{"id":"b339e2f7-09be-48e4-a86a-f9b02476f0b0","arxiv_id":"2501.09746","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"Multiple contacts that each obey speed-independent friction can collectively produce a controllable, speed-dependent friction force, with system speed set by the median contact speed.","lead":"A spinning carousel with ten independently controlled wheels shows that dry friction can be shaped in real time: the system's speed settles at the median wheel speed, and the net friction force can be tuned by choosing the spread of wheel speeds. The work reveals a simple control principle for sliding friction that could inform robot locomotion and smart surfaces.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The unweighted median and zero-force predictions depend on unmeasured equal normal forces; the 91% residual friction and 6.5% slope offset may signal load imbalance.","rationale":"The reader's weakest assumption and my independent read point to the same load-bearing issue: the unweighted median and equal-friction-magnitude predictions are not directly verified. This concern is concrete and testable, and it is acknowledged by the authors only in a general discussion of weighted cases, not tested against the present data. I do not think it warrants rejection: the experimental demonstrations of median behavior and controllable force-speed curves are internally consistent and the deviations are small (9% residual force, 6.5% slope offset). A conditional acceptance that asks for a direct measurement of per-wheel normal forces—or at least a quantified bound on load imbalance—would be appropriate. Other potential concerns (e.g., motor speed droop, backlash, the sign(0) ambiguity) are either addressed by the authors' characterization or are second-order relative to the equal-normal-force issue. Thus the verdict remains CONDITIONAL, and my stress-test does not change the reader's conclusion.","tokens_in":10056,"tokens_out":12759,"duration_ms":156014,"concrete_test":"Place a calibrated load cell or accurate scale under each of the ten wheels to measure the static normal force N_i individually (or instrument each spoke with an in-line load cell). Then, using the recorded wheel speeds from the 660 equilibrium experiments, compute the weighted median (with weights N_i) and the weighted force sum; compare the predicted equilibrium speeds and force-speed curves against the unweighted model. Specifically, check whether using the measured N_i removes the 6.5% slope offset in Fig. 3b and reduces the residual friction in Fig. 6 from 9% toward zero. If all N_i are equal within 2%, the equal-normal-force assumption is validated; if not, the paper must report weighted results.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central model (Eq. 1) and the equilibrium-median prediction require every contact to contribute equal friction magnitude µmg/N. The only evidence for this is the statement that each spoke is attached to the central plate via a \"closet hinge\" to ensure equal normal force. No per-wheel normal force is ever measured. Two measured residuals are the expected signatures of unequal normal forces: (1) the near-frictionless state in Fig. 6 reduces friction by 91±4%, not 100%; (2) the median fit in Fig. 3b has slope 1.065±0.01, not 1. If wheels carry unequal loads, the equilibrium is a weighted median and the force-speed curve is a weighted sum, exactly as the authors acknowledge in the Discussion (\"By treating fi as weights on the probability distribution... we can still solve for equilibrium speed as the weighted median\"). This would not destroy the concept of active-contact friction control, but it would make the headline quantitative claims (unweighted median, Eq. 2's CDF) approximate rather than exact, and it would bias the claimed friction-coefficient values. Because the paper's central demonstration of \"near frictionless sliding\" rests on exact cancellation of equal-magnitude contact forces, this assumption is load-bearing and should be tested directly.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper studies sliding friction of a carousel with ten independently motor-driven wheels in frictional contact with an acrylic ground. The authors model each contact as a speed-independent Coulomb friction element and show that the net force-speed curve obeys Eq. (1), so the equilibrium system speed equals the median of the wheel speeds. Experiments on 660 random speed configurations support the median prediction, force-speed measurements reproduce step-like and programmed linear curves, and measurements of startup time and transients demonstrate controllable effective viscosity and friction coefficient. The paper claims that opposing contact pairs can create near-frictionless sliding.","tokens_in":10303,"tokens_out":4174,"duration_ms":44910,"significance":"If correct, the work establishes that the statistical distribution of active-contact speeds, rather than material properties alone, can shape macroscopic sliding friction, with implications for legged locomotion and engineered surfaces. The model is simple and falsifiable, and the experiments are extensive: 660 equilibrium trials, 490 viscosity trials, and independent mean-vs-median tests. The main quantitative claims rest on the equal-normal-force assumption, which is plausible from the hinge design but not directly measured; this is the central weakness. The paper's proposed CDF formulation in Eq. (2) is a restatement of Eq. (1), but it may be useful for continuous contact distributions.","major_comments":[{"comment":"The equal-normal-force assumption is load-bearing but unverified. Eq. (1) and the median prediction assume each contact contributes the same friction magnitude µmg/N; the only evidence is the hinge design statement. The fitted slope m = 1.065 ± 0.01 in Fig. 3b and the 91 ± 4% reduction in Fig. 6 are both consistent with unequal normal forces, which the authors themselves note would replace the unweighted median with a weighted median. Please either measure per-contact normal forces directly, add a calibration showing equal loads, or explicitly reframe the quantitative claims as approximate.","section":"Equilibrium properties of multi-contact friction, Fig. 3b"},{"comment":"The agreement with the programmed force-speed curves is a central demonstration, but the plotted data appear to be single trials without error bars or repeated runs. Without a measure of trial-to-trial variability, the reader cannot judge whether the deviations at the step transitions (attributed to backlash) are representative or whether the claimed linear viscosity curves are within noise. Please add statistics over repeated force-speed measurements or clearly label representative traces.","section":"Measurement of force-speed curves, Fig. 4"},{"comment":"The abstract and the final experimental section describe the result as near-frictionless, but the measured reduction is 91 ± 4%, leaving a residual friction force of about 9% of the braked value. This is a good result but not 'near frictionless'; the wording should be qualified, and the residual mechanism should be discussed.","section":"Abstract and Sliding friction coefficient control, Fig. 6"}],"minor_comments":[{"comment":"There are several typographical errors: 'counter-exmaples', 'equilirium', 'whe', 'probabiltiy', 'subsitution', 'combing', and 'Communincations' in Ref. [6].","section":"Throughout"},{"comment":"The caption labels subfigures a), c), d) but the text refers to 'Fig. 2a,b'; subfigure b appears to be missing or mislabeled.","section":"Fig. 2"},{"comment":"The text refers to 'Fig. 1b' and 'Fig. 1c' when discussing the median and mean experiments; these cross-references should be to Fig. 3.","section":"Equilibrium properties section"},{"comment":"The CDF definition should specify the integration variable explicitly; the current expression CDF(V) = ∫ P(V) is ambiguous because the dummy variable equals the argument.","section":"Eq. (2)"},{"comment":"The text says the commanded force-speed profile is shown in red in Fig. 4a, while the caption says the programmed curves are black lines; please reconcile these color descriptions.","section":"Fig. 4 caption"}],"recommendation":"major_revision","confidential_remarks":"This is a solid experimental study; the equal-normal-force assumption is the main weakness, and I believe it can be addressed by direct normal-force measurements or by softening the quantitative claims. The paper fits the journal's scope and is not fatally flawed. I would not require new theory, but the force-speed curves need error bars and the 'near frictionless' wording should be made more precise."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The thing to know: this paper builds a ten-wheel carousel with independently speed-controlled contacts and shows, convincingly, that the net friction force is the sum of individual Amonton-Coulomb terms, producing programmable force-speed curves—including near-zero friction. The data are plentiful (660 equilibrium trials, 490 viscosity trials) and the simple model matches them well. The median equilibrium result is not brand new, and the authors credit prior work for it; what is new is the direct force-speed measurement and the real-time friction coefficient control, which are neatly done.\n\nThe soft spots are real but not disabling. The biggest is the equal-normal-force assumption. Each spoke hangs from a closet hinge, but no per-wheel normal force is measured. The two residuals you'd expect from load imbalance are present: the near-frictionless state reduces force by 91±4%, not 100%, and the median fit slope is 1.065±0.01, not 1. The authors themselves note in the Discussion that with unequal forces you get a weighted median and a weighted CDF, so they are aware of the generalization. Still, the headline quantitative claims are stated in unweighted terms, and the paper should either measure per-wheel loads or explicitly characterize how much imbalance the hinge allows. The qualitative message—active contacts can control friction—survives either way.\n\nTwo smaller issues. Figure 4 shows force-speed curves without error bars or repeated trials; given that this is the core measurement, a few repeats would help. And \"near frictionless sliding\" overstates a 91% reduction; it is a large reduction, but not near frictionless, and the phrasing invites hype.\n\nI disagree with the stress-test's implicit worry that the concept collapses under load imbalance. It doesn't; the model simply becomes weighted. The paper is honest about the equations, uses the right citations, and the experiments are substantial. The main deficiencies are presentation and a missing control, not a fatal flaw.\n\nWho is this for: robophysics, biomechanics, and tribology readers interested in active friction control. It deserves a serious referee, with requests for repeated force-curve measurements and a direct or indirect test of the equal-normal-force assumption. I would engage with it and probably cite it if I worked on active friction or multi-legged locomotion.","headline":"A clean experimental demonstration that active contacts can shape the force-speed curve of dry sliding friction; the equal-normal-force assumption is the main soft spot, but it does not break the central claim.","tokens_in":10786,"tokens_out":1378,"would_cite":true,"duration_ms":17742,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Multiple active contacts with speed-independent friction can together produce a controllable, speed-dependent sliding friction law.","keywords":["active contacts","Amonton-Coulomb friction","force-speed curve","median speed","friction control","effective viscosity","sliding locomotion","dry friction"],"falsifier":"Place an individual load cell under each wheel mount (or otherwise measure per-contact normal force) while the carousel slides through its commanded speed range; if the per-wheel loads differ measurably, the equilibrium speed should follow a weighted median of the contact speeds and the measured force-speed staircase should deviate from Eq. (1) in the direction of the weighted sum. The paper reports no such per-contact load measurement.","tokens_in":9865,"feed_emoji":"⚙️","tokens_out":8566,"duration_ms":83967,"temperature":0.7,"pith_summary":"Sliding friction between dry surfaces is usually taken as a speed-independent force set by the materials and normal load. This paper argues that when a surface has many 'active' contacts—points that move at their own controllable speeds, like wheels or robot feet—the collective friction becomes speed-dependent even though every individual contact obeys the speed-independent Amonton-Coulomb law. The net force-speed curve is then shaped by the distribution of contact speeds: it is a staircase that passes through zero at the median contact speed, so the system's equilibrium sliding speed is that median. The authors verify this with a ten-wheel rotating carousel and show that programming the contact speeds can set an effective viscosity, adjust the sliding friction coefficient in real time, and even produce near-frictionless sliding. If the result holds, contact-speed distribution, not just material and load, becomes a design handle for dry friction.","feed_headline":"Wheel speeds set dry sliding friction: median is the key","feed_subtitle":"Ten spinning wheels turn speed-independent friction into a programmable force-speed curve, even near-frictionless.","key_machinery":"The load-bearing object is the sign-sum identity $F(V)=\\frac{1}{N}\\mu mg\\sum_i\\frac{v_i-V}{|v_i-V|}$, which converts each contact's speed-independent Amonton-Coulomb force $\\mu mg/N$ into a $\\pm$ contribution whose sign flips when the system speed $V$ crosses the contact's commanded speed $v_i$. Ordering the $v_i$ makes the net force a staircase between $+\\mu mg$ and $-\\mu mg$, crossing zero at the median speed; with an even number of contacts the two middle speeds produce a flat zero-force plateau. The equivalent cumulative-distribution form $F(V)=\\mu mg(1-2\\,\\mathrm{CDF}(V))$ turns the distribution of contact speeds into the design parameter. Experimentally, the machinery is a ten-wheel 'carousel' with independently speed-controlled motors, hinge-mounted spokes intended to equalize normal forces, and encoders on every wheel and on the central rotation.","core_discovery":"The central claim is that the force-speed law of a multi-contact sliding system is a property of the set of contact speeds, not of the contacts' individual friction laws. For $N$ contacts with equal friction magnitude $\\mu mg/N$ and commanded speeds $v_i$, the net friction is $F(V)=\\frac{1}{N}\\mu mg\\sum_i \\frac{v_i-V}{|v_i-V|}$; each term is $\\pm 1$, so the curve is a descending staircase with a drop at each $v_i$. The equilibrium sliding speed, where $F=0$, is the median of the $v_i$, independent of the friction coefficient and of the mean contact speed, and robust to outliers. The paper verifies this with a ten-wheel carousel whose wheels have independent motor speeds: 660 random-speed trials confirm median equilibrium, direct ground-reaction measurements match the commanded staircases, uniformly spaced speeds create linear force-speed curves with controllable effective viscosity, and opposing equal-speed contact pairs reduce the measured sliding friction coefficient by $91\\pm4\\%$ and can make the system nearly frictionless. The discrete law is recast as $F(V)=\\mu mg\\,(1-2\\,\\mathrm{CDF}(V))$, which the paper proposes as an analytic route to force-speed curves for continuous active-contact systems such as spinning disks and slithering snakes.","pith_inferences":["If the cumulative-distribution rule holds generally, a surface with many small counter-rotating rollers could present a programmable friction landscape: adjusting the roller-speed distribution would shape the in-plane force-speed curve, including zero net sliding friction without lifting the load; the paper does not test this engineered-surface application.","The median's robustness to outliers implies that in legged locomotion with many slipping contacts, an outlier limb speed would not shift steady locomotion speed, so a malfunctioning or freely spinning leg might be masked as long as it stays on one side of the median; this is a testable extension to robots.","For even numbers of contacts, the flat zero-force plateau creates a speed-selective mechanical response: nearly frictionless inside the band between the two middle speeds and strongly braking outside it, which could serve as a speed-selective brake or clutch.","Extending the authors' weighted-median remark, deliberately imposing unequal normal loads or per-contact friction coefficients would turn the force-speed curve into a weighted sum and the equilibrium into a weighted median, providing a direct experimental bridge to biological and granular contacts where loads are inherently uneven."],"forward_implications":["The steady sliding speed of an equal-force active-contact system is the median of the contact speeds, so uniform changes in the friction coefficient or in the mean contact speed leave the equilibrium speed unchanged.","The slope of a programmed force-speed curve behaves like a viscosity: the time constant for approaching steady state scales inversely with the slope, matching exponential-relaxation predictions across 490 trials.","Pairing contacts at opposite speeds effectively deactivates them within the speed band, so the sliding friction coefficient can be lowered in real time; at maximum cancellation the measured reduction is $91\\pm4\\%$ and the sliding is nearly frictionless.","Through the cumulative-distribution formulation, the discrete result gives an analytic force-speed curve for any continuous distribution of contact speeds, which the paper proposes for spinning disks, slithering snakes, and slow granular drag.","Because the force-speed curve is controlled by the speed distribution alone, a single material pair can implement a family of friction laws—step, viscous, near-zero—by reprogramming contact speeds."],"supporting_citations":[{"why":"introduces the active-contact framework for multilegged locomotion that motivates the experiment","marker":"[14]"},{"why":"provides the walking-robot observation of speed-dependent traction from speed-independent contacts and the prior time-averaged viscous-force result that this paper generalizes","marker":"[15]"},{"why":"gives the median-speed argument for equilibrium in slipping multi-legged locomotion, which the carousel experiments test directly","marker":"[18]"},{"why":"is the earlier disk sliding-and-spinning demonstration of non-uniform contact speeds that the CDF formulation would also describe","marker":"[13]"},{"why":"slithering snake locomotion is cited as a continuous active-contact system to which the force-speed CDF curve could apply","marker":"[20]"},{"why":"defines the coasting-number context used to predict when multi-contact gaits will exhibit instantaneous force-speed behavior","marker":"[30]"},{"why":"shows slow drag in granular media is speed-independent, supporting the proposed extension of active-contact physics to granular locomotion","marker":"[31]"}],"fun_headline_variants":["Median wheel speed sets friction in active contacts","Active contacts turn speed curves into tunable friction","Ten wheels show friction can be tuned by speed control","Sliding friction shaped by contact speeds, not just materials","Programmable friction from independent contact speeds"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that all ten contacts press on the ground with equal normal force, so each contributes the same friction magnitude $\\mu mg/N$; the hinge suspension is meant to enforce this, but per-contact normal forces are never measured.","fun_headline_variants_meta":{"raw":{"variants":["Median wheel speed sets friction in active contacts","Active contacts turn speed curves into tunable friction","Ten wheels show friction can be tuned by speed control","Sliding friction shaped by contact speeds, not just materials","Programmable friction from independent contact speeds"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000188,"raw_usage":{"total_tokens":1379,"prompt_tokens":1038,"completion_tokens":341,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":654,"completion_tokens_details":{"reasoning_tokens":270}},"tokens_in":654,"tokens_out":341,"duration_ms":4470,"temperature":1.0,"reasoning_tokens":270,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T19:41:11.991432+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Place an individual load cell under each wheel mount (or otherwise measure per-contact normal force) while the carousel slides through its commanded speed range; if the per-wheel loads differ measurably, the equilibrium speed should follow a weighted median of the contact speeds and the measured force-speed staircase should deviate from Eq. (1) in the direction of the weighted sum. The paper reports no such per-contact load measurement.","supporting_citations":[{"cited_title":"Chong, J","cited_arxiv_id":null,"evidence_quote":"introduces the active-contact framework for multilegged locomotion that motivates the experiment"},{"cited_title":"Chong, J","cited_arxiv_id":null,"evidence_quote":"provides the walking-robot observation of speed-dependent traction from speed-independent contacts and the prior time-averaged viscous-force result that this paper generalizes"},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"gives the median-speed argument for equilibrium in slipping multi-legged locomotion, which the carousel experiments test directly"},{"cited_title":"Farkas, G","cited_arxiv_id":null,"evidence_quote":"is the earlier disk sliding-and-spinning demonstration of non-uniform contact speeds that the CDF formulation would also describe"},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"slithering snake locomotion is cited as a continuous active-contact system to which the force-speed CDF curve could apply"},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"defines the coasting-number context used to predict when multi-contact gaits will exhibit instantaneous force-speed behavior"},{"cited_title":"Albert, M","cited_arxiv_id":null,"evidence_quote":"shows slow drag in granular media is speed-independent, supporting the proposed extension of active-contact physics to granular locomotion"}],"review_version":1}