Pith. sign in

REVIEW 4 major objections 4 minor 1 cited by

Validation of Tumbling Robot Dynamics with Posture Manipulation for Closed-Loop Heading Angle Control

T0 review · 4 major / 4 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper claims that COBRA's tumbling descent can be modeled as a thin elliptical ring whose posture-controlled shape changes steer the heading, and that this reduced-order cascade model reproduces the key behaviors of a high-fidelity…

desk verdict The validation can't be assessed as written because the model's core equations are missing and the printed posture constraint is wrong; the paper needs a major revision before peer review. read the letter →

arxiv 2411.12970 v1 pith:GHMLOQAL submitted 2024-11-20 cs.RO cs.SYeess.SY

Adarsh Salagame , Eric Sihite , Alireza Ramezani This is my paper · ORCID
classification cs.ROcs.SYeess.SY
keywords tumblinglocomotionreduced-ordercascademodelposturemanipulationheadinganglecontrolsnakerobotellipticalringnonholonomicrollingconstraintsSimscapevalidation
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper claims that the tumbling of the snake-like robot COBRA can be captured by a reduced-order model: a thin elliptical ring whose principal-axis lengths $a$ and $b$ act as control inputs adjusted by the robot's joints. It validates this cascade model against a high-fidelity rigid-body simulation by feeding both models the same impulse-shaped posture input while the ring tumbles down a 15-degree slope. Over a four-second run, the heading angle, center-of-mass trajectory, and linear velocity agree well between the two models, including the central effect that a mid-tumble posture pulse deflects the heading. The author argues that this agreement makes the cascade model a valid design model for a closed-loop controller that steers the robot's descent on steep slopes, and that posture manipulation alone can replace dedicated steering actuators.

What carries the argument

The load-bearing object is the two-part cascade model $\Sigma_{tbl}/\Sigma_{pos}$. The posture subsystem $\Sigma_{pos}$ has state $\xi=(\xi_1,\dots,\xi_6)$ built from polar coordinates and the principal-axis lengths, and produces the mass moments of inertia $I_{kk}=\frac{m}{P}\int r_k^2\gamma(\xi)\,d\xi_4$ under a fixed-perimeter constraint; the tumbling subsystem $\Sigma_{tbl}$ evolves the roll-pitch-yaw orientation and contact point through $\dot{x}=M^{-1}(x,y)N(x,y)$ subject to the pure-rolling constraints $v_{c,x}=v_{c,y}=0$ and $v_{c,z}=0$. The validation harness is a Simscape ring assembled from 150 rigid elements linked to the center by prismatic actuators, with the mapping $r_i=ab/(b^2C_{\theta_i}^2+a^2S_{\theta_i}^2)^{1/2}$ converting the shape variables into element radii, and a spring-damper contact model with friction coefficient $\mu=5$ chosen to approximate the no-slip assumption.

What would settle it

Run the cascade model and the high-fidelity simulation with a posture input whose predicted ground reaction force goes negative (for example, a sharper or larger impulse than the tested $b'=0.2$ m, $\gamma=10$ pulse at the two-second mark); if the simulation then leaves the ground while the cascade model pins the ring to the surface, the heading and center-of-mass predictions will diverge and the general validity claim fails.

Watch

Extended reading notes

Core claim

The central claim is that the cascade model, which splits the system into a posture subsystem $\Sigma_{pos}$ that maps the shape inputs $u=[u_1,u_2]^\top$ to the ring's inertia tensor and a tumbling subsystem $\Sigma_{tbl}$ that evolves the ring's orientation and contact point under these time-varying inertias, closely approximates the high-fidelity Simscape simulation when driven by the same external input. The validation shows that both models deflect the center-of-mass trajectory to the same side at the moment of the impulse, keep the heading angle in close agreement, and exhibit the same linear-velocity response, with residual differences attributed to unmodeled slippage, an idealized mass distribution, and Simscape's approximate treatment of initial conditions. From this, the paper concludes that the reduced-order model accurately reflects key behaviors such as changes in center-of-mass trajectory, heading angle, and linear velocity, and that it is suitable for closed-loop trajectory tracking during tumbling.

Load-bearing premise

The model assumes the tumbling ring never slips and never leaves the ground: pure rolling with a non-negative normal force at the contact point is built into the equations, while the high-fidelity simulation allows real contact loss, so the validated agreement holds only for inputs that keep the ground reaction force positive.

Editorial extensions

If this is right

  • Posture manipulation alone is enough to steer the tumbling trajectory: changing the ring's principal axes mid-tumble deflects the heading without any dedicated steering actuator.
  • The cascade model can serve as the design model for a closed-loop heading-angle controller for COBRA's tumbling descent, since it reproduces the input-to-heading response of the high-fidelity simulation.
  • A controller developed on this validated model would first be tested on the reduced-order Simscape representation before being extended to full-scale control of COBRA's dynamics.
  • The results are reported for a single representative impulse signal, which the paper argues is representative because variations of it consistently produce heading and velocity changes in prior studies.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The validation only covers inputs that keep the predicted ground reaction force positive; a natural extension is to test whether the cascade model can still predict the dynamics during the ballistic phases that arise when the ring leaves the ground.
  • Because the model already parameterizes the slope angle $\alpha$, the same cascade structure should extend to variable slopes and uneven terrain, with the terrain profile entering the contact constraint.
  • The inertia-from-posture mapping suggests an onboard observer could estimate the ring's inertia tensor from joint-angle measurements, allowing the closed-loop controller to run without external motion capture.
  • The paper claims qualitative agreement but reports no quantitative error metric; computing heading-angle error or center-of-mass deviation between the two models would sharpen the suitability claim.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 4 minor

Summary. The paper proposes a reduced-order cascade model for the tumbling locomotion of the COBRA snake-like robot, separating posture dynamics (Σpos) from tumbling dynamics (Σtbl). The posture dynamics model an elliptical ring with virtual principal-axis actuators, and the tumbling dynamics are expressed as a first-order system ẋ = f(x,y) = M⁻¹(x,y)N(x,y), with M and N deferred to an appendix. The authors validate the cascade model against a high-fidelity Simscape multi-body simulation for a single impulse-like input on a 15-degree slope, comparing CoM trajectory, heading angle, and linear velocity qualitatively. They report good agreement and conclude that the model is suitable for closed-loop heading control during tumbling.

Significance. If the validation were quantitatively sound and the governing equations fully specified, the paper would provide a useful reduced-order design model for posture-controlled tumbling robots, which is a relatively unexplored locomotion mode. The Simscape setup with 150 discrete elements and mirrored prismatic actuators is a constructive high-fidelity reference. However, the present manuscript does not deliver a verifiable derivation of the tumbling equations, contains a questionable fixed-perimeter constraint, and reports only qualitative agreement. The central claim therefore rests on incomplete and partially erroneous foundations. The work is relevant to the robotics community, but in its current form it cannot serve as a reliable basis for closed-loop controller design.

major comments (4)
  1. [Section III-B, Eqs. (5)-(7)] Equation (6) is not the time derivative of the perimeter expression in Eq. (5). Differentiating P = ∫₀^{2π} sqrt(a²Cθ² + b²Sθ²) dθ with respect to time, using dan = u1 and db = u2, gives dP/dt = ∫₀^{2π} (a u1 Cθ² + b u2 Sθ²)/sqrt(a²Cθ² + b²Sθ²) dθ = 0. The equation printed in Eq. (6), namely dP/dt = sqrt(a²Cθ² + b²Sθ²) θ̇ = 0, does not follow from Eq. (5). As a consequence, the fourth row of Eq. (7) forces γ(ξ) dξ4/dt = 0, i.e., θ̇ = 0, which eliminates the θ-dynamics from the posture model. This is a load-bearing error because the subsequent derivation of Σpos and the cascade structure depends on a dynamical coupling between ȧ, ḃ and θ. The authors must correct the fixed-perimeter constraint and re-derive the posture state-space model, then re-run the validation.
  2. [Section III-C, Eq. (13)] The tumbling equations of motion are stated as ẋ = f(x,y) = M⁻¹(x,y)N(x,y), with the note that 'the nonlinear terms M(.) and N(.) are given in the Appendix Section.' In the arXiv version under review, there is no appendix. Without explicit expressions for M and N, the governing dynamics of the cascade model are not defined, and the numerical integration reported in Section V cannot be reproduced or checked. The full equations of motion must be included in the manuscript or in a freely accessible supplementary document.
  3. [Section V, Figs. 7-10] The validation is entirely qualitative. The conclusions state that the cascade model 'closely approximates' and 'strongly aligns' with the Simscape model, but no quantitative error metrics (e.g., RMSE, peak error, or percentage error) are reported for the heading angle, CoM trajectory, or linear velocity. The plots appear to show reasonable agreement, but the central claim of the paper is a quantitative modeling claim, and it must be supported by numbers. The authors should report error metrics over the 4-second horizon, and ideally for several input amplitudes and slope angles.
  4. [Section V and Section IV] The comparison is limited to a single input signal, and the authors state that the input was selected to keep ground reaction forces positive. The cascade model enforces continuous contact and pure rolling by construction, whereas the Simscape model allows ballistic motion when the normal force vanishes. This restricts the validity domain of the validated model, and the paper should explicitly state this limitation when claiming suitability for closed-loop control. Additional tests with inputs that approach the contact-loss boundary, or with different slopes, would be needed to substantiate the generality of the claim.
minor comments (4)
  1. [Section III-B, after Eq. (7)] The phrase 'it is possible to show' is vague; the derivation of the moments of inertia in Eq. (9) should either be shown explicitly or the reader should be referred to a specific equation or reference.
  2. [Section IV, contact parameters] The contact model parameters (k = 10⁴ N/m, b = 10³ N·s/m, μ = 5) are reported without justification or sensitivity analysis. Some discussion of how these values were chosen and how sensitive the results are to them would strengthen the validation.
  3. [Figure 6 caption] The caption says 'Due to the constraint of fixed perimeter according to equation 2,' but the fixed-perimeter condition is introduced in Eq. (5), not Eq. (2). The citation should be corrected.
  4. [Section VI] The conclusion says 'some deviations arise due to unmodeled effects like slippage,' but slippage was not measured or computed in the Simscape results. It would be more precise to say 'possible slippage' or to infer slip from the contact forces.

Circularity Check

0 steps flagged · score 2.0 of 10

No circular reduction: the cascade model is not fitted to the Simscape output and the validation is an independent model comparison; the main defects are a questionable posture constraint in Eq. 6 and the missing appendix referenced by Eq. 13, not circularity.

full rationale

The paper's central claim is that the cascade model closely approximates the Simscape simulation. The comparison is not circular by construction: the cascade model is integrated with ode45 from Eq. 13 under prescribed inputs, and the same inputs are fed to a Simscape multibody model built from 150 discrete rigid elements with contact dynamics; no parameter of the cascade model is tuned to match the Simscape trajectories. The self-citation to [26] is real but not load-bearing in the circularity sense: the posture and tumbling equations are restated in Sections III-B and III-C, and the validity claim is tested against an independently implemented simulator rather than asserted from the prior paper. The selection of inputs that keep ground reaction forces positive (Section V) restricts the operating domain but is not a fit of the inputs to the predicted heading. Two serious, non-circular defects should be flagged. First, Eq. 6 asserts the fixed-perimeter condition as ˙P = sqrt(a^2 C_theta^2 + b^2 S_theta^2) theta_dot = 0, but differentiating Eq. 5 with a_dot = u1 and b_dot = u2 gives an integral involving u1 and u2, so Eq. 6 is not the derivative of Eq. 5; this affects the posture state-space model in Eq. 7 and therefore the identity of the model being validated. Second, Eq. 13 says M and N are "given in the Appendix Section," but no appendix appears in the arXiv version, so the tumbling equations of motion cannot be independently checked. Neither defect makes the validation equivalent to its inputs; both are correctness and completeness risks. Since the only circularity-adjacent feature is a minor, non-load-bearing self-citation, the score is 2 rather than 0.

Assumptions & free parameters 3 free parameters · 4 assumptions · 1 invented entities

The central claim rests on idealizations (uniform elliptical ring, symmetric posture, pure rolling, continuous contact) that are shared by both the cascade model and the Simscape validation, plus hand-chosen simulation and input parameters. No free parameters are fitted to the cascade model's predicted outputs, which limits circularity, but the absence of an external benchmark means the ledger is mostly assumptions rather than measured evidence.

free parameters (3)
  • Input impulse parameters (gamma, t0, b0, b') = gamma=10, t0=2s, b0=0.3m, b'=0.2m
    Chosen as representative and to keep ground reaction force positive in Section V; no parameter sweep is performed, so generality of the validation is untested.
  • Simscape contact parameters (k, b_c, mu) = k=1e4 N/m, b_c=1e3 Ns/m, mu=5
    Hand-selected in Section IV to emulate no-slip contact in the high-fidelity model; no sensitivity analysis is provided.
  • Initial conditions and slope angle = phidot=2*pi rad/s, psidot=pi/6 rad/s, alpha=15 deg
    Scenario chosen in Section V to produce tumbling with nonzero roll rate; results for other conditions are not shown.
assumptions (4)
  • domain assumption The robot can be modeled as a thin elliptical ring with negligible cross-sectional area and a uniform mass distribution.
    Section III; this idealization determines the inertia tensor and dynamics; COBRA's discrete links and head/tail electronics violate it.
  • domain assumption Posture deformations are symmetric, keeping the ellipse center at the CoM.
    Section III, assumption (4); allows the inertia model to be a function of principal axes only; not validated against real joint configurations.
  • domain assumption Pure rolling with no slip at the contact point and continuous ground contact with non-negative normal force.
    Section III-C and V; the Simscape sim uses mu=5 to approximate no-slip, and the input is selected to avoid negative normal forces; ballistic phases are outside the model's domain.
  • ad hoc to paper The fixed-perimeter constraint can be expressed as gamma(xi) * dxi4/dt = 0 in Eq. 6, treating theta as a fixed coordinate.
    The derivation of Eq. 6 from the constant perimeter integral is not shown rigorously, and the resulting row in Eq. 7 freezes xi4; this step is load-bearing for the posture state-space model.
invented entities (1)
  • Virtual principal-axis actuators u1 and u2
    purpose: To model posture manipulation as direct control of the ellipse semi-axes a and b.
    Introduced as 'two imaginary actuators' in Fig. 3; no physical actuator on COBRA corresponds directly, and the mapping to joint commands is not established here.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Validation of Tumbling Robot Dynamics with Posture Manipulation for Closed-Loop Heading Angle Control." pith.science (2026). https://pith.science/paper/GHMLOQAL

@misc{pith2026241112970,
  author       = {Pith},
  title        = {Pith review of: Validation of Tumbling Robot Dynamics with Posture Manipulation for Closed-Loop Heading Angle Control},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GHMLOQAL}},
  note         = {Machine review of arXiv:2411.12970}
}
read the original abstract

Navigating rugged terrain and steep slopes is a challenge for mobile robots. Conventional legged and wheeled systems struggle with these environments due to limited traction and stability. Northeastern University's COBRA (Crater Observing Bio-inspired Rolling Articulator), a novel multi-modal snake-like robot, addresses these issues by combining traditional snake gaits for locomotion on flat and inclined surfaces with a tumbling mode for controlled descent on steep slopes. Through dynamic posture manipulation, COBRA can modulate its heading angle and velocity during tumbling. This paper presents a reduced-order cascade model for COBRA's tumbling locomotion and validates it against a high-fidelity rigid-body simulation, presenting simulation results that show that the model captures key system dynamics.

Figures

Figures reproduced from arXiv: 2411.12970 by the authors.

Figure 1
Figure 1. Northeastern University’s COBRA robot performing tumbling [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Shows the kinematic structure of COBRA in its snake configura [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 4
Figure 4. ROM simulation built using Simscape Multi-Body Toolbox [PITH_FULL_IMAGE:figures/full_fig_p004_4.png] view at source ↗
Figures from the paper (4 more)
Figure 6
Figure 6. Figure 6: Shows the input signal provided to the models. Due to the constraint [PITH_FULL_IMAGE:figures/full_fig_p004_6.png]
Figure 7
Figure 7. Figure 7: Shows the trajectory of the Center of Mass during tumbling for [PITH_FULL_IMAGE:figures/full_fig_p005_7.png]
Figure 9
Figure 9. Figure 9: Normal forces predicted by Cascade Model [PITH_FULL_IMAGE:figures/full_fig_p005_9.png]
Figure 10
Figure 10. Figure 10: Velocity of Center of Mass in the direction of tumbling [PITH_FULL_IMAGE:figures/full_fig_p006_10.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. NMPC-based Unified Posture Manipulation and Thrust Vectoring for Agile and Fault-Tolerant Flight of a Morphing Aerial Robot

    cs.RO 2025-04 conditional novelty 6.0 of 10

    A nonlinear model predictive controller, tested in high-fidelity simulation, lets the M4 morphing robot recover from a fully failed rotor by reconfiguring its legs and remaining thrusters.

Reference graph

Works this paper leans on

26 extracted references · 25 canonical work pages · cited by 1 Pith paper

  1. [1]

    Efficient Path Planning and Tracking for Multi-Modal Legged- Aerial Locomotion Using Integrated Probabilistic Road Maps (PRM) and Reference Governors (RG),

    E. Sihite, B. Mottis, P. Ghanem, A. Ramezani, and M. Gharib, “Efficient Path Planning and Tracking for Multi-Modal Legged- Aerial Locomotion Using Integrated Probabilistic Road Maps (PRM) and Reference Governors (RG),” in 2022 IEEE 61st Conference on Decision and Control (CDC) , Dec. 2022, pp. 764–770

    2022

  2. [2]

    Multi-Modal Mobility Morphobot (M4) with appendage repurpos- ing for locomotion plasticity enhancement,

    E. Sihite, A. Kalantari, R. Nemovi, A. Ramezani, and M. Gharib, “Multi-Modal Mobility Morphobot (M4) with appendage repurpos- ing for locomotion plasticity enhancement,” en, Nature Communica- tions, vol. 14, no. 1, p. 3323, Jun. 2023

    work page 2023

  3. [3]

    A Finite-State Machine for Accommodating Unexpected Large Ground-Height Variations in Bipedal Robot Walking,

    H.-W. Park, A. Ramezani, and J. W. Grizzle, “A Finite-State Machine for Accommodating Unexpected Large Ground-Height Variations in Bipedal Robot Walking,” en,IEEE Transactions on Robotics, vol. 29, no. 2, pp. 331–345, Apr. 2013

    work page 2013

  4. [4]

    Hoeller, N

    D. Hoeller, N. Rudin, D. Sako, and M. Hutter, ANYmal Parkour: Learning Agile Navigation for Quadrupedal Robots , Jun. 2023

    work page 2023

  5. [5]

    Kolvenbach, P

    H. Kolvenbach, P. Arm, E. Hampp, et al. , Traversing Steep and Granular Martian Analog Slopes With a Dynamic Quadrupedal Robot, Jun. 2021

    work page 2021

  6. [6]

    Performance Analysis and Feedback Control of ATRIAS, A Three- Dimensional Bipedal Robot,

    A. Ramezani, J. W. Hurst, K. Akbari Hamed, and J. W. Grizzle, “Performance Analysis and Feedback Control of ATRIAS, A Three- Dimensional Bipedal Robot,” Journal of Dynamic Systems, Measure- ment, and Control , vol. 136, no. 021012, Dec. 2013

    work page 2013

  7. [7]

    NASA/JPL Tumble- weed polar rover,

    A. Behar, J. Matthews, F. Carsey, and J. Jones, “NASA/JPL Tumble- weed polar rover,” in 2004 IEEE Aerospace Conference Proceedings (IEEE Cat. No.04TH8720) , vol. 1, Mar. 2004, 395 V ol.1

    work page 2004

  8. [8]

    Motion control of a spherical mobile robot,

    A. Halme, T. Schonberg, and Y . Wang, “Motion control of a spherical mobile robot,” in Proceedings of 4th IEEE International Workshop on Advanced Motion Control - AMC ’96 - MIE , vol. 1, Mar. 1996, 259–264 vol.1

    work page 1996

Show all 26 references
  1. [9]

    Introducing the

    A. Bicchi, A. Balluchi, D. Prattichizzo, and A. Gorelli, “Introducing the ”SPHERICLE”: An experimental testbed for research and teach- ing in nonholonomy,” in Proceedings of International Conference on Robotics and Automation , vol. 3, Apr. 1997, 2620–2625 vol.3

    1997

  2. [10]

    Simple motion planning strategies for spherobot: A spherical mobile robot,

    R. Mukherjee, M. Minor, and J. Pukrushpan, “Simple motion planning strategies for spherobot: A spherical mobile robot,” in Proceedings of the 38th IEEE Conference on Decision and Control (Cat. No.99CH36304) , vol. 3, Dec. 1999, 2132–2137 vol.3

    1999

  3. [11]

    Introducing Glory: A Novel Strategy for an Omnidirectional Spherical Rolling Robot,

    A. H. Javadi A. and P. Mojabi, “Introducing Glory: A Novel Strategy for an Omnidirectional Spherical Rolling Robot,” Journal of Dynamic Systems, Measurement, and Control , vol. 126, no. 3, pp. 678–683, Dec. 2004

    2004

  4. [12]

    Ourobot—A sensorized closed-kinematic-chain robot for shape-adaptive rolling in rough terrain,

    J. Paskarbeit, S. Beyer, M. Engel, A. Gucze, J. Schr ¨oder, and A. Schneider, “Ourobot—A sensorized closed-kinematic-chain robot for shape-adaptive rolling in rough terrain,” en, Robotics and Au- tonomous Systems , vol. 140, p. 103 715, Jun. 2021

    2021

  5. [13]

    Design and evolution of a modular tensegrity robot platform,

    J. Bruce, K. Caluwaerts, A. Iscen, A. P. Sabelhaus, and V . SunSpiral, “Design and evolution of a modular tensegrity robot platform,” in 2014 IEEE International Conference on Robotics and Automation (ICRA), May 2014, pp. 3483–3489

    2014

  6. [14]

    Dynamic rolling analysis of triangular- bipyramid robot,

    Y . Tian and Y .-A. Yao, “Dynamic rolling analysis of triangular- bipyramid robot,” en, Robotica, vol. 33, no. 4, pp. 884–897, May 2015

    2015

  7. [15]

    Design and locomotion analysis of a novel modular rolling robot,

    X. Wei, Y . Tian, and S. Wen, “Design and locomotion analysis of a novel modular rolling robot,” en, Mechanism and Machine Theory , vol. 133, pp. 23–43, Mar. 2019

    2019

  8. [16]

    Trajectory planning of a rolling robot of closed five-bow-shaped-bar linkage,

    Y . Wang, C. Wu, L. Yu, and Y . Mei, “Trajectory planning of a rolling robot of closed five-bow-shaped-bar linkage,” en, Robotics and Computer-Integrated Manufacturing , vol. 53, pp. 81–92, Oct. 2018

    2018

  9. [17]

    Dynamics of a rolling robot of closed five-arc-shaped-bar linkage,

    Y . Wang, C. Wu, L. Yu, and Y . Mei, “Dynamics of a rolling robot of closed five-arc-shaped-bar linkage,” en, Mechanism and Machine Theory, vol. 121, pp. 75–91, Mar. 2018

    2018

  10. [18]

    Dynamic Rolling for a Modu- lar Loop Robot,

    J. Sastra, S. Chitta, and M. Yim, “Dynamic Rolling for a Modu- lar Loop Robot,” The International Journal of Robotics Research , vol. 28, no. 6, pp. 758–773, Jun. 2009

    2009

  11. [19]

    Crawling and Jumping by a Deformable Robot,

    Y . Sugiyama and S. Hirai, “Crawling and Jumping by a Deformable Robot,” The International Journal of Robotics Research , vol. 25, no. 5-6, pp. 603–620, May 2006

    2006

  12. [20]

    How Strong a Kick Should be to Topple Northeastern’s Tumbling Robot?

    A. Salagame, N. Bhattachan, A. Caetano, et al. , “How Strong a Kick Should be to Topple Northeastern’s Tumbling Robot?” In 2024 IEEE International Conference on Advanced Intelligent Mechatron- ics (AIM) , Jul. 2024, pp. 76–81

    2024

  13. [21]

    Salagame, K

    A. Salagame, K. Gangaraju, H. K. Nallaguntla, E. Sihite, G. Schirner, and A. Ramezani, Loco-Manipulation with Nonimpulsive Contact- Implicit Planning in a Slithering Robot , Apr. 2024

    2024

  14. [22]

    Salagame, H

    A. Salagame, H. K. Nallaguntla, E. Sihite, G. Schirner, and A. Ramezani, Reinforcement Learning-Based Model Matching to Re- duce the Sim-Real Gap in COBRA , en, Jun. 2024

    2024

  15. [23]

    Snake Robot with Tactile Perception Navigates on Large-scale Challenging Terrain,

    S. Jiang, A. Salagame, A. Ramezani, and L. L. S. Wong, “Snake Robot with Tactile Perception Navigates on Large-scale Challenging Terrain,” in 2024 IEEE International Conference on Robotics and Automation (ICRA) , May 2024, pp. 5090–5096

    2024

  16. [24]

    Hierarchical RL-Guided Large-scale Navigation of a Snake Robot,

    S. Jiang, A. Salagame, A. Ramezani, and L. L. Wong, “Hierarchical RL-Guided Large-scale Navigation of a Snake Robot,” in 2024 IEEE International Conference on Advanced Intelligent Mechatron- ics (AIM) , Jul. 2024, pp. 1347–1352

    2024

  17. [25]

    Sidewinding with minimal slip: Snake and robot ascent of sandy slopes,

    H. Marvi, C. Gong, N. Gravish, et al. , “Sidewinding with minimal slip: Snake and robot ascent of sandy slopes,” Science, vol. 346, no. 6206, pp. 224–229, Oct. 2014

    2014

  18. [26]

    Dynamic Posture Manipulation During Tumbling for Closed-Loop Heading Angle Control,

    A. Salagame, E. Sihite, G. Schirner, and A. Ramezani, “Dynamic Posture Manipulation During Tumbling for Closed-Loop Heading Angle Control,” in 2024 IEEE International Conference on Advanced Intelligent Mechatronics (AIM) , Jul. 2024, pp. 64–69

    2024

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.