{"id":"0c5c4f8b-e6b8-48b7-8bf4-58f2b8b6b260","arxiv_id":"2506.15788","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Mapping leg-ground contacts to communication bits lets the authors design multi-legged robots that traverse noisy terrain with passive and active error correction, approaching a speed limit of about four leg-lengths per cycle.","lead":"This paper treats each leg-ground contact of a multi-legged robot like a 'bit' in a data transmission, using leg redundancy and simple feedback to correct terrain-induced errors. With that framing, the authors show that elongate robots with many legs can traverse rough terrain at speeds approaching a leg-length-dependent limit.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 'approaching the upper bound' claim compares 0.40 BL/cycle to a stale 4l bound: footnote 4 admits the sensor-equipped MIMO robot has a longer effective leg length, which raises the true bound and is never used in the comparison.","rationale":"The paper's central contribution is a communication-theory-inspired framework plus two concrete quantitative claims: a speed upper bound 4l that is independent of leg count, and demonstrations that CI (MIMO) or morphology (C-legs) brings measured speed close to that bound. The upper bound itself is supported by a simple kinematic argument and by the geometric-mechanics saturation data, so I do not see a fatal flaw there. The most load-bearing weak point is the MIMO comparison to the bound. The paper explicitly flags, in footnote 4, that the sensor-equipped robot's effective leg length is larger than the l=8.4cm used to compute 4l=33.6cm/cycle≈0.43 BL/cycle, which means the reference bound used in the comparison is too low. This is an internal inconsistency: the footnote raises the bound but the headline number does not use the raised bound. The severity depends on the actual sensor height, which is not reported in the main text. The reader's concern about correlated bac losses is plausible and worth testing, but it applies to the qualitative MI/CI analogy rather than to the quantitative speed-bound claim, and the paper's robustness results are experimentally demonstrated rather than derived from an independence assumption. Thus the more immediate, checkable issue is the stale bound in the MIMO comparison. Depending on l_eff, this could turn 'approaching the upper bound' into 'reaching ~80% of the upper bound,' which would require toning down the claim. This supports the CONDITIONAL verdict; no change in verdict category is needed, but the revision should address this calculation.","tokens_in":16186,"tokens_out":16294,"duration_ms":169855,"concrete_test":"Report the physical leg length and the added length of the contact sensors on the MIMO robot; compute the corrected bound 4l_eff/BL and the ratio (0.40±0.03)/(4l_eff/BL). If the ratio is materially below the implied 93% (e.g., <0.85), revise the 'approaching the upper bound' claim or state the corrected bound in Fig. 7.A.2; if the ratio stays ≥0.90, the concern is minor. The same correction should be applied wherever the sensor-equipped robot is compared to 4l.","verdict_should_be":"UNCHANGED","load_bearing_attack":"In the MIMO CI section (Fig. 7.A.2), the paper states that the sensor-equipped 12-legged robot reaches 0.40±0.03 BL/cycle, 'approaching the theoretical upper bound of 0.43 BL/cycle predicted in previous sections.' The immediately following footnote 4 says: 'The robot used in this section is equipped with contact sensors at the tips of each foot, effectively increasing leg length and thereby raising the upper bound on achievable speed.' Since the predicted upper bound is 4l (Section 'Emergent upper bound...', l=8.4cm ⇒ 4l=33.6cm/cycle ≈0.43 BL/cycle), an increase in effective leg length l_eff necessarily increases the bound to 4l_eff/BL. The comparison therefore uses the wrong reference value: the measured 0.40 should be judged against a larger denominator. For example, a 1 cm increase in l (8.4→9.4cm) raises the bound to ≈0.48 BL/cycle, making the achieved speed 83% of the bound rather than 93%. This does not destroy the upper-bound result, but it weakens the paper's headline claim that MIMO CI approaches the theoretical speed limit, and it is an internal inconsistency: the footnote acknowledges the issue but the analysis does not correct for it.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a communication-theoretic framework for multi-legged elongate robots (MERs), in which each leg-ground contact is a 'bac' analogous to a bit, passive gravity-driven force redistribution is mechanical intelligence (MI) analogous to a majority-vote decoder, and active feedback is computational intelligence (CI) analogous to ARQ. Using geometric mechanics, the authors derive an emergent upper bound on absolute speed of approximately 4l (about 33 cm/cycle for l = 8.4 cm), predict that speed saturates near N = 7 leg pairs, and validate these predictions numerically and experimentally on flat and rugose terrains. They then introduce vertical-wave contact modulation, a SISO adaptive controller, an RL-trained MIMO controller, and C-shaped legs, reporting speeds that approach the predicted bound on noisy terrain. The communication-theory analogy is used throughout to justify redundancy-based reliability and speed-robustness trade-offs.","tokens_in":16496,"tokens_out":7487,"duration_ms":76267,"significance":"If the results hold, the paper provides a rare predictive design principle for legged locomotion: a simple kinematic speed bound 4l that is independent of leg count, together with a quantitative account of when adding legs ceases to improve speed. The geometric-mechanics predictions match the flat-ground experiments, the N≈7 saturation is tested both in simulation and in hardware, and the C-leg experiments demonstrate terrain-agnostic open-loop performance. These are concrete strengths. The communication analogy is stimulating but not yet fully load-bearing: the reliability guarantees rest on an untested independence assumption for bac errors, several key proofs are deferred to the SI, and the MIMO comparison to the upper bound uses an inconsistent reference leg length. The paper also does not state data/code availability, which limits reproducibility of the simulation and RL components. With targeted revisions, this could be a strong contribution to the field.","major_comments":[{"comment":"The claim that MIMO CI at 0.40±0.03 BL/cycle 'approaches the theoretical upper bound of 0.43 BL/cycle' is internally inconsistent with footnote 4, which states that the contact sensors effectively increase leg length and thereby raise the upper bound. Since the predicted bound is 4l, the comparison must use the effective leg length of the sensor-equipped robot. As a concrete illustration, increasing l from 8.4 cm to 9.4 cm changes the bound to approximately 0.48 BL/cycle, making the achieved speed about 83% of the bound rather than 93%. Please recompute the comparison with the actual effective leg length or rephrase the claim so that it does not overstate proximity to the bound.","section":"Approach the upper bound of speed on noisy landscapes; footnote 4"},{"comment":"The main text twice asserts that a constant thrust profile minimizes the coefficient of variation of cycle-averaged velocity and that a more uniform thrust distribution lowers Cv, with proofs deferred to the SI. The SI is not supplied with the manuscript as provided, so these load-bearing claims are not verifiable from the submission. Please include the proofs or state the exact modeling assumptions under which they hold, and confirm that the vertical-wave robustness results do not depend on an unstated circularity between the thrust-uniformity measure and the performance metric.","section":"Temporal and spatial synchronization for simple repetition"},{"comment":"The reliability framework assumes that terrain-induced bac losses behave like independent bit errors in a communication channel. The experiments use random-height step fields, but the paper does not measure or model correlated bac losses, such as a single obstacle lifting several legs at once or a body tilt causing systematic contact loss. Under correlated noise, the majority-vote and redundancy guarantees need not hold. Please either add experiments or analysis with controlled correlation lengths of terrain failures, or explicitly limit the reliability claims to the tested class of terrains.","section":"Gravity as an MI control scheme; Table 1"},{"comment":"The C-leg result is compared to '4×leg length, 0.44 BL/cycle,' but the effective leg length of the C-shaped leg is not defined. If the C-leg's effective length differs from the point-leg value l = 8.4 cm used for the earlier bound, the reference bound changes and the 'approaching the upper bound' claim is weakened. Please specify the leg length used for the C-leg comparison and recompute the bound accordingly, or report the ratio of achieved speed to the appropriate 4l value for the C-leg geometry.","section":"Embedding MI in bac through foot morphology design"}],"minor_comments":[{"comment":"The text contains the typo 'self-collusion' where 'self-collision' is meant; please correct it.","section":"Emergent upper bound of absolute speed with additional legs"},{"comment":"The phrase 'MIMI CI' appears in the text and should be 'MIMO CI'.","section":"Multi-input-multi-output CI via reinforcement learning"},{"comment":"The table entry 'Automatic-Repeat Qequest' contains a typo; it should read 'Automatic Repeat Request'.","section":"Table 1"},{"comment":"The figure labels the non-MIMO controller 'SIMO CI,' while the text refers to it as 'SISO CI'; please unify the terminology.","section":"Figure 7.A.2"},{"comment":"The SISO controller gain p = 3.2 is chosen from empirical experiments; a sensitivity analysis or cross-validation would help establish that the reported improvement is not specific to this particular value.","section":"Single-input-single-output CI"}],"recommendation":"major_revision","confidential_remarks":"The manuscript relies heavily on the authors' prior work (refs 18-20, 47) and a companion RL paper (47) for the MIMO training details; the novelty relative to these works should be stated more crisply. The 'demonstration' section features a commercial platform from Ground Control Robotics, a company with which the authors have funding ties; a competing-interests statement is advisable. The upper-bound result itself appears sound, but the MIMO 'approaching the bound' comparison and the missing SI proofs are the main correctness risks to resolve before publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This paper deserves a serious referee, but you should know two things going in. First, the core result — a parameter-free kinematic upper bound near 4l on absolute speed, with saturation around 7 leg pairs — is genuinely nice and is backed by both simulation and physical experiments. That part holds up. Second, the headline claim that the MIMO controller approaches this bound has a small internal inconsistency: footnote 4 admits the sensor-equipped robot has an effectively longer leg length, which raises the theoretical bound from 0.43 BL/cycle to something larger. They still compare against the stale 0.43 value, so the 0.40±0.03 result is less impressive than it looks. It doesn't destroy the paper, but it needs correcting.\n\nWhat's actually new: the geometric-mechanics encoding of body undulation that predicts the speed saturation, the adaptive vertical-wave SISO controller, the RL-based MIMO controller, and the C-legged morphology demonstration. The 4l derivation is clean and the experimental match on flat ground is convincing. The communication-theory analogy is built on the authors' prior work, but the GM framework and the controllers are real extensions.\n\nThe soft spots, in order of severity. The proof about uniform thrust minimizing the coefficient of variation is deferred to an SI that wasn't available; that's a referee chase. The SISO gain p=3.2 is fitted to experiments, which is fine for a single scalar but means that part is less predictive. The MIMO policy is trained in simulation, with details in a separate paper, so the CI sections are not self-contained. And the label \"source entropy\" for a kinematic speed bound is rhetorical — that's not what source entropy means in information theory. It's a cool analogy, but it shouldn't be sold as more than that.\n\nOne assumption worth probing: treating terrain-induced bac losses as bit-like errors presumes the noise is fairly independent across legs. A single step block could lift multiple legs at once, and correlated failures would weaken the redundancy argument. The experiments on random stepfields show robustness anyway, so this is a limitation of the theory rather than a fatal flaw, but a referee should ask how far the independence assumption can be pushed.\n\nBottom line: the paper is a solid contribution to legged locomotion on rough terrain, with a reproducible geometric result and honest experiments. It should go to peer review rather than be desk-rejected. The referee should ask for the missing SI, corrected MIMO comparison, and a more careful positioning of the analogy. I'd be willing to cite the 4l bound in my own work. For a reading group, it would spark a good discussion about when communication-theory metaphors are useful in robotics.","headline":"A serious, experimentally grounded robotics paper with a clean geometric speed bound; the MIMO 'approaching the bound' claim is slightly overstated due to a footnote they acknowledge but don't correct.","tokens_in":17048,"tokens_out":2724,"would_cite":true,"duration_ms":29404,"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":"Multi-legged elongate robots have an emergent stride ceiling near four times leg length—about 33 cm per cycle—no matter how many legs are added.","keywords":["multi-legged robots","elongate robots","terrain robustness","mechanical intelligence","geometric mechanics","communication theory analogy","basic active contact","speed upper bound"],"falsifier":"A robot with more legs than the saturation point that achieves a flat-ground stride strictly greater than four times its leg length (4l) would falsify the claimed bound. Separately, a rugose terrain whose height variations are strongly correlated—one obstacle lifting several legs at once—on which open-loop redundancy loses its predicted reliability would falsify the independent-bit noise model.","tokens_in":15971,"feed_emoji":"🦿","tokens_out":8752,"duration_ms":85217,"temperature":0.7,"pith_summary":"The paper claims that multi-legged elongate robots (MERs), built from serially connected bipedal modules, have an emergent upper bound on stride length of roughly four times leg length, about 33 cm per cycle for the tested robots, no matter how many legs are added. It argues this ceiling follows from treating each leg-ground contact as a bit-like 'basic active contact' and applying communication-theory ideas: redundancy gives open-loop robustness on rough terrain, while feedback control plays the role of retransmission. Using geometric mechanics to design body-undulation gaits, the authors show that adding legs only helps up to about seven leg pairs, after which speed saturates. They then demonstrate that a learned multi-input feedback controller reaches 0.40 ± 0.03 body lengths per cycle on noisy terrain, close to the predicted 0.43, and that C-shaped legs reach comparable speeds with no feedback at all. The wider value is a design rule: speed, robustness, and sensor/computation load can be traded off systematically rather than tuned robot-by-robot.","feed_headline":"Robot speed caps at 33 cm per cycle no matter the leg count","feed_subtitle":"The cap is four times leg length; feedback control and C-shaped legs both reach it on rough terrain.","key_machinery":"The load-bearing object is the 'bac'—basic active contact, one leg-ground contact treated as a bit in a communication channel—together with the analogy that maps mechanical intelligence (passive gravity redistribution of ground reaction forces) to a majority-vote decoder, simple repetition/spatial synchronization to repetition coding, and feedback control (computational intelligence) to automatic repeat request. Gait design is carried by geometric mechanics: the body shape space with local connection matrix A(w), whose curl (the forward height function) gives the net stride from a closed gait path via Stokes' theorem. A vertical body wave modulates contact timing to spread thrust uniformly in time, which sets the 'majority vote window size' and is the tuning knob for the speed-robustness trade-off. The claimed 4l upper bound is derived from two elementary models—non-slip kinematics and steady slipping with Coulomb friction—so the bound is a geometric consequence of leg length rather than of the friction model.","core_discovery":"For an elongate robot whose legs all do the same synchronized thrust, additional legs add reliability but not speed; the paper's central discovery is that this speed ceiling is not an engineering accident but a geometric bound. With a lateral undulation wave and one leg pair per module, geometric mechanics predicts that stride length is maximized at a body amplitude set by self-collision constraints, that the optimal amplitude falls as leg count rises, and that the resulting maximum stride converges to 4l (four times leg length, 33.6 cm for l = 8.4 cm). The saturation occurs near N = 7 leg pairs for the tested geometry. The same 4l limit is recovered from a non-slip kinematic model and from a steady-state slipping model with isotropic Coulomb friction, so it does not depend on the no-slip assumption. The authors then show the limit is approachable on rough terrain: a reinforcement-learned MIMO controller, which adapts body undulation, leg amplitude, and vertical wave in response to measured bac loss, reaches 0.40 ± 0.03 BL/cycle against the predicted 0.43, and point-foot robots with C-shaped legs reach comparable speed in open loop because the distributed foot contact makes bac loss rare.","pith_inferences":["Beyond the paper: terrain noise with correlated bac losses (one obstacle lifting several legs at once) should break the independent-bit redundancy guarantee; a terrain characterized by bac-loss correlation length would test this.","Beyond the paper: the 4l result bounds stride per cycle, not velocity; the natural next bound to seek is on cycle frequency, since absolute speed is stride times frequency.","Beyond the paper: the stated equivalence between computational and design complexity implies an iso-performance surface; measuring wall-clock or energy cost of a many-leg open-loop robot versus a few-leg feedback robot on identical terrain would quantify the exchange rate."],"forward_implications":["Adding legs beyond roughly seven pairs does not increase flat-ground stride; robots should be designed with fewer legs plus body undulation or feedback.","A simple single-input single-output controller can give a 12-legged robot reliability comparable to a 16-legged open-loop robot, showing that sensor complexity can substitute for leg count.","C-shaped legs let an MER traverse flat and rugose terrains with statistically indistinguishable open-loop speeds, approaching the four-times-leg-length bound.","The speed-robustness trade-off is controlled by encoding choices: less spatial redundancy or a smaller vertical wave raises smooth-terrain speed but costs robustness."],"supporting_citations":[{"why":"It establishes the locomotion-communication analogy and shows that leg redundancy guarantees open-loop reliability, which this paper extends to speed bounds and feedback.","marker":"(20)"},{"why":"It supplies the noisy-channel coding framework, including source entropy and majority-vote decoding, onto which the bac-bit model is mapped.","marker":"(37)"},{"why":"It defines the multi-legged elongate robot architecture and the lateral body-undulation wave prescribed by shape variables.","marker":"(18)"},{"why":"It gives the quasi-static ground-reaction-force model and coasting-number analysis used to compute body velocity and the four-times-leg-length bound.","marker":"(19)"},{"why":"It introduces the height-function formulation of geometric mechanics used to design gaits from closed paths in shape space.","marker":"(44)"},{"why":"It defines the coasting number and establishes the thrust-dominated regime, justifying the quasi-static force balance.","marker":"(39)"},{"why":"It documents the C-shaped leg whose distributed foot contact underlies the open-loop, terrain-agnostic mechanical-intelligence result.","marker":"(24)"},{"why":"It provides the no-free-lunch argument used to justify that mechanical intelligence alone cannot simultaneously maximize speed and robustness.","marker":"(45)"},{"why":"It describes the Stop-and-Wait ARQ protocol that serves as the template for bac-level catch-and-correct feedback.","marker":"(46)"}],"fun_headline_variants":["Speed ceiling for legged robots: 4 leg lengths per cycle","More legs add reliability, not speed, for elongate robots","Geometric bound caps robot speed at 4 leg lengths per stride","Leg count irrelevant: robot speed limited by leg length","No speed gain from extra legs — only robustness"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that terrain-induced failures of individual leg-ground contacts behave like independent random bit errors, so that redundancy and coding arguments (majority vote, forward error correction, retransmission) apply quantitatively to locomotion.","fun_headline_variants_meta":{"raw":{"variants":["Speed ceiling for legged robots: 4 leg lengths per cycle","More legs add reliability, not speed, for elongate robots","Geometric bound caps robot speed at 4 leg lengths per stride","Leg count irrelevant: robot speed limited by leg length","No speed gain from extra legs — only robustness"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000294,"raw_usage":{"total_tokens":1783,"prompt_tokens":1092,"completion_tokens":691,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":708,"completion_tokens_details":{"reasoning_tokens":609}},"tokens_in":708,"tokens_out":691,"duration_ms":6840,"temperature":1.0,"reasoning_tokens":609,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T19:32:01.002187+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A robot with more legs than the saturation point that achieves a flat-ground stride strictly greater than four times its leg length (4l) would falsify the claimed bound. Separately, a rugose terrain whose height variations are strongly correlated—one obstacle lifting several legs at once—on which open-loop redundancy loses its predicted reliability would falsify the independent-bit noise model.","supporting_citations":[],"review_version":2}