Extreme dynamic symmetry enables omnidirectional and multifunctional robots

Boxi Xia; Boyuan Chen; Jiaxun Liu

arxiv: 2605.29254 · v1 · pith:TX567TOUnew · submitted 2026-05-28 · 💻 cs.RO · cs.AI

Extreme dynamic symmetry enables omnidirectional and multifunctional robots

Jiaxun Liu , Boxi Xia , Boyuan Chen This is my paper

Pith reviewed 2026-06-29 07:25 UTC · model grok-4.3

classification 💻 cs.RO cs.AI

keywords dynamic symmetrydynamic isotropyspherical robotsomnidirectional locomotionrobot resilienceactuation geometrycenter-of-mass dynamics

0 comments

The pith

Maximizing uniformity in attainable center-of-mass accelerations enables omnidirectional locomotion and multifunctionality in robots.

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

The paper defines dynamic symmetry as the uniformity of attainable center-of-mass accelerations and introduces dynamic isotropy to quantify it. Simulations across more than 1000 morphologies show that higher dynamic isotropy improves trajectory tracking, task success, robustness, resiliency, and energy efficiency, with gains largest near the theoretical limit. The Argus family of spherical robots uses radially oriented linear actuators to achieve this symmetry. A physical 20-leg Argus prototype reached near-extreme isotropy and exhibited orientation-invariant movement, agile traversal of cluttered and deformable terrain, rapid self-stabilization, resilience to actuator failures, and omnidirectional perception during motion. The work positions dynamic symmetry, rather than geometric form alone, as a design principle for agile and robust robots.

Core claim

Dynamic symmetry, formalized as dynamic isotropy measuring the uniformity of attainable center-of-mass accelerations from radially oriented linear actuators, when driven close to its theoretical limit, produces robots with orientation-invariant locomotion, high resiliency, and multifunctionality, as shown by consistent gains in over 1000 simulations and the physical 20-leg Argus variant.

What carries the argument

Dynamic isotropy, the metric of uniformity in attainable center-of-mass accelerations shaped directly by radially oriented linear actuators.

If this is right

Higher dynamic symmetry produces better trajectory tracking and task success rates.
Performance gains in robustness, resiliency, and energy efficiency increase as dynamic isotropy approaches its limit.
Radially oriented actuators in spherical forms can achieve near-extreme isotropy and orientation-invariant locomotion.
Distributed sensing on such platforms supports omnidirectional perception and object interaction during continuous motion.
Partial actuator failures do not prevent continued locomotion when dynamic isotropy is high.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The same radial-actuator principle could be tested on non-spherical morphologies to check whether dynamic isotropy remains beneficial outside the Argus family.
Control algorithms might be simplified because uniform center-of-mass dynamics reduce the need for posture-dependent compensation.
In environments with high uncertainty, such as extraterrestrial surfaces, the approach could lower reliance on complex sensing by making dynamics more predictable.
Energy-efficiency claims could be validated by direct power-consumption measurements on the physical prototype across matched tasks.

Load-bearing premise

That the dynamic isotropy metric itself, rather than correlated factors such as actuator count, placement, or control policy, is the primary cause of the observed performance improvements.

What would settle it

A controlled comparison in which a robot with measurably lower dynamic isotropy matches or exceeds the 20-leg Argus performance in trajectory tracking, terrain traversal, and failure resilience would falsify the causal link.

read the original abstract

Symmetry is a central organizing principle in natural systems, yet its use as a unifying design strategy in robotics has largely remained limited to geometric form. We show that symmetry can instead be leveraged at the level of dynamic actuation capability. We introduce dynamic symmetry, the uniformity of a robot's attainable center-of-mass accelerations, and formalize it through a measure coined as dynamic isotropy. Across more than 1000 simulated morphologies, we found that higher dynamic symmetry consistently improved trajectory tracking, task success, robustness, resiliency, and energy efficiency, with the benefits becoming most pronounced as dynamic isotropy approached its theoretical limit. To study this regime systematically, we developed Argus, a family of spherical robots designed to explore the effects of increasing dynamic symmetry. Members of the Argus family vary in their actuation geometry and dynamic symmetry level while sharing a common architectural principle: radially oriented linear actuators that directly shape the robot's center-of-mass dynamics. Among them, we built a physical 20-leg Argus variant that achieved near-extreme dynamic isotropy and demonstrated orientation-invariant locomotion, agile traversal of cluttered and deformable terrain, rapid self-stabilization, and resilience to partial actuator failures. Its distributed sensing further enabled omnidirectional perception and object interaction during continuous motion. These results show that designing robots for symmetry not only in morphology but also in their attainable dynamics provides a powerful and general pathway toward agility, robustness, and multifunctionality in uncertain terrestrial and extraterrestrial environments.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper defines dynamic isotropy as a uniformity metric on attainable CoM accelerations and shows gains in a 20-leg physical Argus prototype plus 1000+ sims, but actuator count and geometry are not cleanly separated from the metric.

read the letter

The paper's main point is that by designing for uniformity in the dynamic capabilities—specifically how evenly a robot can accelerate its center of mass in all directions—you can get robots that are more reliable and versatile, and they've tested this idea with both lots of simulations and a real machine.

They formalize dynamic isotropy as that uniformity measure and show in over 1000 simulated cases that higher values line up with better trajectory tracking, task completion, robustness to disturbances, and lower energy use. Then they create the Argus family of spherical robots using radial linear actuators, with the physical 20-leg one reaching close to the max isotropy. That hardware version demonstrates consistent movement no matter the orientation, good performance on uneven or soft ground, quick recovery from tipping or damage, and even some sensing for interacting with objects while moving. That's concrete and useful.

The part that needs more work is pinning down that the isotropy is what causes the improvements. The different Argus versions change the number and placement of actuators along with the symmetry level, so without a controlled comparison that keeps those other factors steady, the correlation might be picking up on having more actuators or better distributed forces instead. The stress-test note gets at this, and it seems like a fair point given the description. Also, the results are described more qualitatively than with hard numbers, which makes the strength of the effect harder to assess.

This kind of work is aimed at people designing robots for tough environments where reliability matters, like exploration or disaster response. A reader focused on mechanism design or control would find the Argus examples and the physical tests worth looking at. It has enough substance—the new metric, the scale of the sims, and the working prototype—to warrant a serious review rather than a quick pass.

I would recommend sending it out for peer review.

Referee Report

2 major / 2 minor

Summary. The paper introduces dynamic symmetry, formalized as dynamic isotropy (uniformity of attainable center-of-mass accelerations from radially oriented actuators). It reports that across >1000 simulated morphologies, higher dynamic isotropy correlates with improved trajectory tracking, task success, robustness, resiliency, and energy efficiency, with largest gains near the theoretical limit. The Argus family of spherical robots is introduced to explore this, including a physical 20-leg prototype demonstrating orientation-invariant locomotion, cluttered/deformable terrain traversal, rapid self-stabilization, resilience to actuator failures, and omnidirectional perception/interaction.

Significance. If the central claim holds after isolating causality, the work offers a new design principle emphasizing dynamic actuation symmetry (beyond geometric symmetry) for agile, robust, multifunctional robots in uncertain environments. Credit is due for the scale of the morphology sweep in simulation and the physical hardware demonstration of near-extreme isotropy.

major comments (2)

[Simulation study (abstract; results on 1000+ morphologies)] The central claim (abstract and simulation results) that higher dynamic isotropy is the primary causal driver of performance gains across >1000 morphologies requires explicit isolation from confounds. Morphologies vary in both actuation geometry and symmetry level, yet no ablation study, regression controlling for actuator count/placement statistics, or fixed-geometry isotropy sweep is described to orthogonalize the metric from these covariates.
[Abstract and simulation results] The definition and computation of the dynamic isotropy metric (based on attainable CoM accelerations from radially oriented actuators) is presented as independently testable, but the manuscript does not report quantitative metrics, error bars, simulation protocols, or statistical controls for the performance correlations, making it impossible to assess effect sizes or robustness of the reported improvements.

minor comments (2)

[Abstract; hardware results] The abstract describes physical results qualitatively without data; quantitative metrics, figures, or tables for the 20-leg Argus hardware experiments should be added to the main text or supplementary material for reproducibility.
[Methods/theory section] Notation for dynamic isotropy and related quantities should be introduced with explicit equations early in the methods or theory section to allow readers to verify the measure independently.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments. We address each major point below and indicate planned revisions to strengthen the manuscript.

read point-by-point responses

Referee: [Simulation study (abstract; results on 1000+ morphologies)] The central claim (abstract and simulation results) that higher dynamic isotropy is the primary causal driver of performance gains across >1000 morphologies requires explicit isolation from confounds. Morphologies vary in both actuation geometry and symmetry level, yet no ablation study, regression controlling for actuator count/placement statistics, or fixed-geometry isotropy sweep is described to orthogonalize the metric from these covariates.

Authors: We agree that the simulation results demonstrate correlation rather than fully isolated causality, as morphologies were varied in both geometry and symmetry level. No ablation studies or controlled regressions were performed in the original manuscript. Since the full dataset of over 1000 morphologies is available, we will add a multiple regression analysis controlling for actuator count and placement statistics, along with a discussion of effect sizes, to better isolate the contribution of dynamic isotropy in the revised manuscript. revision: yes
Referee: [Abstract and simulation results] The definition and computation of the dynamic isotropy metric (based on attainable CoM accelerations from radially oriented actuators) is presented as independently testable, but the manuscript does not report quantitative metrics, error bars, simulation protocols, or statistical controls for the performance correlations, making it impossible to assess effect sizes or robustness of the reported improvements.

Authors: The Methods section provides the definition and computation procedure for dynamic isotropy. However, the manuscript does not include the requested quantitative metrics such as effect sizes, error bars, or detailed statistical controls. We will revise the Results and Methods sections to report these details, including simulation protocols, confidence intervals, and robustness checks for the reported correlations. revision: yes

Circularity Check

0 steps flagged

No circularity; empirical correlation tested on independently defined metric

full rationale

The paper introduces dynamic isotropy as an independent definition (uniformity of attainable CoM accelerations from radially oriented actuators) and reports an empirical correlation with performance metrics across >1000 simulated morphologies plus hardware validation. No equations, fitted parameters, or self-citations are presented as load-bearing derivations that reduce the performance claims to the isotropy definition by construction. The chain consists of metric definition followed by external simulation/hardware testing, which is self-contained and non-circular.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 1 invented entities

The central claim rests on a newly introduced scalar measure of dynamic isotropy whose explicit definition, computation, and independence from actuator geometry are not supplied in the abstract; the performance benefits are asserted to follow from proximity to its theoretical limit without intermediate derivation steps shown.

axioms (1)

domain assumption Uniformity of attainable center-of-mass accelerations can be meaningfully quantified as a single scalar (dynamic isotropy) that predicts multiple performance metrics
This is the load-bearing definition introduced by the paper and invoked to explain simulation and hardware outcomes.

invented entities (1)

dynamic isotropy no independent evidence
purpose: Scalar metric of uniformity in attainable CoM accelerations
Newly coined quantity whose mathematical construction is not detailed in the abstract.

pith-pipeline@v0.9.1-grok · 5789 in / 1324 out tokens · 24959 ms · 2026-06-29T07:25:54.377484+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

64 extracted references · 13 canonical work pages · 2 internal anchors

[1]

Capek,RUR (Rossum’s universal robots)(Penguin) (2004)

K. Capek,RUR (Rossum’s universal robots)(Penguin) (2004)

2004
[2]

Pfeifer, M

R. Pfeifer, M. Lungarella, F. Iida, Self-organization, embodiment, and biologically inspired robotics.Science318(5853), 1088–1093 (2007)

2007
[3]

Trivedi, C

D. Trivedi, C. D. Rahn, W. M. Kier, I. D. Walker, Soft robotics: Biological inspiration, state of the art, and future research.Appl. Bionics Biomech.5(3), 99–117 (2008), doi:10.1080/ 11762320802557865

2008
[4]

F. Iida, A. J. Ijspeert, Biologically inspired robotics, inSpringer Handbook of Robotics (Springer), pp. 2015–2034 (2016)

2015
[5]

Bar-Cohen, C

Y. Bar-Cohen, C. Breazeal, Biologically inspired intelligent robots.Smart Structures and Materials 2003: Electroactive Polymer Actuators and Devices (EAPAD)5051, 14–20 (2003)

2003
[6]

N. F. Lepora, P. Verschure, T. J. Prescott, The state of the art in biomimetics.Bioinspir. Biomim. 8(1), 013001 (2013)

2013
[7]

Raibert, K

M. Raibert, K. Blankespoor, G. Nelson, R. Playter, Bigdog, the rough-terrain quadruped robot. IFAC Proc. Vol.41(2), 10822–10825 (2008)

2008
[8]

Hutter, C

M. Hutter, C. Gehring, D. Jud, A. Lauber, C. D. Bellicoso, V. Tsounis, J. Hwangbo, K. Bodie, P. Fankhauser, M. Bloesch, R. Diethelm, S. Bachmann, A. Melzer, M. Hoepflinger, Anymal-a highly mobile and dynamic quadrupedal robot, in2016 IEEE/RSJ international conference on intelligent robots and systems (IROS)(IEEE) (2016), pp. 38–44

2016
[9]

Bledt, M

G. Bledt, M. J. Powell, B. Katz, J. Di Carlo, P. M. Wensing, S. Kim, Mit cheetah 3: Design and control of a robust, dynamic quadruped robot, in2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)(IEEE) (2018), pp. 2245–2252

2018
[10]

Metta, L

G. Metta, L. Natale, F. Nori, G. Sandini, D. Vernon, L. Fadiga, C. Von Hofsten, K. Rosander, M. Lopes, J. Santos-Victor, A. Bernardino, L. Montesano, The iCub humanoid robot: An open-systems platform for research in cognitive development.Neural Netw.23(8-9), 1125– 1134 (2010). 37

2010
[11]

SaLoutos, E

A. SaLoutos, E. Stanger-Jones, Y. Ding, M. Chignoli, S. Kim, Design and development of the mit humanoid: A dynamic and robust research platform, in2023 IEEE-RAS 22nd International Conference on Humanoid Robots (Humanoids)(IEEE) (2023), pp. 1–8

2023
[12]

B. Xia, B. Li, J. Lee, M. Scutari, B. Chen, The Duke Humanoid: Design and Control for Energy-Efficient Bipedal Locomotion Using Passive Dynamics, in2025 IEEE/RSJ Interna- tional Conference on Intelligent Robots and Systems (IROS)(IEEE) (2025), pp. 6579–6586

2025
[13]

Q. Liao, B. Zhang, X. Huang, X. Huang, Z. Li, K. Sreenath, Berkeley humanoid: A research platform for learning-based control, in2025 IEEE International Conference on Robotics and Automation (ICRA)(IEEE) (2025), pp. 2897–2904

2025
[14]

Mattar, A survey of bio-inspired robotics hands implementation: New directions in dexterous manipulation.Robot

E. Mattar, A survey of bio-inspired robotics hands implementation: New directions in dexterous manipulation.Robot. Auton. Syst.61(5), 517–544 (2013)

2013
[15]

R. R. Ma, A. M. Dollar, On dexterity and dexterous manipulation, in2011 15th International Conference on Advanced Robotics (ICAR)(IEEE) (2011), pp. 1–7

2011
[16]

B. Sun, W. Li, Z. Wang, Y. Zhu, Q. He, X. Guan, G. Dai, D. Yuan, A. Li, W. Cui, D. Fan, Recent progress in modeling and control of bio-inspired fish robots.J. Mar. Sci. Eng.10(6), 773 (2022)

2022
[17]

Baines, S

R. Baines, S. K. Patiballa, J. Booth, L. Ramirez, T. Sipple, A. Garcia, F. Fish, R. Kramer- Bottiglio, Multi-environment robotic transitions through adaptive morphogenesis.Nature 610(7931), 283–289 (2022)

2022
[18]

G. Li, X. Chen, F. Zhou, Y. Liang, Y. Xiao, X. Cao, Z. Zhang, M. Zhang, B. Wu, S. Yin, Y. Xu, H. Fan, Z. Chen, W. Song, W. Yang, B. Pan, J. Hou, W. Zou, S. He, X. Yang, G. Mao, Z. Jia, H. Zhou, T. Li, S. Qu, Z. Xu, Z. Huang, Y. Luo, T. Xie, J. Gu, S. Zhu, W. Yang, Self-powered soft robot in the Mariana Trench.Nature591(7848), 66–71 (2021)

2021
[19]

W. D. Shin, H.-V. Phan, M. A. Daley, A. J. Ijspeert, D. Floreano, Fast ground-to-air transition with avian-inspired multifunctional legs.Nature636(8041), 86–91 (2024). 38

2024
[20]

Chang, D

E. Chang, D. D. Chin, D. Lentink, Bird-inspired reflexive morphing enables rudderless flight. Sci. Robot.9(96), eado4535 (2024)

2024
[21]

Langowski, P

J. Langowski, P. Sharma, A. L. Shoushtari, In the soft grip of nature.Sci. Robot.5(49), eabd9120 (2020)

2020
[22]

Ocklenburg, A

S. Ocklenburg, A. Mundorf, Symmetry and asymmetry in biological structures.Proc. Natl. Acad. Sci. U.S.A.119(28), e2204881119 (2022)

2022
[23]

Enquist, A

M. Enquist, A. Arak, Symmetry, beauty and evolution.Nature372(6502), 169–172 (1994)

1994
[24]

X. Zhu, D. Wang, O. Biza, G. Su, R. Walters, R. Platt, Sample Efficient Grasp Learning Using Equivariant Models, inRobotics: Science and Systems(2022)

2022
[25]

D. Wang, R. Walters, X. Zhu, R. Platt, Equivariant𝑞learning in spatial action spaces, in Conference on Robot Learning(PMLR) (2022), pp. 1713–1723

2022
[26]

D. Wang, M. Jia, X. Zhu, R. Walters, R. Platt, On-Robot Learning With Equivariant Models, inConference on Robot Learning(PMLR) (2023), pp. 1345–1354

2023
[27]

Z. Su, X. Huang, D. Ordo ˜nez-Apraez, Y. Li, Z. Li, Q. Liao, G. Turrisi, M. Pontil, C. Semini, Y. Wu, Leveraging symmetry in rl-based legged locomotion control, in2024 IEEE/RSJ Inter- national Conference on Intelligent Robots and Systems (IROS)(IEEE) (2024), pp. 6899–6906

2024
[28]

Mittal, N

M. Mittal, N. Rudin, V. Klemm, A. Allshire, M. Hutter, Symmetry considerations for learn- ing task symmetric robot policies, in2024 IEEE International Conference on Robotics and Automation (ICRA)(IEEE) (2024), pp. 7433–7439

2024
[29]

D. O. Apraez, G. Turrisi, V. Kostic, M. Martin, A. Agudo, F. Moreno-Noguer, M. Pontil, C. Semini, C. Mastalli, Morphological symmetries in robotics.Int. J. Robot. Res.44, 1743– 1766 (2024)

2024
[30]

S. Yan, B. Zhang, Y. Zhang, J. Boedecker, W. Burgard, Learning continuous control with geometric regularity from robot intrinsic symmetry, in2024 IEEE International Conference on Robotics and Automation (ICRA)(IEEE) (2024), pp. 49–55. 39

2024
[31]

Frazzoli, M

E. Frazzoli, M. A. Dahleh, E. Feron, Maneuver-based motion planning for nonlinear systems with symmetries.IEEE Trans. Robot.21(6), 1077–1091 (2005)

2005
[32]

D. K. Pai, R. A. Barman, S. K. Ralph, Platonic beasts: a new family of multilimbed robots, in Proceedings of the 1994 IEEE International Conference on Robotics and Automation(IEEE) (1994), pp. 1019–1025

1994
[33]

Nozaki, Y

H. Nozaki, Y. Kujirai, R. Niiyama, Y. Kawahara, T. Yonezawa, J. Nakazawa, Continuous Shape Changing Locomotion of 32-legged Spherical Robot, in2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)(2018), pp. 2721–2726, doi:10.1109/ IROS.2018.8593791

work page arXiv 2018
[34]

Gheorghe, N

V. Gheorghe, N. Alexandrescu, D. Duminica, L. A. Cartal, Rolling robot with radial extending legs, in2010 3rd International Symposium on Resilient Control Systems(2010), pp. 107–112, doi:10.1109/ISRCS.2010.5603951

work page doi:10.1109/isrcs.2010.5603951 2010
[35]

Liu, Y.-a

R. Liu, Y.-a. Yao, Y. Li, Design and analysis of a deployable tetrahedron-based mo- bile robot constructed by Sarrus linkages.Mech. Mach. Theory152, 103964 (2020), doi: 10.1016/j.mechmachtheory.2020.103964

work page doi:10.1016/j.mechmachtheory.2020.103964 2020
[36]

C. Paul, F. J. Valero-Cuevas, H. Lipson, Design and control of tensegrity robots for locomotion. IEEE Trans. Robot.22(5), 944–957 (2006)

2006
[37]

Vespignani, J

M. Vespignani, J. M. Friesen, V. SunSpiral, J. Bruce, Design of SUPERball v2, a Compliant Tensegrity Robot for Absorbing Large Impacts, in2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)(IEEE) (2018), pp. 2865–2871, doi:10.1109/IROS. 2018.8594233

work page doi:10.1109/iros 2018
[38]

K. Kim, A. K. Agogino, A. M. Agogino, Rolling Locomotion of Cable-Driven Soft Spherical Tensegrity Robots.Soft Robot.7(3), 346–361 (2020), doi:10.1089/soro.2019.0056

work page doi:10.1089/soro.2019.0056 2020
[39]

Jeong, I

J. Jeong, I. Kim, Y. Choi, S. Lim, S. Kim, H. Kang, D. Shah, R. Baines, J. W. Booth, R. Kramer- Bottiglio, S. Y. Kim, Spikebot: A Multigait Tensegrity Robot with Linearly Extending Struts. Soft Robot.11(2), 207–217 (2024), doi:10.1089/soro.2023.0030. 40

work page doi:10.1089/soro.2023.0030 2024
[40]

Surovik, K

D. Surovik, K. Wang, M. Vespignani, J. Bruce, K. E. Bekris, Adaptive tensegrity locomotion: Controlling a compliant icosahedron with symmetry-reduced reinforcement learning.Int. J. Robot. Res.40(1), 375–396 (2021), doi:10.1177/0278364919859443

work page doi:10.1177/0278364919859443 2021
[41]

J. Liu, Z. Xu, J. Lu, X. Gu, J. Wu, LCRBot: Load-Carrying Rolling Robot Based on Truncated Hexahedral Tensegrity.J. Field Robot.42(6), 2454–2468 (2025)

2025
[42]

N. Chen, K. Wang, W. R. Johnson III, R. Kramer-Bottiglio, K. Bekris, M. Aanjaneya, Learning Differentiable Tensegrity Dynamics using Graph Neural Networks, inProceedings of the Conference on Robot Learning (CoRL)(2024)

2024
[43]

J. K. Salisbury, J. J. Craig, Articulated hands: Force control and kinematic issues.Int. J. Robot. Res.1(1), 4–17 (1982)

1982
[44]

Yoshikawa, Manipulability of robotic mechanisms.Int

T. Yoshikawa, Manipulability of robotic mechanisms.Int. J. Robot. Res.4(2), 3–9 (1985)

1985
[45]

C. A. Klein, B. E. Blaho, Dexterity measures for the design and control of kinematically redundant manipulators.Int. J. Robot. Res.6(2), 72–83 (1987)

1987
[46]

O. Ma, J. Angeles, The concept of dynamic isotropy and its applications to inverse kinematics and trajectory planning, inProceedings., IEEE International Conference on Robotics and Automation(IEEE) (1990), pp. 481–486

1990
[47]

C. A. Klein, T. A. Miklos, Spatial robotic isotropy.Int. J. Robot. Res.10(4), 426–437 (1991)

1991
[48]

S. Kim, I. Jeong, S. Lee, Systematic isotropy analysis of a mobile robot with three active caster wheels, inInternational Conference on Intelligent Computing(Springer) (2007), pp. 587–597

2007
[49]

O. Ma, J. Angeles, Optimum design of manipulators under dynamic isotropy conditions, in[1993] Proceedings IEEE International Conference on Robotics and Automation(IEEE) (1993), pp. 470–475

1993
[50]

Yoshikawa, Dynamic manipulability of robot manipulators.Trans

T. Yoshikawa, Dynamic manipulability of robot manipulators.Trans. Soc. Instrum. Control Eng.21(9), 970–975 (1985). 41

1985
[51]

A. Dasgupta, SpiderBot DeepRL: A Custom-Designed Spider Robot Trained to Walk Using Deep Reinforcement Learning in a PyBullet Simulation,https://github.com/ arijit-dasgupta/SpiderBot_DeepRL(2026), gitHub repository. Accessed: 2026-02-23

2026
[52]

Nozaki, R

H. Nozaki, R. Niiyama, T. Yonezawa, J. Nakazawa, Shape changing locomotion by spiny mul- tipedal robot, in2017 IEEE International Conference on Robotics and Biomimetics (ROBIO) (2017), pp. 2162–2166, doi:10.1109/ROBIO.2017.8324739

work page doi:10.1109/robio.2017.8324739 2017
[53]

Glasser, A

L. Glasser, A. Every, Energies and spacings of point charges on a sphere.J. Phys. A25(9), 2473 (1992)

1992
[54]

Zheng, Y

Y. Zheng, Y. Li, Y. Lu, M. Wang, X. Xu, C. Zhou, Y. Luo, Robustness evaluation for rolling gaits of a six-strut tensegrity robot.Int. J. Adv. Robot. Syst.18(1), 1–11 (2021), doi:10.1177/ 1729881421993638

2021
[55]

F. Xu, X. Zhao, M. Yue, A Physics-Driven Closed-Loop Motion Planning Method for Spherical Multi-Expandable-Limb Robots.IEEE Trans. Ind. Electron.71(12), 16087–16097 (2024), doi: 10.1109/TIE.2024.3390725

work page doi:10.1109/tie.2024.3390725 2024
[56]

D. Yang, Y. Liu, Y. Yu, A General Locomotion Approach for a Novel Multi-legged Spherical Robot, in2023 IEEE International Conference on Robotics and Automation (ICRA)(2023), pp. 10146–10152, doi:10.1109/ICRA48891.2023.10160881

work page doi:10.1109/icra48891.2023.10160881 2023
[57]

H. Dong, J. Zhang, T. Wang, C. Zhang, Symmetry-aware robot design with structured sub- groups, inInternational Conference on Machine Learning(PMLR) (2023), pp. 8334–8355

2023
[58]

Ghaffari, R

M. Ghaffari, R. Zhang, M. Zhu, C. E. Lin, T.-Y. Lin, S. Teng, T. Li, T. Liu, J. Song, Progress in symmetry preserving robot perception and control through geometry and learning.Front. Robot. AI9, 969380 (2022)

2022
[59]

L. P. Kaelbling, M. L. Littman, A. R. Cassandra, Planning and acting in partially observable stochastic domains.Artif. Intell.101(1-2), 99–134 (1998)

1998
[60]

Isaac Gym: High Performance GPU-Based Physics Simulation For Robot Learning

V. Makoviychuk, L. Wawrzyniak, Y. Guo, M. Lu, K. Storey, M. Macklin, D. Hoeller, N. Rudin, A. Allshire, A. Handa, G. State, Isaac gym: High performance gpu-based physics simulation for robot learning,https://arxiv.org/abs/2108.10470(2021). 42

work page internal anchor Pith review Pith/arXiv arXiv 2021
[61]

Proximal Policy Optimization Algorithms

J. Schulman, F. Wolski, P. Dhariwal, A. Radford, O. Klimov, Proximal policy optimization algorithms,https://arxiv.org/abs/1707.06347(2017)

work page internal anchor Pith review Pith/arXiv arXiv 2017
[62]

C. R. Qi, H. Su, K. Mo, L. J. Guibas, Pointnet: Deep learning on point sets for 3d classification and segmentation, inProceedings of the IEEE conference on computer vision and pattern recognition(2017), pp. 652–660

2017
[63]

robot descriptions, awesome-robot-descriptions, GitHub repository (2026),https:// github.com/robot-descriptions/awesome-robot-descriptions, accessed: 2026- 02-23

2026
[64]

D. Yang, Y. Liu, F. Ding, Y. Yu, Bionic Multi-legged Robot Based on End-to-end Artificial Neural Network Control, in2022 IEEE International Conference on Cyborg and Bionic Systems (CBS)(2023), pp. 104–109, doi:10.1109/CBS55922.2023.10115331. Acknowledgments: The authors would like to thank Edwin Ma, Jacob Lee, and Max Li for their help in 3D printing. We ...

work page doi:10.1109/cbs55922.2023.10115331 2023

[1] [1]

Capek,RUR (Rossum’s universal robots)(Penguin) (2004)

K. Capek,RUR (Rossum’s universal robots)(Penguin) (2004)

2004

[2] [2]

Pfeifer, M

R. Pfeifer, M. Lungarella, F. Iida, Self-organization, embodiment, and biologically inspired robotics.Science318(5853), 1088–1093 (2007)

2007

[3] [3]

Trivedi, C

D. Trivedi, C. D. Rahn, W. M. Kier, I. D. Walker, Soft robotics: Biological inspiration, state of the art, and future research.Appl. Bionics Biomech.5(3), 99–117 (2008), doi:10.1080/ 11762320802557865

2008

[4] [4]

F. Iida, A. J. Ijspeert, Biologically inspired robotics, inSpringer Handbook of Robotics (Springer), pp. 2015–2034 (2016)

2015

[5] [5]

Bar-Cohen, C

Y. Bar-Cohen, C. Breazeal, Biologically inspired intelligent robots.Smart Structures and Materials 2003: Electroactive Polymer Actuators and Devices (EAPAD)5051, 14–20 (2003)

2003

[6] [6]

N. F. Lepora, P. Verschure, T. J. Prescott, The state of the art in biomimetics.Bioinspir. Biomim. 8(1), 013001 (2013)

2013

[7] [7]

Raibert, K

M. Raibert, K. Blankespoor, G. Nelson, R. Playter, Bigdog, the rough-terrain quadruped robot. IFAC Proc. Vol.41(2), 10822–10825 (2008)

2008

[8] [8]

Hutter, C

M. Hutter, C. Gehring, D. Jud, A. Lauber, C. D. Bellicoso, V. Tsounis, J. Hwangbo, K. Bodie, P. Fankhauser, M. Bloesch, R. Diethelm, S. Bachmann, A. Melzer, M. Hoepflinger, Anymal-a highly mobile and dynamic quadrupedal robot, in2016 IEEE/RSJ international conference on intelligent robots and systems (IROS)(IEEE) (2016), pp. 38–44

2016

[9] [9]

Bledt, M

G. Bledt, M. J. Powell, B. Katz, J. Di Carlo, P. M. Wensing, S. Kim, Mit cheetah 3: Design and control of a robust, dynamic quadruped robot, in2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)(IEEE) (2018), pp. 2245–2252

2018

[10] [10]

Metta, L

G. Metta, L. Natale, F. Nori, G. Sandini, D. Vernon, L. Fadiga, C. Von Hofsten, K. Rosander, M. Lopes, J. Santos-Victor, A. Bernardino, L. Montesano, The iCub humanoid robot: An open-systems platform for research in cognitive development.Neural Netw.23(8-9), 1125– 1134 (2010). 37

2010

[11] [11]

SaLoutos, E

A. SaLoutos, E. Stanger-Jones, Y. Ding, M. Chignoli, S. Kim, Design and development of the mit humanoid: A dynamic and robust research platform, in2023 IEEE-RAS 22nd International Conference on Humanoid Robots (Humanoids)(IEEE) (2023), pp. 1–8

2023

[12] [12]

B. Xia, B. Li, J. Lee, M. Scutari, B. Chen, The Duke Humanoid: Design and Control for Energy-Efficient Bipedal Locomotion Using Passive Dynamics, in2025 IEEE/RSJ Interna- tional Conference on Intelligent Robots and Systems (IROS)(IEEE) (2025), pp. 6579–6586

2025

[13] [13]

Q. Liao, B. Zhang, X. Huang, X. Huang, Z. Li, K. Sreenath, Berkeley humanoid: A research platform for learning-based control, in2025 IEEE International Conference on Robotics and Automation (ICRA)(IEEE) (2025), pp. 2897–2904

2025

[14] [14]

Mattar, A survey of bio-inspired robotics hands implementation: New directions in dexterous manipulation.Robot

E. Mattar, A survey of bio-inspired robotics hands implementation: New directions in dexterous manipulation.Robot. Auton. Syst.61(5), 517–544 (2013)

2013

[15] [15]

R. R. Ma, A. M. Dollar, On dexterity and dexterous manipulation, in2011 15th International Conference on Advanced Robotics (ICAR)(IEEE) (2011), pp. 1–7

2011

[16] [16]

B. Sun, W. Li, Z. Wang, Y. Zhu, Q. He, X. Guan, G. Dai, D. Yuan, A. Li, W. Cui, D. Fan, Recent progress in modeling and control of bio-inspired fish robots.J. Mar. Sci. Eng.10(6), 773 (2022)

2022

[17] [17]

Baines, S

R. Baines, S. K. Patiballa, J. Booth, L. Ramirez, T. Sipple, A. Garcia, F. Fish, R. Kramer- Bottiglio, Multi-environment robotic transitions through adaptive morphogenesis.Nature 610(7931), 283–289 (2022)

2022

[18] [18]

G. Li, X. Chen, F. Zhou, Y. Liang, Y. Xiao, X. Cao, Z. Zhang, M. Zhang, B. Wu, S. Yin, Y. Xu, H. Fan, Z. Chen, W. Song, W. Yang, B. Pan, J. Hou, W. Zou, S. He, X. Yang, G. Mao, Z. Jia, H. Zhou, T. Li, S. Qu, Z. Xu, Z. Huang, Y. Luo, T. Xie, J. Gu, S. Zhu, W. Yang, Self-powered soft robot in the Mariana Trench.Nature591(7848), 66–71 (2021)

2021

[19] [19]

W. D. Shin, H.-V. Phan, M. A. Daley, A. J. Ijspeert, D. Floreano, Fast ground-to-air transition with avian-inspired multifunctional legs.Nature636(8041), 86–91 (2024). 38

2024

[20] [20]

Chang, D

E. Chang, D. D. Chin, D. Lentink, Bird-inspired reflexive morphing enables rudderless flight. Sci. Robot.9(96), eado4535 (2024)

2024

[21] [21]

Langowski, P

J. Langowski, P. Sharma, A. L. Shoushtari, In the soft grip of nature.Sci. Robot.5(49), eabd9120 (2020)

2020

[22] [22]

Ocklenburg, A

S. Ocklenburg, A. Mundorf, Symmetry and asymmetry in biological structures.Proc. Natl. Acad. Sci. U.S.A.119(28), e2204881119 (2022)

2022

[23] [23]

Enquist, A

M. Enquist, A. Arak, Symmetry, beauty and evolution.Nature372(6502), 169–172 (1994)

1994

[24] [24]

X. Zhu, D. Wang, O. Biza, G. Su, R. Walters, R. Platt, Sample Efficient Grasp Learning Using Equivariant Models, inRobotics: Science and Systems(2022)

2022

[25] [25]

D. Wang, R. Walters, X. Zhu, R. Platt, Equivariant𝑞learning in spatial action spaces, in Conference on Robot Learning(PMLR) (2022), pp. 1713–1723

2022

[26] [26]

D. Wang, M. Jia, X. Zhu, R. Walters, R. Platt, On-Robot Learning With Equivariant Models, inConference on Robot Learning(PMLR) (2023), pp. 1345–1354

2023

[27] [27]

Z. Su, X. Huang, D. Ordo ˜nez-Apraez, Y. Li, Z. Li, Q. Liao, G. Turrisi, M. Pontil, C. Semini, Y. Wu, Leveraging symmetry in rl-based legged locomotion control, in2024 IEEE/RSJ Inter- national Conference on Intelligent Robots and Systems (IROS)(IEEE) (2024), pp. 6899–6906

2024

[28] [28]

Mittal, N

M. Mittal, N. Rudin, V. Klemm, A. Allshire, M. Hutter, Symmetry considerations for learn- ing task symmetric robot policies, in2024 IEEE International Conference on Robotics and Automation (ICRA)(IEEE) (2024), pp. 7433–7439

2024

[29] [29]

D. O. Apraez, G. Turrisi, V. Kostic, M. Martin, A. Agudo, F. Moreno-Noguer, M. Pontil, C. Semini, C. Mastalli, Morphological symmetries in robotics.Int. J. Robot. Res.44, 1743– 1766 (2024)

2024

[30] [30]

S. Yan, B. Zhang, Y. Zhang, J. Boedecker, W. Burgard, Learning continuous control with geometric regularity from robot intrinsic symmetry, in2024 IEEE International Conference on Robotics and Automation (ICRA)(IEEE) (2024), pp. 49–55. 39

2024

[31] [31]

Frazzoli, M

E. Frazzoli, M. A. Dahleh, E. Feron, Maneuver-based motion planning for nonlinear systems with symmetries.IEEE Trans. Robot.21(6), 1077–1091 (2005)

2005

[32] [32]

D. K. Pai, R. A. Barman, S. K. Ralph, Platonic beasts: a new family of multilimbed robots, in Proceedings of the 1994 IEEE International Conference on Robotics and Automation(IEEE) (1994), pp. 1019–1025

1994

[33] [33]

Nozaki, Y

H. Nozaki, Y. Kujirai, R. Niiyama, Y. Kawahara, T. Yonezawa, J. Nakazawa, Continuous Shape Changing Locomotion of 32-legged Spherical Robot, in2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)(2018), pp. 2721–2726, doi:10.1109/ IROS.2018.8593791

work page arXiv 2018

[34] [34]

Gheorghe, N

V. Gheorghe, N. Alexandrescu, D. Duminica, L. A. Cartal, Rolling robot with radial extending legs, in2010 3rd International Symposium on Resilient Control Systems(2010), pp. 107–112, doi:10.1109/ISRCS.2010.5603951

work page doi:10.1109/isrcs.2010.5603951 2010

[35] [35]

Liu, Y.-a

R. Liu, Y.-a. Yao, Y. Li, Design and analysis of a deployable tetrahedron-based mo- bile robot constructed by Sarrus linkages.Mech. Mach. Theory152, 103964 (2020), doi: 10.1016/j.mechmachtheory.2020.103964

work page doi:10.1016/j.mechmachtheory.2020.103964 2020

[36] [36]

C. Paul, F. J. Valero-Cuevas, H. Lipson, Design and control of tensegrity robots for locomotion. IEEE Trans. Robot.22(5), 944–957 (2006)

2006

[37] [37]

Vespignani, J

M. Vespignani, J. M. Friesen, V. SunSpiral, J. Bruce, Design of SUPERball v2, a Compliant Tensegrity Robot for Absorbing Large Impacts, in2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)(IEEE) (2018), pp. 2865–2871, doi:10.1109/IROS. 2018.8594233

work page doi:10.1109/iros 2018

[38] [38]

K. Kim, A. K. Agogino, A. M. Agogino, Rolling Locomotion of Cable-Driven Soft Spherical Tensegrity Robots.Soft Robot.7(3), 346–361 (2020), doi:10.1089/soro.2019.0056

work page doi:10.1089/soro.2019.0056 2020

[39] [39]

Jeong, I

J. Jeong, I. Kim, Y. Choi, S. Lim, S. Kim, H. Kang, D. Shah, R. Baines, J. W. Booth, R. Kramer- Bottiglio, S. Y. Kim, Spikebot: A Multigait Tensegrity Robot with Linearly Extending Struts. Soft Robot.11(2), 207–217 (2024), doi:10.1089/soro.2023.0030. 40

work page doi:10.1089/soro.2023.0030 2024

[40] [40]

Surovik, K

D. Surovik, K. Wang, M. Vespignani, J. Bruce, K. E. Bekris, Adaptive tensegrity locomotion: Controlling a compliant icosahedron with symmetry-reduced reinforcement learning.Int. J. Robot. Res.40(1), 375–396 (2021), doi:10.1177/0278364919859443

work page doi:10.1177/0278364919859443 2021

[41] [41]

J. Liu, Z. Xu, J. Lu, X. Gu, J. Wu, LCRBot: Load-Carrying Rolling Robot Based on Truncated Hexahedral Tensegrity.J. Field Robot.42(6), 2454–2468 (2025)

2025

[42] [42]

N. Chen, K. Wang, W. R. Johnson III, R. Kramer-Bottiglio, K. Bekris, M. Aanjaneya, Learning Differentiable Tensegrity Dynamics using Graph Neural Networks, inProceedings of the Conference on Robot Learning (CoRL)(2024)

2024

[43] [43]

J. K. Salisbury, J. J. Craig, Articulated hands: Force control and kinematic issues.Int. J. Robot. Res.1(1), 4–17 (1982)

1982

[44] [44]

Yoshikawa, Manipulability of robotic mechanisms.Int

T. Yoshikawa, Manipulability of robotic mechanisms.Int. J. Robot. Res.4(2), 3–9 (1985)

1985

[45] [45]

C. A. Klein, B. E. Blaho, Dexterity measures for the design and control of kinematically redundant manipulators.Int. J. Robot. Res.6(2), 72–83 (1987)

1987

[46] [46]

O. Ma, J. Angeles, The concept of dynamic isotropy and its applications to inverse kinematics and trajectory planning, inProceedings., IEEE International Conference on Robotics and Automation(IEEE) (1990), pp. 481–486

1990

[47] [47]

C. A. Klein, T. A. Miklos, Spatial robotic isotropy.Int. J. Robot. Res.10(4), 426–437 (1991)

1991

[48] [48]

S. Kim, I. Jeong, S. Lee, Systematic isotropy analysis of a mobile robot with three active caster wheels, inInternational Conference on Intelligent Computing(Springer) (2007), pp. 587–597

2007

[49] [49]

O. Ma, J. Angeles, Optimum design of manipulators under dynamic isotropy conditions, in[1993] Proceedings IEEE International Conference on Robotics and Automation(IEEE) (1993), pp. 470–475

1993

[50] [50]

Yoshikawa, Dynamic manipulability of robot manipulators.Trans

T. Yoshikawa, Dynamic manipulability of robot manipulators.Trans. Soc. Instrum. Control Eng.21(9), 970–975 (1985). 41

1985

[51] [51]

A. Dasgupta, SpiderBot DeepRL: A Custom-Designed Spider Robot Trained to Walk Using Deep Reinforcement Learning in a PyBullet Simulation,https://github.com/ arijit-dasgupta/SpiderBot_DeepRL(2026), gitHub repository. Accessed: 2026-02-23

2026

[52] [52]

Nozaki, R

H. Nozaki, R. Niiyama, T. Yonezawa, J. Nakazawa, Shape changing locomotion by spiny mul- tipedal robot, in2017 IEEE International Conference on Robotics and Biomimetics (ROBIO) (2017), pp. 2162–2166, doi:10.1109/ROBIO.2017.8324739

work page doi:10.1109/robio.2017.8324739 2017

[53] [53]

Glasser, A

L. Glasser, A. Every, Energies and spacings of point charges on a sphere.J. Phys. A25(9), 2473 (1992)

1992

[54] [54]

Zheng, Y

Y. Zheng, Y. Li, Y. Lu, M. Wang, X. Xu, C. Zhou, Y. Luo, Robustness evaluation for rolling gaits of a six-strut tensegrity robot.Int. J. Adv. Robot. Syst.18(1), 1–11 (2021), doi:10.1177/ 1729881421993638

2021

[55] [55]

F. Xu, X. Zhao, M. Yue, A Physics-Driven Closed-Loop Motion Planning Method for Spherical Multi-Expandable-Limb Robots.IEEE Trans. Ind. Electron.71(12), 16087–16097 (2024), doi: 10.1109/TIE.2024.3390725

work page doi:10.1109/tie.2024.3390725 2024

[56] [56]

D. Yang, Y. Liu, Y. Yu, A General Locomotion Approach for a Novel Multi-legged Spherical Robot, in2023 IEEE International Conference on Robotics and Automation (ICRA)(2023), pp. 10146–10152, doi:10.1109/ICRA48891.2023.10160881

work page doi:10.1109/icra48891.2023.10160881 2023

[57] [57]

H. Dong, J. Zhang, T. Wang, C. Zhang, Symmetry-aware robot design with structured sub- groups, inInternational Conference on Machine Learning(PMLR) (2023), pp. 8334–8355

2023

[58] [58]

Ghaffari, R

M. Ghaffari, R. Zhang, M. Zhu, C. E. Lin, T.-Y. Lin, S. Teng, T. Li, T. Liu, J. Song, Progress in symmetry preserving robot perception and control through geometry and learning.Front. Robot. AI9, 969380 (2022)

2022

[59] [59]

L. P. Kaelbling, M. L. Littman, A. R. Cassandra, Planning and acting in partially observable stochastic domains.Artif. Intell.101(1-2), 99–134 (1998)

1998

[60] [60]

Isaac Gym: High Performance GPU-Based Physics Simulation For Robot Learning

V. Makoviychuk, L. Wawrzyniak, Y. Guo, M. Lu, K. Storey, M. Macklin, D. Hoeller, N. Rudin, A. Allshire, A. Handa, G. State, Isaac gym: High performance gpu-based physics simulation for robot learning,https://arxiv.org/abs/2108.10470(2021). 42

work page internal anchor Pith review Pith/arXiv arXiv 2021

[61] [61]

Proximal Policy Optimization Algorithms

J. Schulman, F. Wolski, P. Dhariwal, A. Radford, O. Klimov, Proximal policy optimization algorithms,https://arxiv.org/abs/1707.06347(2017)

work page internal anchor Pith review Pith/arXiv arXiv 2017

[62] [62]

C. R. Qi, H. Su, K. Mo, L. J. Guibas, Pointnet: Deep learning on point sets for 3d classification and segmentation, inProceedings of the IEEE conference on computer vision and pattern recognition(2017), pp. 652–660

2017

[63] [63]

robot descriptions, awesome-robot-descriptions, GitHub repository (2026),https:// github.com/robot-descriptions/awesome-robot-descriptions, accessed: 2026- 02-23

2026

[64] [64]

D. Yang, Y. Liu, F. Ding, Y. Yu, Bionic Multi-legged Robot Based on End-to-end Artificial Neural Network Control, in2022 IEEE International Conference on Cyborg and Bionic Systems (CBS)(2023), pp. 104–109, doi:10.1109/CBS55922.2023.10115331. Acknowledgments: The authors would like to thank Edwin Ma, Jacob Lee, and Max Li for their help in 3D printing. We ...

work page doi:10.1109/cbs55922.2023.10115331 2023