agnostic adaptation for safe human-robot handover. IFAC-PapersOnLine, 55(41):175–180, 2022. ISSN 2405-8963. doi: https://doi.org/10.1016/j.ifacol.2023.01

    2022 doi

  14. [22]

    Proactive human-robot co-assembly: Leveraging human intention prediction and robust safe control

    Ruixuan Liu, Rui Chen, Abulikemu Abuduweili, and Changliu Liu. Proactive human-robot co-assembly: Leveraging human intention prediction and robust safe control. In 2023 IEEE Conference on Control Technology and Applications (CCTA) , pages 339–345. IEEE, 2023

    2023

  15. [23]

    Human– robot collaboration in manufacturing applications: A re- view

    Eloise Matheson, Riccardo Minto, Emanuele GG Zampieri, Maurizio Faccio, and Giulio Rosati. Human– robot collaboration in manufacturing applications: A re- view. Robotics, 8(4):100, 2019

    2019

  16. [24]

    Natural multimodal communication for human–robot collaboration

    Inaki Maurtua, Izaskun Fernandez, Alberto Tellaeche, Johan Kildal, Loreto Susperregi, Aitor Ibarguren, and Basilio Sierra. Natural multimodal communication for human–robot collaboration. International Journal of Advanced Robotic Systems , 14(4), 2017

    2017

  17. [25]

    Generative skill chaining: Long-horizon skill planning with diffusion models

    Utkarsh Aashu Mishra, Shangjie Xue, Yongxin Chen, and Danfei Xu. Generative skill chaining: Long-horizon skill planning with diffusion models. In Conference on Robot Learning, pages 2905–2925. PMLR, 2023

    2023

  18. [26]

    Long-Horizon Prediction for Human-Robot Collaboration

    Eley Ng. Long-Horizon Prediction for Human-Robot Collaboration. Stanford University, 2023

    2023

  19. [27]

    Towards seamless human robot collaboration: integrating multi- modal interaction

    Stergios Papanastasiou, Niki Kousi, Panagiotis Kara- giannis, Christos Gkournelos, Apostolis Papavasileiou, Konstantinos Dimoulas, Konstantinos Baris, Spyridon Koukas, George Michalos, and Sotiris Makris. Towards seamless human robot collaboration: integrating multi- modal int...

    2019

  20. [28]

    Modeling long-horizon tasks as sequential interaction landscapes

    S ¨oren Pirk, Karol Hausman, Alexander Toshev, and Mohi Khansari. Modeling long-horizon tasks as sequential interaction landscapes. arXiv preprint arXiv:2006.04843, 2020

    2006 arXiv

  21. [29]

    Mobilenetv2: In- verted residuals and linear bottlenecks

    Mark Sandler, Andrew Howard, Menglong Zhu, Andrey Zhmoginov, and Liang-Chieh Chen. Mobilenetv2: In- verted residuals and linear bottlenecks. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 4510–4520, 2018

    2018

  22. [30]

    3d human pose estimation: A review of the literature and analysis of covariates

    Nikolaos Sarafianos, Bogdan Boteanu, Bogdan Ionescu, and Ioannis A Kakadiaris. 3d human pose estimation: A review of the literature and analysis of covariates. Computer Vision and Image Understanding , 152:1–20, 2016

    2016

  23. [31]

    Matching incomplete time series with dynamic time warping: an algorithm and an application to post-stroke rehabilitation

    Paolo Tormene, Toni Giorgino, Silvana Quaglini, and Mario Stefanelli. Matching incomplete time series with dynamic time warping: an algorithm and an application to post-stroke rehabilitation. Artificial intelligence in medicine, 45(1):11–34, 2009

    2009

  24. [32]

    Human-robot collabo- ration in industry

    Ales Vysocky and Petr Novak. Human-robot collabo- ration in industry. MM Science Journal , 9(2):903–906, 2016

    2016

  25. [33]

    Multimodal human–robot interaction for human-centric smart manufacturing: A survey

    Tian Wang, Pai Zheng, Shufei Li, and Lihui Wang. Multimodal human–robot interaction for human-centric smart manufacturing: A survey. Advanced Intelligent Systems, 6(3):2300359, 2024

    2024

  26. [34]

    Kalman filter

    Gregory F Welch. Kalman filter. In Computer vision: a reference guide, pages 721–723. Springer, 2021

    2021

  27. [35]

    Are transformers effective for time series forecasting? In Proceedings of the AAAI Conference on Artificial Intelligence, volume 37, pages 11121–11128, 2023

    Ailing Zeng, Muxi Chen, Lei Zhang, and Qiang Xu. Are transformers effective for time series forecasting? In Proceedings of the AAAI Conference on Artificial Intelligence, volume 37, pages 11121–11128, 2023. APPENDIX A. Proof That the Multimodal Framework Yields Greater Mutual ...

    2023

  28. [122]

    4th IFAC Workshop on Cyber- Physical and Human Systems CPHS 2022

    URL https://www.sciencedirect.com/science/article/ pii/S2405896323001295. 4th IFAC Workshop on Cyber- Physical and Human Systems CPHS 2022

    2022

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.