pith. sign in

arxiv: 2507.10694 · v2 · submitted 2025-07-14 · 💻 cs.RO

Linking Exteroception and Proprioception through Improved Contact Modeling for Soft Growing Robots

Francesco Fuentes , Serigne Diagne , Zachary Kingston , Laura H. Blumenschein This is my paper

Pith reviewed 2026-05-19 04:22 UTC · model grok-4.3

classification 💻 cs.RO
keywords soft growing robotscontact modelingenvironment mappingtrajectory simulationMonte Carlo samplingunstructured environmentstactile sensing
0
0 comments X

The pith

Soft growing robots can explore and map unstructured environments by using modeled contact deformations to select better growth paths.

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

The paper shows that soft growing robots, which bend when they hit things, can turn those bends into information about the shape of the space around them. By first measuring how the robot deforms during simple turns, the authors create a geometry-based simulator to predict full trajectories in flat environments. They then use Monte Carlo sampling inside that simulator to pick the next best direction for the robot to grow, based on what it has already sensed. If the model holds, this lets the robot map unknown areas quickly and with little active control. Tests in both simple and irregular setups show the chosen actions get close to the best possible ones rapidly.

Core claim

By characterizing collision behavior during discrete turns and building a geometry-based simulator, the work demonstrates that soft growing robots can map unknown 2D environments by using Monte Carlo sampling to estimate the optimal next deployment from current knowledge, with the selection method approaching ideal actions over uniform and non-uniform environments.

What carries the argument

A geometry-based simulator that predicts robot trajectories in 2D environments from characterized collision deformations during turns.

If this is right

  • Soft growing robots become viable tools for exploration and mapping in unstructured settings using passive deformation sensing.
  • The simulator enables prediction of robot behavior without full dynamic simulation.
  • Optimal deployment selection improves efficiency in learning environment structure from tactile data.
  • Contact modeling links external sensing with the robot's own shape changes for better navigation.

Where Pith is reading between the lines

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

  • This modeling could extend to three-dimensional environments or continuous growth paths with further validation experiments.
  • Similar techniques might improve mapping in other compliant robot designs that rely on passive deformation.
  • Real-world factors like friction variations or growth speed changes could be tested as extensions to the current discrete-turn characterization.

Load-bearing premise

The collision behaviors measured in discrete turns will continue to predict robot motion accurately during continuous navigation through unknown and changing spaces.

What would settle it

Running the robot in a new non-uniform environment and finding that actual paths differ substantially from simulator predictions, causing the Monte Carlo method to select poor actions instead of approaching optimal ones.

Figures

Figures reproduced from arXiv: 2507.10694 by Francesco Fuentes, Laura H. Blumenschein, Serigne Diagne, Zachary Kingston.

Figure 1
Figure 1. Figure 1: Leveraging knowledge of passive kinematics for soft growing structures under contact, we can predict robot deformations and extract information about occupied and unoccupied regions of an environment, allowing uses in exploring and mapping unknown environments. kinematic equations and predicts the path of growing robots through multi-object interactions in cluttered environments. This simulation leverages … view at source ↗
Figure 2
Figure 2. Figure 2: Vine robot movement under contact with the environment can be predicted based on the initial contact geometry and friction, and will pivot about a previous point while the tip follows the wall. The motion can generate two pieces of information, the wall location through swept contact, and free space, through swept area. grow as seen in [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Predicted regions of contact behavior with a straight vine robot growing into a flat wall. Dashed lines plots the critical contact angle between pivoting and axial buckling behaviors as a function of slenderness ratio using µ = 0.3, Equation (3). to pressure is larger than the axial force required to buckle. Given that both tangential and axial force components result from pressure, a force balance can be … view at source ↗
Figure 4
Figure 4. Figure 4: The four post-collision morphologies that occur when a vine robot with a turn grows into a wall, each with a physical demonstration of its movement. A) Forward Pivot, B) Pivot-at-Turn, C) Indeterminate Buckling, and D) Backward Pivot [PITH_FULL_IMAGE:figures/full_fig_p005_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Notation used within the passive kinematic model for vine robots with a turn, including forces, robot and contact geometry, and labels for important points. Cartesian frame and robot orientation used in derivation is shown as well. a function of the robot and environment features (as seen in [PITH_FULL_IMAGE:figures/full_fig_p005_5.png] view at source ↗
Figure 7
Figure 7. Figure 7: Experimental results for predicting the morphologies plotted onto the 3-dimensional region sliced by the critical contact angle surfaces (Equations (13) to (15)) to the examples shown in [PITH_FULL_IMAGE:figures/full_fig_p006_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Variables to determine the location of a pivot change, derived purely from tip collision angle and robot length. body from the pivot point to the contact point. Before any collisions occur, the first θB is the initial launch angle. This term only gets updated once ⃗B is updated. As the vine robot tip slides against the wall, we can plot the point of contact, C⃗ i , using measurements of contact angle to up… view at source ↗
Figure 9
Figure 9. Figure 9: Visibility graphs overlaid onto a manufactured environment with the final shape of the vine robot highlighted in red. Vine robot paths under passive deformation follow paths within the visibility graph. 3.1.1 Visibility Graphs as Environment Decompositions A critical observation for the multi-contact simulation of vine robots is that all final vine robot shapes which include collision must lie on the visib… view at source ↗
Figure 10
Figure 10. Figure 10: Overview of how turning is integrated into the visibility graph architecture. A) When the vine robot turns, a new vertex with all visible connections is added to the visibility graph. B) As the vine robot grows, its interactions can cause the vertex to be moved and updated in the main visibility graph. follow during its navigation. Revisiting the kinematic model in Section 2, a straight vine robot collidi… view at source ↗
Figure 12
Figure 12. Figure 12: A) Distribution of the variables used to create the full action space of 570 deployments, showing one example, and B) the rubric used to score deployments and beliefs. selected (Fuentes and Blumenschein 2023). Of course, this perfect selection can only be achieved by already having full knowledge of the environment, so to map without any initial prior knowledge, we need an approach that integrates the col… view at source ↗
Figure 13
Figure 13. Figure 13: Example of a single loop of the algorithm. First, A) the vine robot is deployed and B) its measurements are used to create the individual belief, which then C) updates the cumulative belief. From this, D) the belief is sampled to create potential environments and cells are grouped to E) select obstacles for Monte Carlo (MC) environments (only one example is pictured). F) The entire action space is simulat… view at source ↗
Figure 14
Figure 14. Figure 14: An expanded example of results for a uniform environment using an action space with no turns. A) Selected deployment actions before launching and B) the same selection of deployments, after launch, including the new information gathered. Information from earlier deployments is visually “on top of” that from later deployments. C) Performance comparison of our proposed algorithm compared to ideal selections… view at source ↗
Figure 15
Figure 15. Figure 15: Example results of MC algorithm on uniform environments with straight vine robots. (Left) Gathered information from 20 vine robots. (Center) Cumulative score compared to Ideal and Random. (Right) Example MC environment after final loop. the boundary of the environment, and an action space of 50 straight robot launch angles at each starting position, totaling 100 unique actions [PITH_FULL_IMAGE:figures/fu… view at source ↗
Figure 16
Figure 16. Figure 16: Example results of MC algorithm on uniform environments with vine robots with turns. (Left) Gathered information from 20 vine robots. (Center) Cumulative score compared to Ideal and Random. (Right) Example MC environment after final loop. are reunited, their summed success is still well below the original approach, though still outside the Random selection range. However, different actions were selected b… view at source ↗
Figure 17
Figure 17. Figure 17: Example results of MC algorithm on non-uniform environments with straight vine robots. (Left) Gathered information from 20 vine robots. (Center) Cumulative score compared to Ideal and Random. (Right) Example MC environment after final loop. in the obstacles taking up more space in the boundary than the squares, so scores may not be completely comparable to square environments, as previous work determined … view at source ↗
Figure 18
Figure 18. Figure 18: Example results of MC algorithm on non-uniform environments with vine robots with turns. (Left) Gathered information from 20 vine robots. (Center) Cumulative score compared to Ideal and Random. (Right) Example MC environment after final loop [PITH_FULL_IMAGE:figures/full_fig_p017_18.png] view at source ↗
Figure 19
Figure 19. Figure 19: Setup of the physical environment with non-uniform obstacles, including a double rail for positioning and orienting the vine robot base such that deployments are accurately angled to grow through one of the three launch positions (marked in red). 4.3.1 Physical Environment Setup A pegboard with 2.54 cm (1 in) spaced holes was used to create an environment with non-uniform obstacles. Cardboard walls were c… view at source ↗
Figure 20
Figure 20. Figure 20: Example of how each video frame is transferred into realistic sensor data streams for mapping. A) A frame of the video. B) The mask created comparing the video frame to a reference frame created before deployment. C) Centerline of the mask’s pixels and D) piecewise linear regression of centerline to create a piecewise spline. Finally, E) the accumulation of each frame’s collision angle and robot length (n… view at source ↗
Figure 21
Figure 21. Figure 21: Results of the physical experiment. A) Wall collision (red) and swept area (blue-gray) data extracted from each deployment. B) Scores of the belief and MC environments across the deployments, and C) a generated MC environment from the final cumulative belief. 5.2 Geometric Simulation We see this is similarly true in the construction and testing of the simulator. The largest prediction errors occurred excl… view at source ↗
read the original abstract

Passive deformation due to compliance is a commonly used benefit of soft robots, providing opportunities to achieve robust actuation with few active degrees of freedom. Soft growing robots in particular have shown promise in navigation of unstructured environments due to their passive deformation. If their collisions and subsequent deformations can be better understood, soft robots could be used to understand the structure of the environment from direct tactile measurements. In this work, we propose the use of soft growing robots as mapping and exploration tools. We do this by first characterizing collision behavior during discrete turns, then leveraging this model to develop a geometry-based simulator that models robot trajectories in 2D environments. Finally, we demonstrate the model and simulator validity by mapping unknown environments using Monte Carlo sampling to estimate the optimal next deployment given current knowledge. Over both uniform and non-uniform environments, this selection method rapidly approaches ideal actions, showing the potential for soft growing robots in unstructured environment exploration and mapping.

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

1 major / 1 minor

Summary. The paper claims that soft growing robots can serve as mapping and exploration tools in unstructured environments by characterizing passive collision deformations during discrete turns, constructing a geometry-based 2D simulator from those observations, and then applying Monte Carlo sampling over candidate deployments to select actions that map unknown environments; it reports that this selection rapidly approaches ideal actions in both uniform and non-uniform settings.

Significance. If the simulator predictions hold under continuous growth, the work would provide a concrete route to using passive compliance for tactile exteroception in soft robots, reducing reliance on active sensing for navigation in unknown spaces.

major comments (1)
  1. [Abstract (demonstration of model and simulator validity)] The central claim that the geometry-based simulator supports Monte Carlo estimation of optimal deployments in unknown continuous environments rests on the untested assumption that discrete-turn collision data generalizes to sustained growth trajectories with repeated contacts. No quantitative validation, error metrics, or continuous-trajectory experiments are described that would confirm this generalization (see the demonstration paragraph in the abstract).
minor comments (1)
  1. [Abstract] The abstract refers to 'improved contact modeling' without stating the specific modeling steps or baseline comparisons used.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the thoughtful review and for identifying a key point about the scope of our simulator validation. We address the major comment below and will revise the manuscript to better clarify the relationship between discrete characterization and continuous trajectories.

read point-by-point responses

  1. Referee: [Abstract (demonstration of model and simulator validity)] The central claim that the geometry-based simulator supports Monte Carlo estimation of optimal deployments in unknown continuous environments rests on the untested assumption that discrete-turn collision data generalizes to sustained growth trajectories with repeated contacts. No quantitative validation, error metrics, or continuous-trajectory experiments are described that would confirm this generalization (see the demonstration paragraph in the abstract).

    Authors: We agree that the model characterization relies on discrete-turn collision data and that the Monte Carlo demonstrations occur within the geometry-based simulator rather than through direct physical validation of long continuous trajectories. The simulator constructs continuous paths by sequentially applying the observed deformation rules at each contact point, which we posit is sufficient for 2D mapping because local collision geometry dominates the behavior. The abstract demonstration therefore shows that actions selected via Monte Carlo sampling in the simulator approach ideal performance in both uniform and non-uniform simulated environments. We acknowledge the absence of quantitative error metrics (e.g., trajectory deviation or contact-force residuals) between simulator predictions and real-robot continuous growth with repeated contacts. In the revised manuscript we will add a dedicated subsection in the results that reports such metrics from supplementary continuous-growth trials and will update the abstract demonstration paragraph to explicitly state the scope of the current validation. revision: yes

Circularity Check

0 steps flagged

No load-bearing circularity; modeling chain rests on empirical characterization

full rationale

The derivation proceeds by characterizing collision behavior in discrete turns, building a geometry-based simulator from that data, and then applying Monte Carlo sampling for deployment selection. No equations, fitted parameters, or self-citations are shown to reduce any reported performance metric to its own inputs by construction. The generalization from discrete to continuous trajectories is an explicit modeling assumption rather than a definitional loop, and the central claims remain independently testable against real robot behavior in the described environments.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only view yields no identifiable free parameters, axioms, or invented entities; the central claims rest on unstated modeling assumptions about collision generalization.

pith-pipeline@v0.9.0 · 5693 in / 1054 out tokens · 40657 ms · 2026-05-19T04:22:09.228169+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

51 extracted references · 51 canonical work pages

  1. [1]

    arXiv preprint arXiv:2306.01166

    Agharese N and Okamura AM (2023) Configuration and fabrication of preformed vine robots. arXiv preprint arXiv:2306.01166

    work page arXiv 2023

  2. [2]

    IEEE Transactions on Robotics

    Akinola I, Xu J, Carius J, Fox D and Narang Y (2025) Tacsl: A library for visuotactile sensor simulation and learning. IEEE Transactions on Robotics

    work page 2025

  3. [3]

    IEEE Transactions on Haptics

    Athar S, Zhang X, Ueda J, Zhao Y and She Y (2025) Vibtac: A high-resolution high-bandwidth tactile sensing finger for multi-modal perception in robotic manipulation. IEEE Transactions on Haptics

    work page 2025

  4. [4]

    Soft Matter 19: 44--56

    Baaij T, Holkenborg MK, Stölzle M, van der Tuin D, Naaktgeboren J, Babuška R and Santina CD (2023) Learning 3d shape proprioception for continuum soft robots with multiple magnetic sensors. Soft Matter 19: 44--56. doi:10.1039/D2SM00914E

    work page doi:10.1039/d2sm00914e 2023

  5. [5]

    Frontiers in Robotics and AI 7: 548266

    Blumenschein LH, Coad MM, Haggerty DA, Okamura AM and Hawkes EW (2020) Design, modeling, control, and application of everting vine robots. Frontiers in Robotics and AI 7: 548266

    work page 2020

  6. [6]

    IEEE Transactions on Robotics 38: 1820--1840

    Blumenschein LH, Koehler M, Usevitch NS, Hawkes EW, Rucker DC and Okamura AM (2022) Geometric solutions for general actuator routing on inflated-beam soft growing robots. IEEE Transactions on Robotics 38: 1820--1840. doi:10.1109/TRO.2021.3115230

    work page doi:10.1109/tro.2021.3115230 2022

  7. [7]

    In: 2018 IEEE International Conference on Soft Robotics (RoboSoft)

    Blumenschein LH, Usevitch NS, Do BH, Hawkes EW and Okamura AM (2018) Helical actuation on a soft inflated robot body. In: 2018 IEEE International Conference on Soft Robotics (RoboSoft). IEEE. ISBN 978-1-5386-4516-1, pp. 245--252. doi:10.1109/ROBOSOFT.2018.8404927

    work page doi:10.1109/robosoft.2018.8404927 2018

  8. [8]

    In: 2025 IEEE 8th International Conference on Soft Robotics (RoboSoft)

    Chen L, Gao Y, Wang S, Fuentes F, Blumenschein LH and Kingston Z (2025) Physics-grounded differentiable simulation for soft growing robots. In: 2025 IEEE 8th International Conference on Soft Robotics (RoboSoft). IEEE

    work page 2025

  9. [9]

    IEEE Robotics and Automation Magazine 27: 120--132

    Coad MM, Blumenschein LH, Cutler S, Zepeda JAR, Naclerio ND, El-Husseiny H, Mehmood U, Ryu JH, Hawkes EW and Okamura AM (2020) Vine robots: Design, teleoperation, and deployment for navigation and exploration. IEEE Robotics and Automation Magazine 27: 120--132. doi:10.1109/MRA.2019.2947538

    work page doi:10.1109/mra.2019.2947538 2020

  10. [10]

    AIAA Journal 1(7): 1652--1655

    Comer RL and Levy S (1963) Deflections of an inflated circular-cylindrical cantilever beam. AIAA Journal 1(7): 1652--1655. doi:10.2514/3.1873

    work page doi:10.2514/3.1873 1963

  11. [11]

    National Air and Space Administration (NASA) Technical Note 3466: 1--19

    Fichter W (1966) A theory for inflated thin-wall cylindrical beams. National Air and Space Administration (NASA) Technical Note 3466: 1--19

    work page 1966

  12. [12]

    and Fallah, S

    Frias-Miranda E, Srivastava A, Wang S and Blumenschein LH (2023) Vine robot localization via collision. In: 2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). pp. 2515--2521. doi:10.1109/IROS55552.2023.10342238

    work page doi:10.1109/iros55552.2023.10342238 2023

  13. [13]

    and Fallah, S

    Fuentes F and Blumenschein LH (2023) Mapping unknown environments through passive deformation of soft, growing robots. In: IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). pp. 2522--2527. doi:10.1109/IROS55552.2023.10341705

    work page doi:10.1109/iros55552.2023.10341705 2023

  14. [14]

    The International Journal of Robotics Research 39(14): 1724--1738

    Greer JD, Blumenschein LH, Alterovitz R, Hawkes EW and Okamura AM (2020) Robust navigation of a soft growing robot by exploiting contact with the environment. The International Journal of Robotics Research 39(14): 1724--1738

    work page 2020

  15. [15]

    In: IEEE International Conference on Robotics and Automation

    Greer JD, Blumenschein LH, Okamura AM and Hawkes EW (2018) Obstacle-aided navigation of a soft growing robot. In: IEEE International Conference on Robotics and Automation. IEEE. ISBN 9781538630815, pp. 4165--4172. doi:10.1109/ICRA.2018.8460777

    work page doi:10.1109/icra.2018.8460777 2018

  16. [16]

    In: 2021 IEEE 4th International Conference on Soft Robotics (RoboSoft)

    Gruebele AM, Zerbe AC, Coad MM, Okamura AM and Cutkosky MR (2021) Distributed sensor networks deployed using soft growing robots. In: 2021 IEEE 4th International Conference on Soft Robotics (RoboSoft). IEEE, pp. 66--73

    work page 2021

  17. [17]

    and Jackson, Brian E

    Haggerty DA, Naclerio ND and Hawkes EW (2019) Characterizing environmental interactions for soft growing robots. IEEE International Conference on Intelligent Robots and Systems : 3335--3342doi:10.1109/IROS40897.2019.8968137

    work page doi:10.1109/iros40897.2019.8968137 2019

  18. [18]

    Science Robotics 2: eaan3028

    Hawkes EW, Blumenschein LH, Greer JD and Okamura AM (2017) A soft robot that navigates its environment through growth. Science Robotics 2: eaan3028. doi:10.1126/scirobotics.aan3028

    work page doi:10.1126/scirobotics.aan3028 2017

  19. [19]

    Science Robotics 7: 2119

    Heiden A, Preninger D, Lehner L, Baumgartner M, Drack M, Woritzka E, Schiller D, Gerstmayr R, Hartmann F and Kaltenbrunner M (2022) 3d printing of resilient biogels for omnidirectional and exteroceptive soft actuators. Science Robotics 7: 2119. doi:10.1126/scirobotics.abk2119

    work page doi:10.1126/scirobotics.abk2119 2022

  20. [20]

    In: International Conference on Robotics and Automation

    Hoepflinger MA, Remy CD, Hutter M, Spinello L and Siegwart R (2010) Haptic terrain classification for legged robots. In: International Conference on Robotics and Automation. ISBN 9781424450381, pp. 2828--2833. doi:10.1109/ROBOT.2010.5509309

    work page doi:10.1109/robot.2010.5509309 2010

  21. [21]

    IEEE Robotics and Automation Letters 6(4): 8694--8701

    Jamil B, Yoo G, Choi Y and Rodrigue H (2021) Proprioceptive soft pneumatic gripper for extreme environments using hybrid optical fibers. IEEE Robotics and Automation Letters 6(4): 8694--8701

    work page 2021

  22. [22]

    2021 , url =

    Jitosho R, Agharese N, Okamura A and Manchester Z (2021) A dynamics simulator for soft growing robots. In: Proceedings - IEEE International Conference on Robotics and Automation, volume 2021-May. Institute of Electrical and Electronics Engineers Inc. ISBN 9781728190778, pp. 11775--11781. doi:10.1109/ICRA48506.2021.9561420

    work page doi:10.1109/icra48506.2021.9561420 2021

  23. [23]

    Technical report, National Aeronautics and Space Administration

    Leonard RW, Brooks GW and McComb Jr HG (1960) Structural considerations of inflatable reentry vehicles. Technical report, National Aeronautics and Space Administration

    work page 1960

  24. [24]

    Nature Communications 13: 79

    Liu M, Zhang Y, Wang J, Qin N, Yang H, Sun K, Hao J, Shu L, Liu J, Chen Q, Zhang P and Tao TH (2022) A star-nose-like tactile-olfactory bionic sensing array for robust object recognition in non-visual environments. Nature Communications 13: 79. doi:10.1038/s41467-021-27672-z

    work page doi:10.1038/s41467-021-27672-z 2022

  25. [25]

    In: 2024 IEEE International Symposium on Safety Security Rescue Robotics (SSRR)

    McFarland C, Dhawan A, Kumari R, Council C, Coad M and Hanson N (2024) Field insights for portable vine robots in urban search and rescue. In: 2024 IEEE International Symposium on Safety Security Rescue Robotics (SSRR). IEEE, pp. 190--197

    work page 2024

  26. [26]

    ISBN 9781467380263, pp

    McGarey P, MacTavish K, Pomerleau F and Barfoot TD (2016) The line leading the blind: Towards nonvisual localization and mapping for tethered mobile robots. ISBN 9781467380263, pp. 4799--4806. doi:10.1109/ICRA.2016.7487684

    work page doi:10.1109/icra.2016.7487684 2016

  27. [27]

    International Journal of Robotics Research 36: 1363--1386

    McGarey P, MacTavish K, Pomerleau F and Barfoot TD (2017) Tslam: Tethered simultaneous localization and mapping for mobile robots. International Journal of Robotics Research 36: 1363--1386. doi:10.1177/0278364917732639

    work page doi:10.1177/0278364917732639 2017

  28. [28]

    Autonomous Robots 42: 1443--1458

    Patel R, Cox R and Correll N (2018) Integrated proximity, contact and force sensing using elastomer-embedded commodity proximity sensors. Autonomous Robots 42: 1443--1458. doi:10.1007/s10514-018-9751-4

    work page doi:10.1007/s10514-018-9751-4 2018

  29. [29]

    International Journal of Advanced Robotic Systems 2

    Pearson MJ, Gilhespy I, Melhuish C, Mitchinson B, Nibouche M, Pipe AG and Prescott TJ (2005) A biomimetic haptic sensor. International Journal of Advanced Robotic Systems 2. doi:10.5772/5774

    work page doi:10.5772/5774 2005

  30. [30]

    Philosophical Transactions of the Royal Society B: Biological Sciences 366: 2989--2995

    Prescott TJ, Diamond ME and Wing AM (2011) Active touch sensing. Philosophical Transactions of the Royal Society B: Biological Sciences 366: 2989--2995. doi:10.1098/rstb.2011.0167

    work page doi:10.1098/rstb.2011.0167 2011

  31. [31]

    Physiological Reviews 92: 1651--1697

    Proske U and Gandevia SC (2012) The proprioceptive senses: Their roles in signaling body shape, body position and movement, and muscle force. Physiological Reviews 92: 1651--1697. doi:10.1152/physrev.00048.2011.-This

    work page doi:10.1152/physrev.00048.2011.-this 2012

  32. [32]

    Available: https://doi.org/10.1007/s43154-021-00065-2

    Roberts P, Zadan M and Majidi C (2021) Soft tactile sensing skins for robotics. Current Robotics Reports 2: 343--354. doi:10.1007/s43154-021-00065-2

    work page doi:10.1007/s43154-021-00065-2 2021

  33. [33]

    IEEE/ASME Transactions on Mechatronics 26: 1877--1885

    Scharff RBN, Fang G, Tian Y, Wu J, Geraedts JMP and Wang CC (2021) Sensing and reconstruction of 3-d deformation on pneumatic soft robots. IEEE/ASME Transactions on Mechatronics 26: 1877--1885. doi:10.1109/TMECH.2021.3078263

    work page doi:10.1109/tmech.2021.3078263 2021

  34. [34]

    International Journal of Robotics Research 40: 1385--1401

    She Y, Wang S, Dong S, Sunil N, Rodriguez A and Adelson E (2021) Cable manipulation with a tactile-reactive gripper. International Journal of Robotics Research 40: 1385--1401. doi:10.1177/02783649211027233

    work page doi:10.1177/02783649211027233 2021

  35. [35]

    In: 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

    Shih B, Drotman D, Christianson C, Huo Z, White R, Christensen HI and Tolley MT (2017) Custom soft robotic gripper sensor skins for haptic object visualization. In: 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE. ISBN 978-1-5386-2682-5, pp. 494--501. doi:10.1109/IROS.2017.8202199

    work page doi:10.1109/iros.2017.8202199 2017

  36. [36]

    In: 2018 IEEE International Conference on Robotics and Automation (ICRA)

    Soter G, Conn A, Hauser H and Rossiter J (2018) Bodily aware soft robots: Integration of proprioceptive and exteroceptive sensors. In: 2018 IEEE International Conference on Robotics and Automation (ICRA). IEEE. ISBN 978-1-5386-3081-5, pp. 2448--2453. doi:10.1109/ICRA.2018.8463169

    work page doi:10.1109/icra.2018.8463169 2018

  37. [37]

    Science Robotics 4: 1488

    Thuruthel TG, Shih B, Laschi C and Tolley MT (2019) Soft robot perception using embedded soft sensors and recurrent neural networks. Science Robotics 4: 1488. doi:10.1126/scirobotics.aav1488

    work page doi:10.1126/scirobotics.aav1488 2019

  38. [38]

    Royal Society Open Science 3

    Umedachi T, Kano T, Ishiguro A and Trimmer BA (2016) Gait control in a soft robot by sensing interactions with the environment using self-deformation. Royal Society Open Science 3. doi:10.1098/rsos.160766

    work page doi:10.1098/rsos.160766 2016

  39. [39]

    In: 2024 IEEE 7th International Conference on Soft Robotics (RoboSoft)

    Wang S and Blumenschein LH (2024) Refined modeling for serial pneumatic artificial muscles enables model-based actuation design. In: 2024 IEEE 7th International Conference on Soft Robotics (RoboSoft). IEEE, pp. 800--807

    work page 2024

  40. [40]

    In: American Control Conference

    Wang T, Whitman J, Travers M and Choset H (2020) Directional compliance in obstacle-aided navigation for snake robots. In: American Control Conference. ISBN 9781538682661, pp. 2458--2463. doi:10.0/Linux-x86_64

    work page 2020

  41. [41]

    IEEE Robotics and Automation Letters 5: 2666--2673

    Watson C and Morimoto TK (2020) Permanent magnet-based localization for growing robots in medical applications. IEEE Robotics and Automation Letters 5: 2666--2673. doi:10.1109/LRA.2020.2972890

    work page doi:10.1109/lra.2020.2972890 2020

  42. [42]

    Sensors and Actuators A: Physical 253: 188--197

    White EL, Case JC and Kramer RK (2017) Multi-mode strain and curvature sensors for soft robotic applications. Sensors and Actuators A: Physical 253: 188--197. doi:10.1016/j.sna.2016.11.031

    work page doi:10.1016/j.sna.2016.11.031 2017

  43. [43]

    IEEE Robotics and Automation Letters

    Wu H, Sun F, Huang C, Huang H and Chu Z (2025) Soft growing robot explore unknown environments through obstacle interaction. IEEE Robotics and Automation Letters

    work page 2025

  44. [44]

    Sensors 17: 2762

    Yuan W, Dong S and Adelson E (2017) Gelsight: High-resolution robot tactile sensors for estimating geometry and force. Sensors 17: 2762. doi:10.3390/s17122762

    work page doi:10.3390/s17122762 2017

  45. [45]

    In: 2015 IEEE International Conference on Robotics and Automation (ICRA)

    Yuan W, Li R, Srinivasan MA and Adelson EH (2015) Measurement of shear and slip with a gelsight tactile sensor. In: 2015 IEEE International Conference on Robotics and Automation (ICRA). IEEE. ISBN 978-1-4799-6923-4, pp. 304--311. doi:10.1109/ICRA.2015.7139016

    work page doi:10.1109/icra.2015.7139016 2015

  46. [46]

    IEEE Transactions on Robotics

    Zhang D, Fan W, Lin J, Li H, Cong Q, Liu W, Lepora NF and Luo S (2025) Design and benchmarking of a multi-modality sensor for robotic manipulation with gan-based cross-modality interpretation. IEEE Transactions on Robotics

    work page 2025

  47. [47]

    IEEE Sensors Journal 22(22): 21410--21427

    Zhang S, Chen Z, Gao Y, Wan W, Shan J, Xue H, Sun F, Yang Y and Fang B (2022) Hardware technology of vision-based tactile sensor: A review. IEEE Sensors Journal 22(22): 21410--21427

    work page 2022

  48. [48]

    , " * write output.state after.block = add.period write newline

    ENTRY address author booktitle chapter doi edition editor eid howpublished institution isbn journal key month note number organization pages publisher school series title type url volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts #0 'before.all := #1 'mid...

    work page

  49. [49]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize ":" * " " *...

    work page

  50. [50]

    , " * write output.state after.block = add.period write newline

    ENTRY address archive author booktitle chapter doi edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type url volume year label INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts #0 'before.all := #1 'mid.sentence := #2 'af...

    work page

  51. [51]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

    work page