REVIEW 3 major objections 3 minor 15 references
A Compact 3D-Printed Soft Finger with Cyclic Hydraulic Actuation
T0 review · 3 major / 3 minor · reviewed 2026-08-01 · deepseek-v4-flash
Pith's one-line read A compact 3D-printed hydraulic finger, driven by a miniature cyclic peristaltic pump, can be designed with finite-element analysis and controlled by vision feedback to bend reliably and grasp fragile objects gently.
desk verdict Solid, honest engineering with a working compact hydraulic finger and a clear design flow, but the FEA-guided selection is only tested on the winning geometry, and the missing error bars and replicate prints leave the predictive claim under-supported. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The load-bearing mechanism is cyclic hydraulic actuation inside a dual-chamber bellows: a miniature peristaltic pump moves fluid from one chamber to the other, creating a pressure difference that bends the finger, and reversing the pump bends it the other way. The geometry is chosen by a finite-element screening procedure that models the soft body as a three-term Ogden hyperelastic material, filters simulated states for approximate volume conservation, and compares candidate wall thicknesses, pitch angles, and bellows lengths. A two-angle descriptor, theta1 and theta2, extracted from multiple ArUco markers along the finger, is used to describe contact-induced bending beyond a single tip angl
What would settle it
Print the same finger geometry as a single monolithic part (or with a different joining method) and compare its pressure-angle curve to the FEA prediction; if the curve still falls well below the simulation, then bonding-induced stiffness is not the main cause of the offset and the model is missing another effect. Alternatively, repeatedly print and bond several identical fingers: if their pressure-angle responses vary substantially from one print to the next, the claimed reproducible pipeline from FEA to validation does not transfer across reprints.
Extended reading notes
Core claim
The paper's central claim is a design-and-validation workflow: a two-chamber bellows finger, printed from soft and rigid materials and driven by a miniature peristaltic pump that shuttles water between chambers, bends toward the lower-pressure side. A quasi-static finite-element model using an Ogden hyperelastic description and volume-conservation filtering is used to select pitch angle, wall thickness, and bellows length. Experiments confirm the predicted chamber-coupling and pressure-angle trends, with systematic offsets the paper attributes to the bonded mid-plane, hydraulic losses, and residual bubbles. Vision-feedback control compensates for these offsets, and a two-angle descriptor cap
Load-bearing premise
The finite-element model assumes the finger is a single monolithic printed part, but the actual finger is printed in two halves and bonded with resin; the paper itself attributes the experimental offset to this bonded mid-plane, so if bonding stiffness is large or inconsistent between prints, the simulated geometry choice loses its advantage.
Editorial extensions
If this is right
- Hydraulic soft fingers no longer require bulky external pumps, valves, and long tubing; a single recirculating peristaltic loop can drive bidirectional bending.
- Quasi-static FEA can serve as a geometry-selection tool for such fingers, even when fabrication introduces bonded seams, as long as trend-level agreement is sufficient.
- Vision-feedback control can compensate for residual bubbles, hydraulic losses, and soft-structure settling, keeping the bending angle within a ±1° band.
- The same validated unit-cell geometry can be extended to longer fingers and to two-finger grippers that grasp fragile and deformable objects without visible damage.
- A two-angle representation can distinguish planar contact from curved-surface contact, which is useful for grasp monitoring.
Reading between the lines
- Because the FEA model treats the finger as monolithic while fabrication introduces a bonded mid-plane, a natural next step is to print the finger monolithically or to model the adhesive seam explicitly; the paper's own data suggest this would reduce the systematic offset.
- The sparse, representative parameter sweep implies that a denser optimization over wall thickness and pitch angle could yield geometries with even better bending efficiency for the same footprint.
- The trapped-air-pocket pressure sensing used here could be replaced by direct inline hydraulic transducers if faster, higher-bandwidth control is needed, at the cost of some compactness.
- The two-angle descriptor could be extended to full curvature estimation along the finger, enabling in-hand slip detection or adaptive grasping of objects with unknown shapes.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper presents a compact PolyJet-printed dual-chamber hydraulic soft finger driven by a miniature cyclic peristaltic pump. An Abaqus FEA model with an Ogden hyperelastic material description is used to compare representative pitch angles, wall-thickness distributions, and bellows lengths, selecting T1=1.5 mm, T2=T3=2.0 mm, tan(α)=3/15 as a compromise between bending efficiency, stress, sealing, and manufacturability. Experiments on 3- and 6-unit fingers measure baseline-corrected chamber pressures and vision-based bending angles, showing trend-level agreement with FEA; remaining offsets are attributed mainly to the bonded mid-plane, viscous losses, and residual air. A 12-unit finger is controlled by vision feedback to track step targets within ±1°, and grasping demonstrations on tofu and blueberries show slip-free, visually damage-free contact. The authors frame the work as a reproducible FEA-guided design-to-validation pipeline for compact hydraulic soft fingers.
Significance. If the central claim is accepted, the paper provides a useful workflow for designing miniaturized hydraulic fingers: externally characterized Ogden parameters, no fitting of the FEA to the presented experiments, and an explicit volume-conservation filtering criterion are notable strengths. The modest scope and transparent discussion of model-experiment offsets are commendable. However, the core claim that FEA 'guides selection' of the geometry is not directly validated, because the ranking of candidate designs is never checked against physical experiments. The paper also lacks trial counts and error bars in the validation data. These gaps weaken the reproducibility claim but are addressable within the scope of a revision.
major comments (3)
- [Section II-B, Fig. 2, Section III-A] The FEA screening in Fig. 2 compares three pitch angles (3/15, 3/12, 3/9), three wall-thickness distributions, and two bellows lengths, but only the selected geometry (3/15, T1=1.5, T2=T3=2.0) is fabricated and tested. No rejected candidate is printed to check whether the simulated ordering (e.g., 3/15 > 3/12 > 3/9 in bending efficiency) holds experimentally. Since the paper's central claim is that FEA guides geometry selection, the absence of a rank-reversal test leaves the design-pipeline claim under-supported. Please add at least one non-selected geometry comparison, or explicitly temper the claim to 'the selected geometry behaves acceptably' rather than 'FEA-guided selection is validated.'
- [Section III-A, Fig. 4(B)] The experimental validation reports single pressure-angle curves without error bars, trial counts, or repeated prints. The systematic offset between FEA and experiment is attributed mainly to the bonded mid-plane, but the bond stiffness is neither measured nor incorporated into the FEA model; the attribution is therefore qualitative. If bond-line stiffness is print-dependent or interacts with pitch angle and wall thickness, the transferability of the FEA-selected geometry across reprints is not established. Please provide repeated trials (at least a few prints and repeated runs) and a quantitative discussion or measurement of the bond-line effect.
- [Section III-B, Fig. 5(A.c,d)] The closed-loop tracking results are presented for a single step-up and step-down trial on the 12-unit finger. Given the acknowledged variability from residual bubbles, hydraulic losses, and settling, one trial per target is insufficient to support the word 'repeatable' (Abstract, Section III-B). Please report multiple tracking runs with statistics, or reduce the claim to demonstrate feasibility on the shown trial.
minor comments (3)
- [Section III-C] The grasping characterization is qualitative: 'gentle' is supported only by visual inspection (no visible damage) and the two-angle descriptor is introduced without a quantitative contact-force or deformation metric. This is acceptable for a demonstrative claim but should be phrased as qualitative.
- [Section II-B, Eq. (4)] The volume-mismatch criterion is undefined if min(|ΔVR|,|ΔVL|)=0. Consider adding a small epsilon or a guard for initial states.
- [Section II-D] The pressure measurement uses trapped air pockets in horizontal columns; the text could more explicitly state that the air-pocket pressure is assumed equal to the chamber pressure in static equilibrium. A schematic of the sensor placement would improve reproducibility.
Circularity Check
No circularity found: FEA uses independently characterized material parameters and the validation does not fit the model to the measured curves.
full rationale
The claimed derivation chain is not circular. The Abaqus FEA model uses Ogden hyperelastic parameters taken from an external material-characterization study (ref. [14], Table I), not fitted to the experiments in this paper. Geometry selection is described as a comparative screening over representative candidates (pitch angle, wall thickness, bellows length) under consistent loading, and the selected geometry is then fabricated and tested. The experimental validation compares measured pressure–angle trends with FEA predictions; the paper does not tune the FEA parameters or the simulated pressure inputs to make the predictions match. The acknowledged systematic offsets are attributed to bonding-induced stiffness and hydraulic losses, but this attribution is an explanation of discrepancy, not a parameter fitted to the output. No self-citations are load-bearing, and no predicted quantity is defined in terms of a measured quantity. The absence of rejected-candidate validation and repeated-print statistics weakens the demonstrated strength of the FEA-guided ranking, but that is a validation-gap/correctness concern, not circularity.
Assumptions & free parameters
free parameters (3)
- Wall thicknesses T1/T2/T3 =
1.5 / 2.0 / 2.0 mm
- Pitch angle (tan α) =
3/15 (0.20)
- Bellows unit count =
6 (validation) and 12 (control/grasping)
assumptions (3)
- domain assumption Agilus30 is modeled by the three-term Ogden hyperelastic parameters in Table I, taken from ref [14].
- domain assumption Cyclic hydraulic actuation can be approximated by pressure-driven fluid cavities with opposite-signed volume changes and <3% volume mismatch (Eqs. 2-4).
- domain assumption The split-and-bonded mid-plane does not fundamentally alter the deformation modes captured by the monolithic FEA model.
Cite this review
Pith. "Pith review of A Compact 3D-Printed Soft Finger with Cyclic Hydraulic Actuation." pith.science (2026). https://pith.science/paper/W4JAFQIM
@misc{pith2026260717840,
author = {Pith},
title = {Pith review of: A Compact 3D-Printed Soft Finger with Cyclic Hydraulic Actuation},
year = {2026},
howpublished = {\url{https://pith.science/paper/W4JAFQIM}},
note = {Machine review of arXiv:2607.17840}
}
read the original abstract
Hydraulic soft fingers offer compliant and gentle manipulation, but their practical deployment is limited by bulky fluidic hardware, fabrication complexity, and insufficient design validation. This paper presents a compact 3D-printed soft hydraulic finger driven by a miniature cyclic peristaltic loop. The finger integrates compliant bellows, rigid connectors, and fluidic ports, while an Abaqus fluid-structure model is used to guide selection of wall thickness, pitch angle, and bellows length. The selected design is validated through baseline-corrected chamber-pressure measurements and vision-based angle tracking. Results show that the quasi-static finite-element model captures the main pressure-angle trends, with remaining offsets mainly attributed to bonding-induced stiffness and hydraulic losses. Vision-feedback control further enables repeatable angle tracking over a large bending range. Finally, grasping tests on fragile and deformable objects, including tofu and blueberries, demonstrate gentle, slip-free contact without visible damage. Overall, this work establishes a reproducible pipeline from FEA-guided design to closed-loop validation for compact 3D-printed hydraulic soft fingers.
Figures
Figures from the paper (2 more)
Reference graph
Works this paper leans on
-
[1]
Soft manipulators and grippers: A review,
J. Hughes, U. Culha, F. Giardina, F. Guenther, A. Rosendo, and F. Iida, “Soft manipulators and grippers: A review,”Frontiers in Robotics and AI, vol. 3, p. 69, 2016
2016
-
[2]
Review on research progress of hydraulic powered soft actuators,
H. Shi, K. Tan, B. Zhang, and W. Liu, “Review on research progress of hydraulic powered soft actuators,”Energies, vol. 15, no. 23, p. 9048, 2022
2022
-
[3]
Design and control of pneumatic systems for soft robotics: A simulation approach,
M. S. Xavier, A. J. Fleming, and Y . K. Yong, “Design and control of pneumatic systems for soft robotics: A simulation approach,”IEEE Robotics and Automation Letters, vol. 6, no. 3, pp. 5800–5807, 2021
2021
-
[4]
A dual-mode soft gripper for food packaging,
Z. Wang, K. Or, and S. Hirai, “A dual-mode soft gripper for food packaging,”Robotics and Autonomous Systems, vol. 125, p. 103427, 2020
2020
-
[5]
Me- chanically programmable bend radius for fiber-reinforced soft actuators,
K. C. Galloway, P. Polygerinos, C. J. Walsh, and R. J. Wood, “Me- chanically programmable bend radius for fiber-reinforced soft actuators,” in2013 16th International Conference on Advanced Robotics (ICAR). IEEE, 2013, pp. 1–6
2013
-
[6]
Cyclic hydraulic actuation for soft robotic devices,
R. K. Katzschmann, A. De Maille, D. L. Dorhout, and D. Rus, “Cyclic hydraulic actuation for soft robotic devices,” in2016 IEEE/RSJ Inter- national Conference on Intelligent Robots and Systems (IROS). IEEE, 2016, pp. 3048–3055
2016
-
[7]
Evaluation of lateral and vertical dimensions of micromolds fabricated by a polyjet™ printer,
S. Vijayan, P. Parthiban, and M. Hashimoto, “Evaluation of lateral and vertical dimensions of micromolds fabricated by a polyjet™ printer,” Micromachines, vol. 12, no. 3, p. 302, 2021
2021
-
[8]
A review of 3d printing processes and materials for soft robotics,
Y . L. Yap, S. L. Sing, and W. Y . Yeong, “A review of 3d printing processes and materials for soft robotics,”Rapid Prototyping Journal, vol. 26, no. 8, pp. 1345–1361, 2020
2020
Show all 15 references
-
[9]
The role of 3d printing technologies in soft grippers,
Y . Xin, X. Zhou, H. Bark, and P. S. Lee, “The role of 3d printing technologies in soft grippers,”Advanced materials, vol. 36, no. 34, p. 2307963, 2024
2024
-
[10]
Computational and experimental design exploration of 3d-printed soft pneumatic actuators,
K. G. Demir, Z. Zhang, J. Yang, and G. X. Gu, “Computational and experimental design exploration of 3d-printed soft pneumatic actuators,” Advanced Intelligent Systems, vol. 2, no. 7, p. 2000013, 2020
2020
-
[11]
Printable hy- draulics: A method for fabricating robots by 3d co-printing solids and liquids,
R. MacCurdy, R. Katzschmann, Y . Kim, and D. Rus, “Printable hy- draulics: A method for fabricating robots by 3d co-printing solids and liquids,” in2016 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2016, pp. 3878–3885
2016
-
[12]
A modular and self-contained fluidic engine for soft actuators,
M. A. Bell, B. Gorissen, K. Bertoldi, J. C. Weaver, and R. J. Wood, “A modular and self-contained fluidic engine for soft actuators,”Advanced Intelligent Systems, vol. 4, no. 1, p. 2100094, 2022
2022
-
[13]
Materials and methods for designing 3d- printed soft robot grippers in low-hardness tpu (60a–70a),
K. Meitani and S. Nisar, “Materials and methods for designing 3d- printed soft robot grippers in low-hardness tpu (60a–70a),”IEEE Access, vol. 13, pp. 210 465–210 473, 2025
2025
-
[14]
Material characterisation of additively manufactured elastomers at different strain rates and build orientations,
F. F. Abayazid and M. Ghajari, “Material characterisation of additively manufactured elastomers at different strain rates and build orientations,” Additive Manufacturing, vol. 33, p. 101160, 2020
2020
-
[15]
Fitting hyperelastic models to experimental data,
R. W. Ogden, G. Saccomandi, and I. Sgura, “Fitting hyperelastic models to experimental data,”Computational Mechanics, vol. 34, no. 6, pp. 484– 502, 2004
2004
Reviewed August 1, 2026 · model on record in the stance chip above.
Discussion (0). Sign in to comment.