Pith. sign in

REVIEW 2 major objections 3 minor 1 cited by

Universal Reconfiguration of Facet-Connected Modular Robots by Pivots: The $O(1)$ Musketeers

T0 review · 2 major / 3 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read The paper proves that five extra helper modules suffice to reconfigure any facet-connected square-grid robot into any other shape of the same size using pivot moves, with an O(n^2)-move algorithm that is worst-case optimal.

desk verdict Genuinely new and important result; the five-musketeer bound has a real unproved step (d≤6 in Lemma 13) that a referee should pin down before acceptance. read the letter →

arxiv 1908.07880 v2 pith:WGJFU23H submitted 2019-08-21 cs.CG cs.RO

Hugo A. Akitaya , Esther M. Arkin , Mirela Damian , Erik D. Demaine , Vida Dujmović , Robin Flatland , Matias Korman , Belén Palop
show 3 more authors
Irene Parada André van Renssen Vera Sacristán
This is my paper · ORCID
classification cs.CGcs.RO
keywords modularrobotspivotmovesfacet-connectedconfigurationssquaregridreconfigurationgraphmonkeymusketeermodulesuniversal
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

This paper tries to establish that five extra "helper" modules are enough to make any facet-connected square-grid modular robot transform into any other shape of the same size using pivot moves, and that this can be done in O($n^{2}$) moves, the best possible worst-case bound. The authors show that without helpers the space of configurations is badly disconnected, with exponentially many connected components, some of them rigid traps, so some outside assistance is genuinely needed. Their algorithm moves modules one at a time around the outer boundary of the robot into a canonical strip, using helpers only when a module is a cut vertex whose removal would split the robot. If correct, this gives the first universal reconfiguration guarantee for pivot-based facet-connected robots and explains why a small constant number of spare modules solves a problem that local shape conditions cannot.

What carries the argument

The device that carries the argument is the outer-shell traversal: a module placed just outside the robot's boundary can pivot clockwise around the entire outer shell using only straight monkey jumps, following the right-hand rule, and return to its start (Lemma 10). Progress is measured by the lexicographic potential $\Phi(m)=(x+y,x)$: the algorithm repeatedly removes the maximum-potential module into a canonical strip. When that module is a cut vertex joining two components, the bridging procedure slides a small window $S$ along the boundary until it touches both components, then places up to five helper modules to reconnect them; the case analysis is organized by the distance $d$ between the closest green and blue modules adjacent to the bridge rectangle.

What would settle it

Simulate Algorithm 1 on small facet-connected configurations, say up to 20 squares, and for each maximum-potential module compute the grid distance between the closest pair of modules from the two components cut by that module; finding any configuration with distance 7 or greater, or any bridge needing a sixth helper, would refute the claim that five musketeers always suffice.

Watch

Extended reading notes

Core claim

The central claim is Theorem 20: Algorithm 1 turns any facet-connected configuration of n pivot modules in the square grid into a straight strip of n modules, and by reversing the move sequence any strip into any target shape, using O($n^{2}$) monkey pivot moves and at most five added helper modules; the O($n^{2}$) count is worst-case optimal. The route through the strip means the result is equivalent to proving that the reconfiguration graph of n modules plus up to five free helpers is connected. On the way, the paper proves that with no helpers the graph is far from connected, with an exponential number of components, some of exponential size and some isolated rigid configurations. It also defines three nested families of pivot moves, restrictive, leapfrog, and monkey, and shows the resulting reconfiguration graphs are strictly different.

Load-bearing premise

The load-bearing premise is the assertion in the bridging lemma that the grid distance between the closest pair of modules from the two separated pieces can only be 2 through 6; if the distance could be 7 or more, the five-helper case analysis would not cover it and five helpers might not suffice.

Editorial extensions

If this is right

  • Any two facet-connected configurations of the same number of modules become mutually reachable once five spare modules are allowed, because both can be reduced to the same canonical strip and the moves reversed.
  • The move count O(n^2) is asymptotically optimal: some pairs of configurations, such as a vertical strip and a horizontal strip, force each module to travel distance $\Theta(n)$, giving an $\Omega(n^2)$ lower bound.
  • The unassisted reconfiguration graph's exponential fragmentation shows why previous forbidden-pattern restrictions were not just an artifact of proof technique; without helpers, universal reconfiguration is impossible.
  • The algorithm only needs straight monkey jumps, not diagonal ones, so the guarantee covers pivot hardware that can perform the simpler straight jump.
  • The three move models are genuinely different: some configurations are reachable under monkey moves but not leapfrog, and some under leapfrog but not restrictive.

Reading between the lines

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

  • I infer that if a counterexample to the assertion that the bridging distance $d$ is at most 6 exists, the framework would not collapse: it would merely push the helper constant upward, and the same outer-shell strategy would likely give universality with a larger fixed number of helpers.
  • I infer that the outer-shell and bridging ideas are portable to cube lattices in three dimensions, where the bridge window would need more helpers; the paper itself lists this as an open direction, and I would expect the constant to grow with dimension.
  • I infer that because the algorithm is explicit, a simulator could test the five-helper promise directly on small random configurations; the bridged cases $d=2$ through $d=6$ are all drawn, so one can check whether the figures cover every local pattern that actually occurs.
  • I infer that the musketeer idea suggests a general principle for lattice reconfiguration: a constant number of freely placeable agents can bypass deadlocks even when the unassisted state space is exponentially disconnected, which may apply to other lattices and move sets.
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

2 major / 3 minor

Summary. The paper studies reconfiguration of facet-connected square-grid modular robots under pivot moves. It introduces three nested move sets (restrictive, leapfrog, and monkey) and proves that they give strictly different reconfiguration graphs. It shows that configurations with a certain forbidden pattern can have exponentially many connected components, including exponentially many singleton (rigid) components in some models. The main algorithmic result, Theorem 20, claims that any two facet-connected configurations of n modules can be reconfigured into each other using at most five additional 'musketeer' modules and O(n^2) monkey-move pivoting steps, which is worst-case optimal. The algorithm repeatedly moves outer-free modules along the external boundary to a canonical strip; when no outer-free module exists, a bridging procedure uses up to five musketeer modules to break the cut vertex and make progress. The proof of the constant five rests on a case analysis over a distance parameter d defined in the bridging procedure (Lemma 13).

Significance. If correct, the main theorem is a substantial advance: it removes the forbidden-pattern admissibility restriction from prior pivot-reconfiguration results and gives a precise constant number of helper modules, together with a matching worst-case move bound. The model separations and the exponential connected-component results are also valuable contributions. The paper is self-contained, the algorithm is concrete, and the potential-function framework is a clear structure for proving termination and monotonicity. However, the proof of the constant five rests on an unproved bound on the parameter d in Lemma 13, so the central claim is not yet fully established.

major comments (2)
  1. [Section 4.3.2, Lemma 13] The proof of Lemma 13 states without argument that 'It is easy to see that d can only be 2, 3, 4, 5, or 6.' This is a load-bearing claim because the subsequent case analysis in Figures 19–23 covers only these values, and the maximum number of musketeer modules needed in the bridging procedure is determined by the largest d considered. The rectangle R is the union of two 3x3 squares, so modules facet-adjacent to R on opposite sides can in principle be at L1 distance larger than 6. The precondition that the configuration has no outer-free modules may rule this out, but no proof is supplied. If d could be 7 or larger, the bridging procedure would require more than five musketeer modules in at least one case, contradicting Theorem 20. The authors must either prove d <= 6 or modify the argument.
  2. [Section 4.3.2, bridging case analysis (Figures 19–23)] The case analyses for d = 2 through 6 are presented almost entirely through figures, and the text asserts for each case that the indicated 'must' occupancies follow from the absence of outer-free modules and that the displayed musketeer sequences are feasible. The completeness of the enumeration—up to rotations and reflections, and the listed options—is not argued formally. Because the correctness and the helper bound of Lemma 13 depend on this enumeration being exhaustive, the proof needs either a formal symmetry argument or a more explicit derivation of each case. Without this, the reader cannot verify that all configurations with the stated invariants are covered.
minor comments (3)
  1. [Figure 13 caption] The caption says 'If a1 or a2 are occupied, then both b1 and b2 are empty', but the proof text defines the relevant positions as a0, a1, b0, and b1. The caption labels should be updated to match the proof.
  2. [Theorem 20 statement] The phrase 'into a canonical strip of the same size, adding at most five extra modules' is ambiguous: the final strip contains n original modules plus up to five helpers, so the total length is n + h with h <= 5. This should be stated precisely.
  3. [Section 4.3.2, Observation 12] The observation that the outer shell has exactly two green-blue changes of color, one at m, is used to justify the traversal of the square S, but its proof is not given. A short justification would improve readability.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the central derivation is self-contained; the unproved d<=6 bound in Lemma 13 is a proof gap, not a circular reduction.

full rationale

Walking the derivation chain: the paper defines three pivot-move models and proves separations between them using explicit configurations and connectivity arguments; the reconfiguration graph results in Section 3 are proved directly from the definitions and external prior work. Section 4 introduces the outer shell via Lemma 10, whose right-hand-rule traversal proof is carried out case-by-case in the text. The bridging procedure in Section 4.3.2 enumerates cases d=2 through d=6 with explicit musketeer placements and proves connectivity and potential arguments for each. Lemmas 14-19 handle the remaining neighborhood configurations, and Theorem 20 combines these into the O(n^2) bound with the earth-moving lower bound. No fitted parameter is later called a prediction; no input definition is equivalent to the output theorem; and no load-bearing claim is justified solely by a self-citation. The references to prior pivot-move work [20], the idea of helper modules [12], and the cactus graph [7] are all external to the present author set, so self-citation is not an issue. The only notable weakness is the sentence in Lemma 13, 'It is easy to see that d can only be 2, 3, 4, 5, or 6,' which asserts the key upper bound on the green-blue distance without proof. That is a genuine missing argument and a correctness risk for the five-musketeer sufficiency claim, but it is not circularity: the bound is not assumed as an input, derived from a fit, or obtained by renaming a prior result. A proof gap is different from a self-referential reduction, so the circularity score remains 0.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

The central claim rests on the model definitions (pivot sets, facet-connectivity), the availability of up to five extra modules, and the geometric fact that an active module can traverse the outer shell. No free parameters are fitted, and no new physical entities are introduced.

assumptions (3)
  • domain assumption Module motion is restricted to pivot moves on a square lattice with facet-connectivity maintenance.
    Section 2.1 defines the three sets of moves and the connectivity requirement; this is the model under which the claims are made.
  • domain assumption Up to five extra modules (musketeers) may be added to the configuration and are not part of the start or goal shapes.
    Section 4.3.1 introduces musketeer modules; the algorithm adds them when the strip is shorter than five modules.
  • standard math The active module's right-hand-rule traversal of the boundary is well-defined on the square lattice outer shell.
    Used in Lemma 10; relies on standard lattice geometry facts about the external boundary of a polyomino.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Universal Reconfiguration of Facet-Connected Modular Robots by Pivots: The $O(1)$ Musketeers." pith.science (2026). https://pith.science/paper/WGJFU23H

@misc{pith2026190807880,
  author       = {Pith},
  title        = {Pith review of: Universal Reconfiguration of Facet-Connected Modular Robots by Pivots: The $O(1)$ Musketeers},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WGJFU23H}},
  note         = {Machine review of arXiv:1908.07880}
}
abstract

We present the first universal reconfiguration algorithm for transforming a modular robot between any two facet-connected square-grid configurations using pivot moves. More precisely, we show that five extra "helper" modules ("musketeers") suffice to reconfigure the remaining $n$ modules between any two given configurations. Our algorithm uses $O(n^2)$ pivot moves, which is worst-case optimal. Previous reconfiguration algorithms either require less restrictive "sliding" moves, do not preserve facet-connectivity, or for the setting we consider, could only handle a small subset of configurations defined by a local forbidden pattern. Configurations with the forbidden pattern do have disconnected reconfiguration graphs (discrete configuration spaces), and indeed we show that they can have an exponential number of connected components. But forbidding the local pattern throughout the configuration is far from necessary, as we show that just a constant number of added modules (placed to be freely reconfigurable) suffice for universal reconfigurability. We also classify three different models of natural pivot moves that preserve facet-connectivity, and show separations between these models.

Figures

Figures reproduced from arXiv: 1908.07880 by the authors.

Figure 1
Figure 1. Two ways a module a starting above module s can move to the adjacent lattice position, above module s 0 . Left: sliding. Right: pivoting. Pivoting requires more free space to execute. destination square/cube to be empty. Unfortunately, some configurations are rigid in this model, meaning that no module can move without disconnecting the robot. Rigid configurations appear also in the sliding square model when the sli… view at source ↗
Figure 2
Figure 2. The possible sets of moves for a pivoting module [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Left: a rigid configuration of edge-connected pivoting squares. Right: A configuration [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (29 more)
Figure 4
Figure 4. Figure 4: The three forbidden patterns for facet-connected pivoting squares; solid squares [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: Two shapes that can be reconfigured into each other using the leapfrog set of moves, [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: A shape that can be reconfigured into a strip after moving modules [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 7
Figure 7. Figure 7: Configurations showing only the forbidden pattern Γ. Top: Rigid configuration for [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 8
Figure 8. Figure 8: A configuration showing only pattern I (corridor bottleneck). It is locked for both the [PITH_FULL_IMAGE:figures/full_fig_p009_8.png]
Figure 9
Figure 9. Figure 9: The adjacency graph G of a facet-connected configuration (left) and the corresponding cactus graph T(G) (right); leaves and maximal cycles are dark-shaded [PITH_FULL_IMAGE:figures/full_fig_p009_9.png]
Figure 10
Figure 10. Figure 10: If only pattern I is allowed some corner can pivot using monkey-pivoting moves. [PITH_FULL_IMAGE:figures/full_fig_p010_10.png]
Figure 11
Figure 11. Figure 11: If only pattern Z is allowed, any corner can pivot regardless of the set of pivoting [PITH_FULL_IMAGE:figures/full_fig_p011_11.png]
Figure 12
Figure 12. Figure 12: Three configurations with only instances of pattern Z. They cannot be reconfigured [PITH_FULL_IMAGE:figures/full_fig_p011_12.png]
Figure 13
Figure 13. Figure 13: Illustration of the invariant. The active module [PITH_FULL_IMAGE:figures/full_fig_p012_13.png]
Figure 14
Figure 14. Figure 14: When b0 is occupied, a0 and a1 are empty. The move, when b1 or c2 are occupied. Striped cells are empty, filled cells are occupied. If this cell is empty, m can move there and after pivoting the invariant is satisfied (in a rotated version). Finally, if b2 and a2 are …
Figure 15
Figure 15. Figure 15: When b0 is occupied, a0 and a1 are empty. The move, when b1 or c2 are both empty. Striped cells are empty, filled cells are occupied. Now, consider the case where b0 is empty, illustrated in [PITH_FULL_IMAGE:figures/full_fig_p013_15.png]
Figure 16
Figure 16. Figure 16: The move, when b0 is empty. Striped cells are empty, filled cells are occupied. equivalent to the case where b0 is occupied. This is made visibly clear after rotating the entire configuration 90◦ . Since the static configuration C has a finite number of modules and is…
Figure 17
Figure 17. Figure 17: A robot configuration (in gray) and its associated outer shell (striped in pink). [PITH_FULL_IMAGE:figures/full_fig_p014_17.png]
Figure 18
Figure 18. Figure 18: Top: 3 × 3 square S in its initial position s0. The outer thick line indicates the path traversed by the center of S. Dots correspond to the center positions where S is adjacent to a boundary edge. Bottom: the rectangular union R of S centered at sk and at sk−1. with …
Figure 19
Figure 19. Figure 19: Bridging when d = 2. Striped cells are empty, filled cells are occupied. Green and blue indicate different connected components of C \ {m}. Dotted pairs of cells of the same color mean that a module must exist in at least one of the two cells. The orange cells labelle…
Figure 20
Figure 20. Figure 20: Bridging when d = 3. Striped cells are empty, filled cells are occupied. Green and blue indicate different connected components of C \ {m}. The dotted pair of cells indicates that a module must exist in at least one of the two cells. The orange cells labelled ai corre…
Figure 21
Figure 21. Figure 21: Bridging when d = 4. Striped cells are empty, filled cells are occupied. Green and blue indicate different connected components of C \ {m}. The dotted pair of cells indicates that a module must exist in at least one of the two cells. The orange cells labelled ai corre…
Figure 22
Figure 22. Figure 22: Bridging when d = 5. Striped cells are empty, filled cells are occupied. Green and blue indicate different connected components of C \ {m}. The dotted pairs of cells indicate that a module must exist in at least one of the two cells. The orange cells labelled ai corre…
Figure 23
Figure 23. Figure 23: Bridging when d = 6. Striped cells are empty, filled cells are occupied. Green and blue indicate different connected components of C \ {m}. The orange cells labelled ai correspond to the positions of the bridging musketeer modules. Their subindexes indicate the order …
Figure 24
Figure 24. Figure 24: A rigid configuration that requires the addition of five musketeer modules for bridging. [PITH_FULL_IMAGE:figures/full_fig_p020_24.png]
Figure 25
Figure 25. Figure 25: (a) If g 0 exists, it is outer-free; if b1 is of degree 3, b 0 is outer-free (b.1) b1 has degree 1 (b.2) b1 has a North neighbor (b.3) b1 has a West neighbor. Since the positions West and East of g1 are empty, g1 has maximum degree 2. Assume first that g1 has degree 1…
Figure 26
Figure 26. Figure 26: Reconfiguring when b1 has degree 1. outer shell of C and join the canonical strip; see [PITH_FULL_IMAGE:figures/full_fig_p022_26.png]
Figure 27
Figure 27. Figure 27: Reconfiguring when g2 and b2 have degree 2. Lemma 16. This reconfiguration step uses O(n) pivoting operations to transform C into a facet-connected configuration of smaller size and smaller potential gap ∆Φ. Proof. In addition to the O(n) pivoting operations used by m…
Figure 28
Figure 28. Figure 28: Notice that b2 cannot have a North neighbor since it would have a higher potential b1 m b2 b2 b1 m b1 m b2 b2 b1 m b2 b1 m m g1 g2 m g1 g2 m g1 g2 [PITH_FULL_IMAGE:figures/full_fig_p023_28.png]
Figure 29
Figure 29. Figure 29: Reconfiguring when g2 has degree greater than 2 (a) g2 does not have a West neighbor (b) g2 has a West neighbor and b2 lies North of b1. If the position b 0 is occupied, then b1 is blocked; see [PITH_FULL_IMAGE:figures/full_fig_p024_29.png]
Figure 30
Figure 30. Figure 30: Reconfiguring when g2 has degree greater than 2, g2 has a West neighbor and b2 lies West of b1. 24 [PITH_FULL_IMAGE:figures/full_fig_p024_30.png]
Figure 31
Figure 31. Figure 31: Reconfiguring when b2 has degree greater than 2 and a South neighbor. Assume first that b2 has a South neighbor; see [PITH_FULL_IMAGE:figures/full_fig_p025_31.png]
Figure 32
Figure 32. Figure 32: Reconfiguring when b2 has degree greater than 2 but no South neighbor. We use a similar sequence of pivoting steps for the case when b 0 1 has a North neighbor b 00; see [PITH_FULL_IMAGE:figures/full_fig_p026_32.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. Stress-Sharing for Decentralized Fault Repair in Modular Spacecraft

    cs.RO 2026-07 conditional novelty 5.0 of 10

    A local-information, connectivity-safe pivot policy restores most connectivity in simulated modular spacecraft after random damage, although full reconnection collapses with size and fault density.

Reference graph

Works this paper leans on

23 extracted references · 23 canonical work pages · cited by 1 Pith paper

  1. [20]

    C. Sung, J. Bern, J. Romanishin, and D. Rus. Reconfiguration planning for pivoting cube modular robots. In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), pages 1933–1940, 2015

    work page 1933

  2. [1]

    A Universal In-Place Reconfiguration Algorithm for Sliding Cube-Shaped Robots in a Quadratic Number of Moves

    Z. Abel and S. D. Kominers. Pushing hypercubes around. CoRR, abs/0802.3414, 2008. arXiv:0802.3414

    work page Pith review arXiv 2008

  3. [2]

    B. K. An. EM-Cube: cube-shaped, self-reconfigurable robots sliding on structure surfaces. In Proc. IEEE International Conference on Robotics and Automation (ICRA), pages 3149–3155, 2008

    work page 2008

  4. [3]

    Ayanian, P

    N. Ayanian, P. J. White, ´A. H´ al´ asz, M. Yim, and V. Kumar. Stochastic control for self- assembly of XBots. In Proceedings of the ASME International Design Engineering Technical Conferences and Computers and Information in Engineering Conference , 2008

    work page 2008

  5. [4]

    N. M. Benbernou. Geometric algorithms for reconfigurable structures . PhD thesis, Mas- sachusetts Institute of Technology, 2011

    work page 2011

  6. [5]

    Chennareddy, A

    S. Chennareddy, A. Agrawal, and A. Karuppiah. Modular self-reconfigurable robotic systems: a survey on hardware architectures. Journal of Robotics, 2017(5013532), 2017

    work page 2017

  7. [6]

    G. S. Chirikjian. Kinematics of a metamorphic robotic system. In Proc. IEEE International Conference on Robotics and Automation (ICRA) , volume 1, pages 449–455, 1994

    work page 1994

  8. [7]

    Dumitrescu and J

    A. Dumitrescu and J. Pach. Pushing squares around. Graphs and Combinatorics , 22(1):37– 50, 2006

    work page 2006

Show all 23 references
  1. [8]

    Fitch, Z

    R. Fitch, Z. Butler, and D. Rus. Reconfiguration planning for heterogeneous self- reconfiguring robots. In Proc. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) , volume 3, pages 2460–2467, 2003

    2003

  2. [9]

    Hemmerling

    A. Hemmerling. Labyrinth problems – labyrinth-searching abilities of automata , volume 14 of Teubner-Texte zur Mathematik (TTZM) . Springer-Verlag, 1989

    1989

  3. [10]

    Kurokawa, S

    H. Kurokawa, S. Murata, E. Yoshida, K. Tomita, and S. Kokaji. A 3-D self-reconfigurable structure and experiments. In Proc. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) , volume 2, pages 860–865, 1998

    1998

  4. [11]

    Larkworthy and S

    T. Larkworthy and S. Ramamoorthy. A characterization of the reconfiguration space of self-reconfiguring robotic systems. Robotica, 29(1):73–85, 2011

    2011

  5. [12]

    Michail, G

    O. Michail, G. Skretas, and P. G. Spirakis. On the transformation capability of feasible mechanisms for programmable matter. J. Comput. Syst. Sci. , 102:18–39, 2019

    2019

  6. [13]

    Murata, H

    S. Murata, H. Kurokawa, and S. Kokaji. Self-assembling machine. In Proc. IEEE In- ternational Conference on Robotics and Automation (ICRA) , volume 1, pages 441–448, 1994

    1994

  7. [14]

    Murata, E

    S. Murata, E. Yoshida, A. Kamimura, H. Kurokawa, K. Tomita, and S. Kokaji. M-TRAN: self-reconfigurable modular robotic system. IEEE/ASME Transactions on Mechatronics, 7(4):431–441, 2002. 28

    2002

  8. [15]

    Nguyen, L

    A. Nguyen, L. J. Guibas, and M. Yim. Controlled module density helps reconfiguration planning. In Algorithmic and Computational Robotics: New Dimensions (WAFR) , pages 23–25. A. K. Peters, 2001

    2001

  9. [16]

    E. H. Østergaard, K. Kassow, R. Beck, and H. H. Lund. Design of the ATRON lattice-based self-reconfigurable robot. Autonomous Robots, 21(2):165–183, 2006

    2006

  10. [17]

    Rus and M

    D. Rus and M. Vona. A physical implementation of the self-reconfiguring crystalline robot. In Proc. IEEE International Conference on Robotics and Automation (ICRA) , volume 2, pages 1726–1733, 2000

    2000

  11. [18]

    Salemi, M

    B. Salemi, M. Moll, and W.-M. Shen. SUPERBOT: a deployable, multi-functional, and modular self-reconfigurable robotic system. In Proc. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) , pages 3636–3641, 2006

    2006

  12. [19]

    K. Stoy, D. Brandt, and D. J. Christensen. Self-reconfigurable robots: an introduction. MIT Press, 2010

    2010

  13. [21]

    Unsal, H

    C. Unsal, H. Kiliccote, and P. Khosla. I(CES)-Cubes: a modular self-reconfigurable bipartite robotic system. In Proc. SPIE Conference on Mobile Robots and Autonomous Systems , volume 3839, pages 258–269. SPIE, 1999

    1999

  14. [22]

    M. Yim, W. Shen, B. Salemi, D. Rus, M. Moll, H. Lipson, E. Klavins, and G. S. Chirikjian. Modular self-reconfigurable robot systems. IEEE Robotics & Automation Magazine, 14(1):43– 52, 2007

    2007

  15. [23]

    Zykov, A

    V. Zykov, A. Chan, and H. Lipson. Molecubes: an open-source modular robotic kit. In IROS-2007 Self-Reconfigurable Robotics Workshop, 2007. 29

    2007

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.