pith. sign in

arxiv: 2508.02204 · v4 · submitted 2025-08-04 · 💻 cs.RO

TacMan-Turbo: Proactive Tactile Control for Robust and Efficient Articulated Object Manipulation

Zihang Zhao , Zhenghao Qi , Yuyang Li , Leiyao Cui , Zhi Han , Lecheng Ruan , Yixin Zhu This is my paper

Pith reviewed 2026-05-19 01:20 UTC · model grok-4.3

classification 💻 cs.RO
keywords articulated object manipulationtactile controlproactive controlrobot manipulationkinematic estimationtactile sensingefficiency in robotics
0
0 comments X

The pith

TacMan-Turbo shows that proactive interpretation of tactile deviations as kinematic information enables both robust and efficient articulated object manipulation without prior models.

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

The paper aims to show that robots can handle articulated objects effectively and efficiently by shifting from reactive to proactive use of tactile feedback. It argues that contact deviations provide local kinematic cues that allow prediction of future interactions instead of just correcting errors. This matters because it removes the need for accurate object models while avoiding the inefficiency of step-by-step exploration. If true, robots could manipulate varied objects in homes and workplaces with fewer actions and smoother motions.

Core claim

TacMan-Turbo is introduced as a proactive tactile control framework that interprets tactile contact deviations as rich sources of local kinematic information rather than error signals. This enables the controller to predict optimal future interactions and make proactive adjustments. In evaluations on 200 simulated objects and real-world tests, it achieves a 100% success rate and outperforms previous tactile-informed methods in time efficiency, action efficiency, and trajectory smoothness with high statistical significance.

What carries the argument

The proactive tactile controller that converts contact deviations into predictions for future adjustments to enhance efficiency.

If this is right

  • Robots achieve reliable manipulation on diverse articulated objects without any kinematic priors.
  • Manipulation becomes faster and requires fewer actions compared to reactive approaches.
  • Trajectories are smoother due to proactive rather than compensatory control.
  • The effectiveness-efficiency trade-off is resolved through this new interpretation of tactile data.

Where Pith is reading between the lines

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

  • Similar proactive sensing strategies could apply to other robot tasks involving uncertain structures.
  • Integration with visual or other sensors might further improve performance in complex environments.
  • This suggests rethinking sensor data in control systems as information for prediction rather than solely for correction.

Load-bearing premise

Tactile contact deviations provide enough rich and reliable local kinematic information to accurately predict and enable optimal future interactions without any prior model.

What would settle it

Demonstrating cases where tactile deviations do not correlate with actual kinematic variations, leading to incorrect proactive adjustments and lower success rates than reactive methods.

Figures

Figures reproduced from arXiv: 2508.02204 by Lecheng Ruan, Leiyao Cui, Yixin Zhu, Yuyang Li, Zhenghao Qi, Zhi Han, Zihang Zhao.

Figure 1
Figure 1. Figure 1: Robust and efficient tactile-informed manipulation of articulated objects. TacMan-Turbo enables robots to achieve both effective and efficient manipulation of articulated objects through a novel tactile-based control framework.Unlike previous approaches, TacMan-Turbo leverages past tactile experience to extract local kinematic information, allowing the controller to predict optimal interaction positions an… view at source ↗
Figure 2
Figure 2. Figure 2: Schematic overview of proactive tactile proactive control framework, TacMan-Turbo. Our framework integrates three sequential components that enable proactive manipulation: (a) In-hand pose estimation extracts tactile contact patterns from gripper sensors to determine the relative transformation between current gripper pose Tgi and handle pose Thi . (b) Handle pose prediction analyzes sequential pose estima… view at source ↗
Figure 3
Figure 3. Figure 3: Simulation test environment and object categories. Our comprehensive evaluation framework encompasses three progressively complex articulated object categories: (a) 50 prismatic-joint objects featuring linear motion paths, representing the most kinematically straightforward mechanisms; (b) 50 revolute-joint objects with circular motion paths, introducing rotational complexity; and (c) 100 complex articulat… view at source ↗
Figure 4
Figure 4. Figure 4: Qualitative results of simulation studies. This visualization demonstrates TacMan-Turbo’s performance across three representative cases from our object categories: prismatic (left), revolute (middle), and complex articulation (right). The top row displays trajectory tracking performance with actual handle positions (blue), predicted handle positions (orange), gripper positions (green), and contact points (… view at source ↗
Figure 5
Figure 5. Figure 5: Time efficiency comparison (logarithmic scale, seconds). Evalua￾tion of task completion times reveals TacMan-Turbo’s superior efficiency com￾pared to Tac-Man across all tested articulation types. Our method consistently completes manipulation tasks in significantly less time, with performance ad￾vantages particularly pronounced for complex articulations. Statistical analysis confirms these improvements are… view at source ↗
Figure 6
Figure 6. Figure 6: Analysis of action efficiency across manipulation strategies. Comprehensive evaluation reveals TacMan-Turbo’s superior motion efficiency through three complementary measures. Temporal profiles highlight TacMan￾Turbo’s continuous productive motion vs. Tac-Man’s alternating execution￾recovery cycles. Mean efficiency measurements confirm TacMan-Turbo’s con￾sistently higher percentage of productive motion acro… view at source ↗
Figure 7
Figure 7. Figure 7: Analysis of motion smoothness through jerk characteristics. Mo￾tion smoothness evaluation reveals TacMan-Turbo maintains consistently lower jerk magnitudes by eliminating the disruptive transition phases inherent to Tac￾Man’s approach. The temporal comparison shows TacMan-Turbo’s continuous motion profile contrasted with Tac-Man’s oscillatory pattern. Statistical analy￾sis across 200 test objects confirms … view at source ↗
Figure 8
Figure 8. Figure 8: Physical experimental setup. The experimental platform consists of a Kinova Gen3 7-DoF robotic arm equipped with a Robotiq 2F-85 parallel gripper. Each gripper finger incorporates a GelSight-type tactile sensor featur￾ing an 8×8 marker grid for precise contact force measurement and deformation tracking. 12 [PITH_FULL_IMAGE:figures/full_fig_p012_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Test objects for real-world experiments. The experimental vali￾dation employs three articulated objects with distinct kinematic characteristics: (a) a wooden drawer representing pure prismatic joint motion along a linear path, (b) a microwave oven featuring revolute joint dynamics through its door hinge mechanism, and (c) a bench vise with a hand crank generating complex helical motion. These objects colle… view at source ↗
Figure 10
Figure 10. Figure 10: Tactile-informed pose estimation of in-hand objects. Precise tracking of object movements during manipulation is demonstrated through tactile feedback across three different objects subjected to five distinct manipulations: initial state, pressing down, dragging upward, and two types of twisting motions. Contact points (marked with purple dots) are tracked to compute relative pose transformations using Eq… view at source ↗
Figure 11
Figure 11. Figure 11: Real-world experiment results. Comparative analysis across three articulated objects (drawer, microwave, vise) demonstrates successful manipulation by both methods with clear performance differences. Each row presents a complete manipulation sequence showing initial stable contact (left), manipulation stages (middle), and corresponding 3D trajectories (right). Tactile feedback visualizations display conta… view at source ↗
Figure 12
Figure 12. Figure 12: Tests on various common household objects. Our TacMan-Turbo approach successfully generalizes to diverse everyday objects without requiring additional tuning. The system effectively manipulates: (a) a battery charging dock requiring precise device extraction, (b) a power switch with discrete on/off state transitions, (c) a coffee maker with rotational lid mechanism, and (d) a sewing machine featuring comp… view at source ↗
Figure 13
Figure 13. Figure 13: Manipulation under disturbance. The efficiency gains enabled by TacMan-Turbo do not compromise its robustness against unexpected disturbances that frequently occur in human-centric environments. Unlike previous reactive paradigms [73] that require stopping for reactive adjustments, TacMan-Turbo can maintain manipulation progress continuously. Detailed manipulation processes are available in the Supplement… view at source ↗
Figure 14
Figure 14. Figure 14: Effects of base velocity direction. We investigate how initial directional alignment impacts manipulation performance. (a) The simulation setup measures the angle θ between the correct interaction direction (green arrow) and the initial motion direction u gi 0 (orange arrow). (b) Time efficiency results reveal significant performance deterioration as θ increases in magnitude, with optimal performance near… view at source ↗
read the original abstract

Adept manipulation of articulated objects is essential for robots to operate successfully in human environments. Such manipulation requires both effectiveness--reliable operation despite uncertain object structures--and efficiency--swift execution with minimal redundant steps and smooth actions. Existing approaches struggle to achieve both objectives simultaneously: methods relying on predefined kinematic models lack effectiveness when encountering structural variations, while tactile-informed approaches achieve robust manipulation without kinematic priors but compromise efficiency through reactive, step-by-step exploration-compensation cycles. This paper introduces TacMan-Turbo, a novel proactive tactile control framework for articulated object manipulation that mitigates this fundamental trade-off. Unlike previous approaches that treat tactile contact deviations merely as error signals requiring compensation, our method interprets these deviations as rich sources of local kinematic information. This new perspective enables our controller to predict optimal future interactions and make proactive adjustments, significantly enhancing manipulation efficiency. In comprehensive evaluations across 200 diverse simulated articulated objects and real-world experiments, our approach maintains a 100% success rate while significantly outperforming the previous tactile-informed method in time efficiency, action efficiency, and trajectory smoothness (all p-values < 0.0001). These results demonstrate that the long-standing trade-off between effectiveness and efficiency in articulated object manipulation can be successfully resolved without relying on prior kinematic knowledge.

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 / 1 minor

Summary. The paper introduces TacMan-Turbo, a proactive tactile control framework for articulated object manipulation. It interprets tactile contact deviations as rich local kinematic signals rather than mere error signals, enabling prediction of optimal future interactions without prior kinematic models. Evaluations on 200 diverse simulated articulated objects and real-world experiments report a 100% success rate, with statistically significant improvements (p < 0.0001) over prior tactile-informed methods in time efficiency, action efficiency, and trajectory smoothness.

Significance. If the empirical results hold under rigorous controls, the work would be significant for robotics by resolving the effectiveness-efficiency trade-off in articulated manipulation using only tactile feedback, without relying on predefined kinematic structures. This could enable more robust operation in unstructured human environments and advance proactive tactile sensing approaches.

major comments (2)
  1. [Abstract] Abstract and evaluations: The central claims of 100% success rate and p < 0.0001 improvements are presented without any description of experimental design details, such as how the 200 simulated objects were varied in joint types/locations, contact geometries, or noise levels; how optimal actions are predicted from deviations; or statistical controls for object distribution bias. This leaves the proactive interpretation's contribution to the results under-supported and load-bearing for the effectiveness claim.
  2. [Introduction] Introduction and method: The assumption that measured tactile deviations at contact points supply sufficiently rich and unique local kinematic information to determine joint axes, types, or locations for forward prediction is stated but not analyzed for under-determination. A single deviation vector is consistent with multiple possible articulations, so the mapping to proactive adjustments requires additional (unstated) regularization or assumptions that may not generalize beyond the tested objects.
minor comments (1)
  1. [Abstract] The abstract mentions 'comprehensive evaluations' but provides no quantitative baselines or ablation studies on the proactive component versus pure reactive compensation.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive comments on our work. We have carefully considered each point and provide point-by-point responses below. Where appropriate, we will revise the manuscript to address the concerns raised.

read point-by-point responses

  1. Referee: [Abstract] Abstract and evaluations: The central claims of 100% success rate and p < 0.0001 improvements are presented without any description of experimental design details, such as how the 200 simulated objects were varied in joint types/locations, contact geometries, or noise levels; how optimal actions are predicted from deviations; or statistical controls for object distribution bias. This leaves the proactive interpretation's contribution to the results under-supported and load-bearing for the effectiveness claim.

    Authors: The abstract is intentionally concise to highlight the key contributions and results. Detailed descriptions of the experimental design are provided in Section IV of the manuscript, including the variation of 200 simulated objects across different joint types (revolute and prismatic), locations, and contact geometries. Noise is incorporated as sensor noise in tactile readings. The method for predicting optimal actions from tactile deviations is explained in Section III, where local kinematic signals are used to anticipate joint movements and adjust proactively. Statistical analysis includes controls for object distribution by sampling from a diverse parameter space, with results reported using mean and standard deviation, and p-values from paired t-tests. To better support the claims, we will update the abstract to include a brief mention of the experimental variations and add cross-references to the relevant sections. revision: yes

  2. Referee: [Introduction] Introduction and method: The assumption that measured tactile deviations at contact points supply sufficiently rich and unique local kinematic information to determine joint axes, types, or locations for forward prediction is stated but not analyzed for under-determination. A single deviation vector is consistent with multiple possible articulations, so the mapping to proactive adjustments requires additional (unstated) regularization or assumptions that may not generalize beyond the tested objects.

    Authors: We appreciate this observation regarding potential under-determination. In our framework, the proactive control relies on temporal sequences of tactile deviations rather than a single vector, allowing the system to accumulate information over multiple time steps to resolve ambiguities in joint parameters. Additionally, the controller incorporates a smoothness prior and minimal intervention principle as regularization to select among possible articulations. These aspects are outlined in the method description. We agree that a more explicit analysis of identifiability and generalization would be beneficial. We will add a subsection discussing the assumptions, potential ambiguities, and how the proactive approach mitigates them through online adaptation. revision: partial

Circularity Check

0 steps flagged

No circularity: empirical framework with independent experimental validation

full rationale

The paper introduces TacMan-Turbo as a proactive tactile control framework that interprets contact deviations as local kinematic signals to enable forward prediction of actions. Its central claims rest on comprehensive evaluations across 200 simulated articulated objects and real-world experiments, reporting 100% success rate and statistically significant efficiency gains (p<0.0001) over a prior tactile-informed baseline. No equations, parameter fits, uniqueness theorems, or self-citations are invoked that reduce the performance metrics or the proactive interpretation to inputs by construction. The derivation chain is self-contained because the method is presented as an algorithmic control policy whose effectiveness is demonstrated through external benchmarking rather than tautological redefinition or fitted renaming of results.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The framework rests on the domain assumption that tactile signals alone can yield usable kinematic predictions; no free parameters or new entities are explicitly introduced in the abstract.

axioms (1)
  • domain assumption Tactile contact deviations contain sufficient local kinematic information to support proactive prediction of future interactions
    This premise is invoked when the abstract states that deviations are interpreted as rich sources of local kinematic information rather than mere errors.

pith-pipeline@v0.9.0 · 5775 in / 1173 out tokens · 35203 ms · 2026-05-19T01:20:30.016657+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

75 extracted references · 75 canonical work pages

  1. [1]

    ATI Industrial Automation, ATI Multi-Axis Force/Torque Sensor Sys- tem, Available online: https://www.ati-ia.com/products/ft/ sensors.aspx, accessed: 2025, 2025

    work page 2025

  2. [2]

    A. D. Berger, P. K. Khosla, Using tactile data for real-time feedback, International Journal of Robotics Research (IJRR) 10 (2) (1991) 88–102

    work page 1991

  3. [3]

    C. M. Boutry, M. Negre, M. Jorda, O. Vardoulis, A. Chortos, O. Khatib, Z. Bao, A hierarchically patterned, bioinspired e-skin able to detect the direction of applied pressure for robotics, Science Robotics 3 (24) (2018) eaau6914

    work page 2018

  4. [4]

    Brahmbhatt, C

    S. Brahmbhatt, C. Ham, C. C. Kemp, J. Hays, Contactdb: Analyzing and predicting grasp contact via thermal imaging, in: Proceedings of Confer- ence on Computer Vision and Pattern Recognition (CVPR), 2019

    work page 2019

  5. [5]

    Burget, A

    F. Burget, A. Hornung, M. Bennewitz, Whole-body motion planning for manipulation of articulated objects, in: IEEE International Conference on Robotics and Automation (ICRA), 2013

    work page 2013

  6. [6]

    N. Chen, H. Zhang, R. Rink, Edge tracking using tactile servo, in: IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 1995

    work page 1995

  7. [7]

    Y . Chen, T. Wu, S. Wang, X. Feng, J. Jiang, Z. Lu, S. McAleer, H. Dong, S.-C. Zhu, Y . Yang, Towards human-level bimanual dexterous manipula- tion with reinforcement learning, in: Proceedings of Advances in Neural Information Processing Systems (NeurIPS), 2022

    work page 2022

  8. [8]

    Cheng, M

    C.-A. Cheng, M. Mukadam, J. Issac, S. Birchfield, D. Fox, B. Boots, N. Ratliff, Rmpflow: A geometric framework for generation of multitask motion policies, IEEE Transactions on Automation Science and Engi- neering (T-ASE) 18 (3) (2021) 968–987

    work page 2021

  9. [9]

    Chitta, B

    S. Chitta, B. Cohen, M. Likhachev, Planning for autonomous door open- ing with a mobile manipulator, in: IEEE International Conference on Robotics and Automation (ICRA), 2010

    work page 2010

  10. [10]

    T. Dai, J. Wong, Y . Jiang, C. Wang, C. Gokmen, R. Zhang, J. Wu, L. Fei- Fei, ACDC: Automated Creation of Digital Cousins for Robust Policy Learning, in: Conference on Robot Learning (CoRL), 2024

    work page 2024

  11. [11]

    A. J. Davison, I. D. Reid, N. D. Molton, O. Stasse, MonoSLAM: Real- time single camera SLAM, Transactions on Pattern Analysis and Machine Intelligence (TPAMI) 29 (6) (2007) 1052–1067

    work page 2007

  12. [12]

    A. B. Dawood, C. Coppola, K. Althoefer, Learning Decoupled Multi- touch Force Estimation, Localization and Stretch for Soft Capacitive E- skin, in: IEEE International Conference on Robotics and Automation (ICRA), 2023

    work page 2023

  13. [13]

    J. A. Fishel, G. E. Loeb, Sensing tactile microvibrations with the Bio- Tac—Comparison with human sensitivity, in: International Conference on Biomedical Robotics and Biomechatronics (BioRob), 2012

    work page 2012

  14. [14]

    Z. Fu, T. Z. Zhao, C. Finn, Mobile ALOHA: Learning Bimanual Mobile Manipulation with Low-Cost Whole-Body Teleoperation, in: Conference on Robot Learning (CoRL), 2024

    work page 2024

  15. [15]

    H. Geng, Z. Li, Y . Geng, J. Chen, H. Dong, H. Wang, Partmanip: Learn- ing cross-category generalizable part manipulation policy from point cloud observations, in: Proceedings of Conference on Computer Vision and Pattern Recognition (CVPR), 2023

    work page 2023

  16. [16]

    H. Geng, H. Xu, C. Zhao, C. Xu, L. Yi, S. Huang, H. Wang, Gapartnet: Cross-category domain-generalizable object perception and manipulation via generalizable and actionable parts, in: Proceedings of Conference on Computer Vision and Pattern Recognition (CVPR), 2023

    work page 2023

  17. [17]

    R. Gong, J. Huang, Y . Zhao, H. Geng, X. Gao, Q. Wu, W. Ai, Z. Zhou, D. Terzopoulos, S.-C. Zhu, ARNOLD: A Benchmark for Language- Grounded Task Learning With Continuous States in Realistic 3D Scenes, in: Proceedings of International Conference on Computer Vision (ICCV), 2023

    work page 2023

  18. [18]

    P. L. Gould, Y . Feng, Introduction to linear elasticity, vol. 2, Springer, 1994

    work page 1994

  19. [19]

    Hausman, S

    K. Hausman, S. Niekum, S. Osentoski, G. S. Sukhatme, Active articula- tion model estimation through interactive perception, in: IEEE Interna- tional Conference on Robotics and Automation (ICRA), 2015

    work page 2015

  20. [20]

    R. Hu, W. Li, O. Van Kaick, A. Shamir, H. Zhang, H. Huang, Learning to predict part mobility from a single static snapshot, ACM Transactions on Graphics (TOG) 36 (6) (2017) 1–13

    work page 2017

  21. [21]

    W. Kabsch, A solution for the best rotation to relate two sets of vectors, Acta Crystallographica Section A: Crystal Physics, Diffraction, Theoret- ical and General Crystallography 32 (5) (1976) 922–923

    work page 1976

  22. [22]

    Kappassov, J.-A

    Z. Kappassov, J.-A. Corrales, V . Perdereau, Touch driven controller and tactile features for physical interactions, Robotics and Autonomous Sys- tems 123 (2020) 103332

    work page 2020

  23. [23]

    Karayiannidis, C

    Y . Karayiannidis, C. Smith, F. E. V . Barrientos, P. ¨Ogren, D. Kragic, An adaptive control approach for opening doors and drawers under uncertain- ties, IEEE Transactions on Robotics (T-RO) 32 (1) (2016) 161–175

    work page 2016

  24. [24]

    Open sesame!

    Y . Karayiannidis, C. Smith, F. E. Vina, P. Ogren, D. Kragic, “Open sesame!” adaptive force/velocity control for opening unknown doors, in: IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2012

    work page 2012

  25. [25]

    N. F. Lepora, J. Lloyd, Pose-based tactile servoing: Controlled soft touch 19 using deep learning, IEEE Robotics and Automation Magazine (RA-M) 28 (4) (2021) 43–55

    work page 2021

  26. [26]

    Q. Li, C. Sch ¨urmann, R. Haschke, H. J. Ritter, A Control Framework for Tactile Servoing., in: Robotics: Science and Systems (RSS), 2013

    work page 2013

  27. [27]

    W. Li, M. Wang, J. Li, Y . Su, D. K. Jha, X. Qian, K. Althoefer, H. Liu, L3 F-TOUCH: A Wireless GelSight with Decoupled Tactile and Three-axis Force Sensing, IEEE Robotics and Automation Letters (RA-L)

    work page

  28. [28]

    W. Li, Z. Zhao, L. Cui, W. Zhang, H. Liu, L.-A. Li, Y . Zhu, MiniTac: An Ultra-Compact 8 mm Vision-Based Tactile Sensor for Enhanced Palpa- tion in Robot-Assisted Minimally Invasive Surgery, IEEE Robotics and Automation Letters (RA-L) 9 (12) (2024) 11170–11177

    work page 2024

  29. [29]

    X. Li, H. Wang, L. Yi, L. J. Guibas, A. L. Abbott, S. Song, Category- level articulated object pose estimation, in: Proceedings of Conference on Computer Vision and Pattern Recognition (CVPR), 2020

    work page 2020

  30. [30]

    Y . Li, W. Du, C. Yu, P. Li, Z. Zhao, T. Liu, C. Jiang, Y . Zhu, S. Huang, Taccel: Scaling Up Vision-based Tactile Robotics via High-performance GPU Simulation, arXiv preprint arXiv:2504.12908

    work page arXiv

  31. [31]

    Y . Li, W. H. Leng, Y . Fang, B. Eisner, D. Held, FlowBotHD: History- Aware Diffuser Handling Ambiguities in Articulated Objects Manipula- tion, in: Conference on Robot Learning (CoRL), 2024

    work page 2024

  32. [32]

    Y . Li, B. Liu, Y . Geng, P. Li, Y . Yang, Y . Zhu, T. Liu, S. Huang, Grasp multiple objects with one hand, IEEE Robotics and Automation Letters (RA-L)

    work page

  33. [33]

    L. Liu, W. Xu, H. Fu, S. Qian, Q. Yu, Y . Han, C. Lu, AKB-48: a real- world articulated object knowledge base, in: Proceedings of Conference on Computer Vision and Pattern Recognition (CVPR), 2022

    work page 2022

  34. [34]

    Y . Liu, B. Jia, R. Lu, J. Ni, S.-C. Zhu, S. Huang, Building Interactable Replicas of Complex Articulated Objects via Gaussian Splatting, in: Proceedings of International Conference on Learning Representations (ICLR), 2025

    work page 2025

  35. [35]

    Lloyd, N

    J. Lloyd, N. F. Lepora, Pose-and-shear-based tactile servoing, Interna- tional Journal of Robotics Research (IJRR) 43 (7) (2024) 1024–1055

    work page 2024

  36. [36]

    J. Lv, Q. Yu, L. Shao, W. Liu, W. Xu, C. Lu, Sagci-system: Towards sample-efficient, generalizable, compositional, and incremental robot learning, in: IEEE International Conference on Robotics and Automation (ICRA), 2022

    work page 2022

  37. [37]

    K. M. Lynch, F. C. Park, Modern Robotics, Cambridge University Press, 2017

    work page 2017

  38. [38]

    Macfarlane, E

    S. Macfarlane, E. A. Croft, Jerk-bounded manipulator trajectory plan- ning: design for real-time applications, IEEE Transactions on Robotics and Automation (T-RA) 19 (1) (2003) 42–52

    work page 2003

  39. [39]

    Mart ´ın-Mart´ın, O

    R. Mart ´ın-Mart´ın, O. Brock, Coupled recursive estimation for online interactive perception of articulated objects, International Journal of Robotics Research (IJRR) 41 (8) (2022) 741–777

    work page 2022

  40. [40]

    Mittal, D

    M. Mittal, D. Hoeller, F. Farshidian, M. Hutter, A. Garg, Articulated ob- ject interaction in unknown scenes with whole-body mobile manipulation, in: IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2022

    work page 2022

  41. [41]

    K. Mo, S. Zhu, A. X. Chang, L. Yi, S. Tripathi, L. J. Guibas, H. Su, Part- net: A large-scale benchmark for fine-grained and hierarchical part-level 3D object understanding, in: Proceedings of Conference on Computer Vision and Pattern Recognition (CVPR), 2019

    work page 2019

  42. [42]

    Moses, M

    C. Moses, M. Noseworthy, L. P. Kaelbling, T. Lozano-P ´erez, N. Roy, Visual Prediction of Priors for Articulated Object Interaction, in: IEEE International Conference on Robotics and Automation (ICRA), 2020

    work page 2020

  43. [43]

    Mur-Artal, J

    R. Mur-Artal, J. M. M. Montiel, J. D. Tardos, ORB-SLAM: A versatile and accurate monocular SLAM system, IEEE Transactions on Robotics (T-RO) 31 (5) (2015) 1147–1163

    work page 2015

  44. [44]

    J. Ni, Y . Liu, R. Lu, Z. Zhou, S.-C. Zhu, Y . Chen, S. Huang, Decom- positional neural scene reconstruction with generative diffusion prior, in: Proceedings of Conference on Computer Vision and Pattern Recognition (CVPR), 2025

    work page 2025

  45. [45]

    Pastor, H

    P. Pastor, H. Hoffmann, T. Asfour, S. Schaal, Learning and generalization of motor skills by learning from demonstration, in: IEEE International Conference on Robotics and Automation (ICRA), 2009

    work page 2009

  46. [46]

    Y . Qin, H. Su, X. Wang, From one hand to multiple hands: Imitation learning for dexterous manipulation from single-camera teleoperation, IEEE Robotics and Automation Letters (RA-L) 7 (4) (2022) 10873– 10881

    work page 2022

  47. [47]

    Y . Qin, W. Yang, B. Huang, K. Van Wyk, H. Su, X. Wang, Y .-W. Chao, D. Fox, Anyteleop: A general vision-based dexterous robot arm-hand teleoperation system, in: Robotics: Science and Systems (RSS), 2023

    work page 2023

  48. [48]

    Sch ¨oller, V

    C. Sch ¨oller, V . Aravantinos, F. Lay, A. Knoll, What the constant velocity model can teach us about pedestrian motion prediction, IEEE Robotics and Automation Letters (RA-L) 5 (2) (2020) 1696–1703

    work page 2020

  49. [49]

    Y . She, S. Wang, S. Dong, N. Sunil, A. Rodriguez, E. Adelson, Ca- ble manipulation with a tactile-reactive gripper, International Journal of Robotics Research (IJRR) 40 (12-14) (2021) 1385–1401

    work page 2021

  50. [50]

    Siciliano, O

    B. Siciliano, O. Khatib, T. Kr ¨oger, Springer handbook of robotics, vol. 200, Springer, 2008

    work page 2008

  51. [51]

    Sikka, H

    P. Sikka, H. Zhang, S. Sutphen, Tactile servo: Control of touch-driven robot motion, in: Experimental Robotics III: The 3rd International Sym- posium, Kyoto, Japan, October 28–30, 1993, Springer, 219–233, 2005

    work page 1993

  52. [52]

    Tekden, M

    A. Tekden, M. P. Deisenroth, Y . Bekiroglu, Grasp Transfer based on Self- Aligning Implicit Representations of Local Surfaces, IEEE Robotics and Automation Letters (RA-L)

    work page

  53. [53]

    Urakami, A

    Y . Urakami, A. Hodgkinson, C. Carlin, R. Leu, L. Rigazio, P. Abbeel, Doorgym: A scalable door opening environment and baseline agent, in: Proceedings of Advances in Neural Information Processing Systems (NeurIPS), 2019

    work page 2019

  54. [54]

    W. Wang, Z. Zhao, Z. Jiao, Y . Zhu, S.-C. Zhu, H. Liu, Rearrange indoor scenes for human-robot co-activity, in: IEEE International Conference on Robotics and Automation (ICRA), 2023

    work page 2023

  55. [55]

    Ward-Cherrier, N

    B. Ward-Cherrier, N. Pestell, L. Cramphorn, B. Winstone, M. E. Gian- naccini, J. Rossiter, N. F. Lepora, The tactip family: Soft optical tactile sensors with 3D-printed biomimetic morphologies, Soft Robotics 5 (2) (2018) 216–227

    work page 2018

  56. [56]

    Weiss, A

    L. Weiss, A. Sanderson, C. Neuman, Dynamic sensor-based control of robots with visual feedback, IEEE Journal on Robotics and Automation 3 (5) (1987) 404–417

    work page 1987

  57. [57]

    Welschehold, C

    T. Welschehold, C. Dornhege, W. Burgard, Learning mobile manipulation actions from human demonstrations, in: IEEE/RSJ International Confer- ence on Intelligent Robots and Systems (IROS), 2017

    work page 2017

  58. [58]

    Wilson, H

    A. Wilson, H. Jiang, W. Lian, W. Yuan, Cable routing and assembly using tactile-driven motion primitives, in: IEEE International Conference on Robotics and Automation (ICRA), 2023

    work page 2023

  59. [59]

    J. Wong, A. Tung, A. Kurenkov, A. Mandlekar, L. Fei-Fei, S. Savarese, R. Mart´ın-Mart´ın, Error-aware imitation learning from teleoperation data for mobile manipulation, in: Conference on Robot Learning (CoRL), 2022

    work page 2022

  60. [60]

    Xiang, Y

    F. Xiang, Y . Qin, K. Mo, Y . Xia, H. Zhu, F. Liu, M. Liu, H. Jiang, Y . Yuan, H. Wang, et al., Sapien: A simulated part-based interactive environment, in: Proceedings of Conference on Computer Vision and Pattern Recogni- tion (CVPR), 2020

    work page 2020

  61. [61]

    Xiong, Q

    H. Xiong, Q. Li, Y .-C. Chen, H. Bharadhwaj, S. Sinha, A. Garg, Learn- ing by watching: Physical imitation of manipulation skills from human videos, in: IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2021

    work page 2021

  62. [62]

    Z. Xu, Z. He, S. Song, Universal manipulation policy network for articu- lated objects, IEEE Robotics and Automation Letters (RA-L) 7 (2) (2022) 2447–2454

    work page 2022

  63. [63]

    Z. Xu, Y . She, LeTac-MPC: Learning Model Predictive Control for Tactile-Reactive Grasping, IEEE Transactions on Robotics (T-RO) 40 (2024) 4376–4395

    work page 2024

  64. [64]

    Y . Yang, X. Wei, N. Zhang, J. Zheng, X. Chen, Q. Wen, X. Luo, C.-Y . Lee, X. Liu, X. Zhang, et al., A non-printed integrated-circuit textile for wireless theranostics, Nature Communications 12 (1) (2021) 4876

    work page 2021

  65. [65]

    Yousef, M

    H. Yousef, M. Boukallel, K. Althoefer, Tactile sensing for dexterous in- hand manipulation in robotics—A review, Sensors and Actuators A: phys- ical 167 (2) (2011) 171–187

    work page 2011

  66. [66]

    Y . Yu, J. Li, S. A. Solomon, J. Min, J. Tu, W. Guo, C. Xu, Y . Song, W. Gao, All-printed soft human-machine interface for robotic physico- chemical sensing, Science Robotics 7 (67) (2022) eabn0495

    work page 2022

  67. [67]

    W. Yuan, S. Dong, E. H. Adelson, Gelsight: High-resolution robot tactile sensors for estimating geometry and force, Sensors 17 (12) (2017) 2762

    work page 2017

  68. [68]

    V . Zeng, T. E. Lee, J. Liang, O. Kroemer, Visual identification of articu- lated object parts, in: IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2021

    work page 2021

  69. [69]

    Zhang, N

    H. Zhang, N. N. Chen, Control of contact via tactile sensing, IEEE Trans- actions on Robotics and Automation (T-RA) 16 (5) (2000) 482–495

    work page 2000

  70. [70]

    Zhang, Z

    T. Zhang, Z. McCarthy, O. Jow, D. Lee, X. Chen, K. Goldberg, P. Abbeel, Deep imitation learning for complex manipulation tasks from virtual re- 20 ality teleoperation, in: IEEE International Conference on Robotics and Automation (ICRA), 2018

    work page 2018

  71. [71]

    Z. Zhao, L. Cui, S. Xie, S. Zhang, Z. Han, L. Ruan, Y . Zhu, B*: Efficient and Optimal Base Placement for Fixed-Base Manipulators, arXiv preprint arXiv:2504.12719

    work page arXiv

  72. [72]

    Z. Zhao, W. Li, Y . Li, T. Liu, B. Li, M. Wang, K. Du, H. Liu, Y . Zhu, Q. Wang, et al., Embedding high-resolution touch across robotic hands enables adaptive human-like grasping, Nature Machine Intelligence 7 (6) (2025) 889–900

    work page 2025

  73. [73]

    Z. Zhao, Y . Li, W. Li, Z. Qi, L. Ruan, Y . Zhu, K. Althoefer, Tac-Man: Tactile-Informed Prior-Free Manipulation of Articulated Objects, IEEE Transactions on Robotics (T-RO) 41 (2024) 538–557

    work page 2024

  74. [74]

    Zheng, S

    B. Zheng, S. Verma, J. Zhou, I. W. Tsang, F. Chen, Imitation learning: Progress, taxonomies and challenges, IEEE Transactions on Neural Net- works and Learning Systems 35 (99) (2022) 1–16

    work page 2022

  75. [75]

    Y . Zhu, T. Gao, L. Fan, S. Huang, M. Edmonds, H. Liu, F. Gao, C. Zhang, S. Qi, Y . N. Wu, et al., Dark, beyond deep: A paradigm shift to cognitive ai with humanlike common sense, Engineering 6 (3) (2020) 310–345. 21

    work page 2020