{"id":"191b12de-e919-4852-b7c7-1458d7dc4e56","arxiv_id":"2506.04547","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"low","formal_verification":"none","parameter_count":3,"one_line_summary":"A two-segment pneumatic soft robot with a folded kirigami skin achieves rectilinear crawling and steering, with fastest forward motion at a quarter-cycle phase offset between segments.","lead":"A soft, worm-like robot that uses an inflatable body covered with a foldable kirigami skin can crawl straight and steer around obstacles on rough surfaces. It shows how directional friction plus body deformation can give limbless robots useful maneuverability for confined spaces.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Dynamic validity of the kirigami skin's friction asymmetry is the load-bearing gap: the paper's own quasi-static data show it nearly vanishes in the AP state, and the authors explicitly defer dynamic tests; a reversed-skin locomotion trial would settle causality.","rationale":"The reader's conditional verdict already captures this uncertainty, and I agree that the weakest load-bearing assumption is the transfer of quasi-static friction asymmetry to the dynamic gait. The robot demonstrably crawls and steers, as supported by videos, force traces, and the speed–force correlation, so there is no basis for rejection. However, the paper's headline contribution is the kirigami skin's friction anisotropy, and the current evidence does not separate that causal mechanism from other dynamic effects. The paper itself flags the quasi-static-to-dynamic gap in the Friction response section, so this is a fair condition rather than an invented objection. A reversed-skin locomotion test is a minimal, decisive experiment that would either confirm the anisotropy's causal role or show that the robot's propulsion is driven by other factors. The post-hoc tuning in Note S1 reinforces the need for such a test, since the model deliberately increases backward friction beyond the experimentally measured values. Overall, the conditional verdict should stand: the empirical platform is credible, but the mechanistic claim requires one additional experiment before full acceptance.","tokens_in":14476,"tokens_out":5697,"duration_ms":72831,"concrete_test":"Run the phi=T/4, f=0.5 Hz rectilinear gait on the PPI10 substrate with the kirigami skin mounted in the reversed orientation, so the scale/fold direction that was 'forward' in Fig. 4 now faces backward. Measure average speed, per-cycle displacement, and direction over 1 minute, n≥5, and compare with the normal orientation. If speed remains approximately 10.8 mm/s with unchanged direction, the locomotion does not depend on the skin's friction anisotropy; if speed drops substantially or direction reverses, the anisotropy is confirmed as the load-bearing propulsion mechanism.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central mechanism claim—that directional kirigami friction anchors the posterior segment so cyclic, phase-shifted inflation produces net forward motion—rests on quasi-static friction coefficients (Fig. 4). The paper reports that for the AP (both segments inflated) configuration mu_R/mu_C ≈ 1, and it states that quasi-static tests may differ from dynamic conditions. Because the phi=T/4 gait includes intervals of dual inflation, the critical anchor state is precisely the one whose asymmetry is unverified. The dynamic pulling-force measurements (Fig. 5C–F) show net traction and correlate with speed, but a whole-robot load cell measures aggregate output, not the direction-dependent skin–ground friction distribution; it cannot distinguish anisotropic kirigami anchoring from other dynamic effects such as foam hysteresis, body-shape changes, or actuator–ground contact. The Note S1 model further weakens the mechanistic support by increasing mu_b1 to 0.5 relative to the measured values; it is a qualitative consistency check, not independent validation. The empirical demonstration that the robot crawls and steers is solid, but the attribution of that crawling to the skin's friction anisotropy is not yet causally established.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents a two-segment pneumatic soft robot covered with a foldable kirigami skin that provides directional friction, and demonstrates rectilinear crawling, steering, and obstacle avoidance on uniform substrates. The authors characterize the kirigami skin's force-displacement response, the actuators' elongation and bending, quasi-static friction coefficients in several inflation states, crawling speed as a function of CPG frequency and phase shift, and dynamic pulling force. They report a speed optimum at a phase shift of phi = T/4 (6.33 mm/s on fine foam, 10.83 mm/s on coarse foam), a correlation between pulling force and speed, and successful teleoperated navigation through obstacles. A three-node spring-damper model in Note S1 is used to argue that the phase shift qualitatively explains the observed speed differences.","tokens_in":14745,"tokens_out":3236,"duration_ms":35661,"significance":"If the central mechanism claim holds, the paper makes a useful contribution to soft crawling robots by combining large deformations, rhythmic actuation, and directional friction in a single platform with steering and sensor-based teleoperation. The experimental characterization is thorough: systematic speed sweeps over frequency and phase, friction measurements with n=6, and dynamic traction tests with reported correlations. The authors honestly flag the quasi-static/dynamic limitation of their friction data and make the MATLAB model code publicly available. The main weakness is causal attribution: the robot demonstrably crawls and steers, but the specific role of the kirigami skin's friction anisotropy is not directly tested, and the theoretical model is fitted rather than independently validated.","major_comments":[{"comment":"The quasi-static friction measurements show that for the AP configuration (both segments inflated) the friction asymmetry ratio mu_R/mu_C is approximately 1 (Fig. 4C and 4E). Since the optimal rectilinear gait phi = T/4 includes intervals in which both segments are inflated, the state that is supposed to provide the anchoring asymmetry is precisely the one whose asymmetry is unverified under dynamic conditions. The text's caveat that 'these quasi-static tests may differ from dynamic conditions' is appropriate but leaves the central mechanism unsupported. A control experiment—for example, crawling with the kirigami skin reversed so that the pop-up orientation opposes the intended direction, or with an isotropic skin—would establish whether the skin's directional friction is causally responsible for the measured propulsion. This is load-bearing because the paper attributes the robot's locomotion to the skin's friction anisotropy.","section":"Friction response, Fig. 4"},{"comment":"The theoretical model is not an independent validation of the phase-shift effect. The friction coefficients are assumed to vary linearly with segment elongation (Eqs. S4-S5), with parameter values that are not derived from the measured friction data. More importantly, the text states that mu_b1 was increased to 0.5 relative to experiments 'to be able to qualitatively reproduce the observed behaviors in experiments' (Note S1). The numerical agreement between Fig. S1 and the experimental tracking in Fig. S2 is therefore a fitted consistency check, not a verification. Please reframe the claim in the Results section that the model is used 'to qualitatively verify whether the phase shift can influence locomotion speed' as a consistency check, or provide parameters measured independently from the reported friction experiments.","section":"Note S1, Eqs. (S4)-(S5)"},{"comment":"The correlation between maximum pulling force and crawling speed (R = 0.77 on fine, 0.86 on coarse) is interpreted as evidence for the anchoring mechanism, but the load cell measures only the aggregate reaction force at the tail. This measurement cannot distinguish the directional skin-ground friction from other dynamic effects such as actuator hysteresis, body-shape changes, or contact between the actuators and the substrate. A test that isolates the skin's contribution—such as a reversed-skin or skinless control—would make the causal claim defensible. In the absence of such a test, the Discussion should explicitly acknowledge that the mechanism is inferred rather than directly established.","section":"Dynamic pulling force, Fig. 5C-F"}],"minor_comments":[{"comment":"Typo: 'cap able' should be 'capable'.","section":"Abstract"},{"comment":"The text contains 'Fig. Fig. 3C' twice; please remove the duplicate 'Fig.'.","section":"Results, Elongation and bending response"},{"comment":"The claim that 'unifying all of these in a crawling robot remains unexplored' is stronger than the cited literature supports, since several recent soft robots combine deformation and steering; consider softening the wording.","section":"Introduction, first paragraph"},{"comment":"The speed data are reported only as mean values; please indicate standard deviations or confidence intervals and state whether the phi = T/4 advantage is statistically significant across the n=5 trials.","section":"Fig. 5A,B"},{"comment":"The model parameters (k = 100 N/m, m_i = 60 g, L0 = 100 mm) are introduced without justification; a sentence explaining how these values relate to the physical prototype would improve reproducibility.","section":"Note S1, Fig. S1"},{"comment":"The link to the MATLAB script is a positive feature; consider also making the raw speed and friction datasets available to strengthen reproducibility.","section":"Data Availability"}],"recommendation":"major_revision","confidential_remarks":"The paper's empirical core—demonstrating a kirigami-skinned soft robot that crawls, steers, and navigates—is solid and likely publishable after revision. The main obstacle is the causal attribution of locomotion to the skin's friction anisotropy, given that the quasi-static AP-state asymmetry vanishes and no reversed-skin control is provided. If the authors can add such a control or substantially temper the mechanistic claims, the paper would meet the standard for acceptance. The theoretical model's fitted nature should also be presented as a consistency check rather than independent verification."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a solid incremental paper, and the new thing is real. Adding fold lines to a kirigami skin so it bends without crumpling is what lets this two-segment pneumatic crawler steer instead of only crawling straight — a genuine step beyond Rafsanjani et al. 2018 and the group's own 2024 earthworm skin. The experimental work is competently executed: clean elongation and bending characterization (up to 74% elongation, bending angles around 49–63°), a repeatable speed optimum at φ = T/4 (6.3 mm/s on fine, 10.8 mm/s on coarse), pulling forces that track speed, and a working assisted-teleoperation obstacle course. They also ship the MATLAB model on GitHub.\n\nThe soft spots, in proportion. First, Note S1 is a fitted consistency check, not validation. The authors openly state they raised μ_b1 to 0.5 relative to the measured values to reproduce the experiments. The model does reproduce the qualitative ordering the experiments show, so it is useful for illustrating the phase-shift effect, but it should be clearly labeled as such. Minor, and mostly already in the text.\n\nSecond, the mechanism gap the stress-test flags is real but not fatal. Quasi-static friction tests show the asymmetry nearly vanishes when both segments are inflated (AP), and the T/4 gait includes dual-inflation intervals. The authors honestly note that quasi-static conditions may differ from dynamic ones. A reversed-skin or skin-removed control trial would settle whether the kirigami anisotropy is actually driving propulsion rather than foam hysteresis or body-shape changes. I think this concern lands, though I would not call it load-bearing for the paper's core empirical claims: the robot demonstrably crawls and steers, and the asymmetry data support the anchoring mechanism in the single-inflation states.\n\nThe citation pattern is appropriate. Prior kirigami skin work and peristaltic crawlers are covered; self-citation is fine here because those priors are the direct baseline.\n\nThe paper is for soft robotics and bioinspired locomotion researchers, especially those building kirigami skins. The design innovation is clear and the experimental claims are reproducible, so it deserves a serious referee. I would send it out, asking for the reversed-skin control and dynamic friction data before acceptance, plus a reframed Note S1.","headline":"Solid incremental advance: the foldable kirigami skin lets this soft crawler steer as well as crawl straight, with clean experiments; the main gaps are a fitted (not validating) model and unverified dynamic friction in the dual-inflation state.","tokens_in":15288,"tokens_out":4400,"would_cite":true,"duration_ms":49434,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper demonstrates that a two-segment pneumatic soft robot wrapped in a foldable kirigami skin with asymmetric friction can crawl rectilinearly and steer, with fastest propulsion at a quarter-period phase shift.","keywords":["soft robotics","kirigami skin","limbless locomotion","peristaltic crawling","asymmetric friction","central pattern generator","pneumatic actuators","assisted teleoperation"],"falsifier":"Measure the robot's forward and backward friction coefficients while the chambers are actively cycled at 0.5 to 1.0 Hz, or run the three-node model with $\\mu_R/\\mu_C = 1$ and show that predicted net displacement drops to zero; either observation would contradict the proposed anchoring mechanism.","tokens_in":14278,"feed_emoji":"🐛","tokens_out":5294,"duration_ms":48877,"temperature":0.7,"pith_summary":"This paper reports a two-segment, pneumatically driven soft robot whose body is wrapped in a foldable kirigami skin with asymmetric friction. The authors aim to show that combining longitudinal elongation and bending of antagonistic chambers with directional gripping of the skin produces rectilinear crawling, in-place rotation, and steering on flat surfaces of moderate roughness. The central quantitative claim is that a phase shift of one quarter period between anterior and posterior actuation maximizes speed, reaching about 6.3 mm/s on fine foam and 10.8 mm/s on coarse foam. If correct, the design is a proof that kirigami skins can support multimodal limbless locomotion without the crumpling that limited earlier stretchable kirigami crawlers. The robot also carries proximity sensors and steers under assisted teleoperation to avoid obstacles.","feed_headline":"Soft robot with kirigami skin crawls at 10.8 mm/s","feed_subtitle":"A one-quarter-cycle delay between its two body segments gives fastest straight-line speed on coarse foam.","key_machinery":"The load-bearing component is the foldable kirigami skin, a $20 \\times 9$ array of unit cells with elliptical cuts and alternating folds that opens in snapping transitions between folded and unfolded states. When the underlying pneumatic chambers elongate, the skin's asymmetric cut pattern gives higher friction in the backward (caudal) direction than in the forward (rostral) direction, preventing slip. The other central mechanism is the central-pattern-generator-based postprocessing controller, which splits each oscillation period into four regions and applies phase shifts $\\varphi = nT/4$ to the four chambers. A three-mass, two-spring model with elongation-dependent friction coefficients carries the argument that phase shift changes the distribution of friction forces and therefore the crawling speed.","core_discovery":"The central discovery is that a foldable, multistable kirigami skin, cut with overlapping elliptical openings and pre-folded hinges, can provide the friction anisotropy needed for two-anchor crawling while remaining compliant enough to bend without kinking. When two antagonistic fiber-reinforced inflatable segments are cyclically inflated with a quarter-period delay, the robot advances rectilinearly, and actuating opposite chambers in the two segments produces turning and on-the-spot rotation. The paper identifies activation order as the reason $\\varphi = T/4$ beats $\\varphi = 3T/4$: with the anterior segment leading, the posterior segment stays anchored against backward slip during propulsion. Measurements of pulling force correlate with speed (R = 0.77 on fine foam, 0.86 on coarse foam), supporting the anchoring-limit explanation. A three-node model with direction-dependent friction reproduces the observed phase-shift ordering.","pith_inferences":["If the dynamic friction asymmetry remains as strong as the quasi-static tests suggest, the same skin geometry could be ported to other actuator arrangements, and the optimal phase shift may shift with the number of segments.","The model's assumption that friction coefficients vary linearly with elongation is an idealization; measuring friction during active inflation would test whether the anchoring mechanism holds under dynamic loading.","The reported optimum at $T/4$ suggests a generic temporal coordination principle for two-anchor crawlers: the rear anchor must stay engaged while the front segment extends, which may transfer to other peristaltic and inchworm robots.","With onboard sensing and teleoperation, the natural next step is closed-loop autonomous path planning; the 18-minute obstacle course time reflects conservative assisted control rather than an inherent speed limit."],"forward_implications":["A phase shift of $T/4$ with anterior-leading activation is the operating point for fastest rectilinear locomotion on both tested surfaces.","Steering can be achieved without additional steering actuators by activating opposite or single chambers in the anterior and posterior segments.","The measured pulling force at 0.5 Hz is highest at $\\varphi = T/4$, and pulling force correlates with crawling speed, so anchoring force can serve as a proxy for locomotion performance.","The multistable skin keeps actuation force in a bounded range during unfolding, allowing large elongations (60 to 74 percent) at 140 kPa.","Obstacle avoidance can be implemented with two proximity sensors and a human-machine interface that overrides user commands within 5 cm of an obstacle."],"supporting_citations":[{"why":"introduces kirigami skins that make a soft actuator crawl, the baseline this work extends by adding folds for steerability","marker":"[23]"},{"why":"provides the design strategy of tailoring kirigami patterns into robot skins with asymmetric friction","marker":"[22]"},{"why":"explains the snapping force-displacement response used to characterize the multistable skin","marker":"[30]"},{"why":"supplies the central pattern generator framework for rhythmic locomotion control","marker":"[27]"},{"why":"provides the discrete-time neural oscillator dynamics used in the central pattern generator","marker":"[28]"},{"why":"gives the biological kinematic basis for peristaltic crawling that motivates the two-segment phase coordination","marker":"[1]"},{"why":"demonstrates antagonistic fiber-reinforced actuators in a peristaltic soft robot, the actuator lineage used here","marker":"[12]"},{"why":"shows that a kirigami skin improves anchoring and locomotion in another substrate, supporting the friction-anchoring role","marker":"[25]"},{"why":"prior soft-skin crawling robot with asymmetric friction whose setae-like skin approach this design extends","marker":"[19]"}],"fun_headline_variants":["Kirigami skin lets soft robot crawl and steer","Soft crawler uses kirigami skin and phase-shifted inflation","Bioinspired soft robot navigates obstacles with kirigami skin","Phase shift drives kirigami-skinned soft robot forward"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The robot only moves forward if, during actual inflation cycles, the skin's friction is genuinely higher in the backward direction than in the forward direction; the paper's quasi-static friction tests leave this open, especially when both segments are inflated, where the asymmetry nearly vanishes.","fun_headline_variants_meta":{"raw":{"variants":["Kirigami skin lets soft robot crawl and steer","Soft crawler uses kirigami skin and phase-shifted inflation","Bioinspired soft robot navigates obstacles with kirigami skin","Phase shift drives kirigami-skinned soft robot forward"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000172,"raw_usage":{"total_tokens":1261,"prompt_tokens":917,"completion_tokens":344,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":533,"completion_tokens_details":{"reasoning_tokens":274}},"tokens_in":533,"tokens_out":344,"duration_ms":4289,"temperature":1.0,"reasoning_tokens":274,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T10:39:55.376267+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the robot's forward and backward friction coefficients while the chambers are actively cycled at 0.5 to 1.0 Hz, or run the three-node model with $\\mu_R/\\mu_C = 1$ and show that predicted net displacement drops to zero; either observation would contradict the proposed anchoring mechanism.","supporting_citations":[{"cited_title":"M., & Bertoldi, K","cited_arxiv_id":null,"evidence_quote":"introduces kirigami skins that make a soft actuator crawl, the baseline this work extends by adding folds for steerability"},{"cited_title":"P., Parvaresh, A ., & Rafsanjani, A","cited_arxiv_id":null,"evidence_quote":"provides the design strategy of tailoring kirigami patterns into robot skins with asymmetric friction"},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"explains the snapping force-displacement response used to characterize the multistable skin"},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"supplies the central pattern generator framework for rhythmic locomotion control"},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"provides the discrete-time neural oscillator dynamics used in the central pattern generator"},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"gives the biological kinematic basis for peristaltic crawling that motivates the two-segment phase coordination"},{"cited_title":"D., Cho, K","cited_arxiv_id":null,"evidence_quote":"demonstrates antagonistic fiber-reinforced actuators in a peristaltic soft robot, the actuator lineage used here"},{"cited_title":"I., & Hammond, F","cited_arxiv_id":null,"evidence_quote":"shows that a kirigami skin improves anchoring and locomotion in another substrate, supporting the friction-anchoring role"},{"cited_title":"D., Moisson de Vaux, J ., Jørgensen, J., & Rafsanjani, A","cited_arxiv_id":null,"evidence_quote":"prior soft-skin crawling robot with asymmetric friction whose setae-like skin approach this design extends"}],"review_version":1}