pith. sign in

arxiv: 2511.14479 · v2 · submitted 2025-11-18 · ⚛️ physics.flu-dyn · math.CV

Metric Geometry Governs Optimal Control in Driven Stokes Flows: Magnetic Driving and Beyond

Pith reviewed 2026-05-17 20:44 UTC · model grok-4.3

classification ⚛️ physics.flu-dyn math.CV
keywords Stokes flowoptimal controlRiemannian metricHele-Shaw cellLorentz forcegeodesicsanisotropic diffusionparticle control
0
0 comments X

The pith

Energy-optimal control paths in driven Stokes flows correspond to geodesics of an emergent Riemannian metric over the fluid domain.

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

The paper examines particle control in a Hele-Shaw cell where Lorentz forces create tunable circulations in a conducting fluid. It establishes that the lowest-energy paths for steering particles match the shortest routes, or geodesics, on a Riemannian metric that arises naturally from the Stokes flow equations. These same paths are also the fastest possible when power is capped at a maximum value. When random forces act at the boundaries, particle motion becomes anisotropic diffusion shaped by the same metric. The geometric picture is presented as applicable beyond the specific magnetic-driving setup to any driven Stokes flow.

Core claim

In the Hele-Shaw cell, Lorentz forces induce circulations that allow control of immersed particles. Energy-optimal control paths are the geodesics of a Riemannian metric that emerges directly from the driven Stokes equations; these geodesics are simultaneously time-optimal under a maximum-power constraint. Particle trajectories subject to random boundary forcing display anisotropic diffusion governed by the same metric. The construction, developed for circulation-driven flows, extends to generic driven Stokes flows and accounts for related observations in three dimensions.

What carries the argument

An emergent Riemannian metric defined over the fluid domain whose geodesics encode the energy-minimizing control paths for particles driven by the Stokes flow.

If this is right

  • Control trajectories can be obtained by solving a geodesic problem on the metric rather than a full optimal-control optimization.
  • Under a hard power limit the same geodesics are the shortest-time paths.
  • Random boundary forcing produces diffusion whose anisotropy is dictated by the metric tensor.
  • The same geometric structure applies to three-dimensional driven Stokes flows.

Where Pith is reading between the lines

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

  • The metric may supply a practical way to plan particle trajectories in microfluidic devices without repeated forward simulations.
  • Similar emergent metrics could be derived for other linear flow regimes, offering a geometric route to optimal actuation in low-Reynolds-number settings.
  • Testing the predicted anisotropic diffusion against experiments with controlled random forcing would directly probe the metric's role in stochastic transport.

Load-bearing premise

A Riemannian metric arises directly from the driven Stokes equations so that optimal control paths are precisely its geodesics without extra adjustments to the flow model.

What would settle it

Compute the true energy-minimizing trajectory by direct numerical optimization of the control problem in the Hele-Shaw cell and compare it to the geodesic curve obtained from the proposed metric; any systematic deviation falsifies the correspondence.

Figures

Figures reproduced from arXiv: 2511.14479 by Kyle McKee.

Figure 1
Figure 1. Figure 1: FIG. 1. Control of an [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2. Metric visualization. (a),(b),(c) For three conductor geometries, log [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3. (A) Four circular conductors are centered on the real axis. Geodesics between a square and four target points (stars) [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
read the original abstract

In a canonical Stokes flow geometry, the Hele-Shaw cell, we show that tunable circulations induced by Lorentz forces in a conducting fluid enable particle control. We reveal that energy-optimal control paths correspond to geodesics of an emergent Riemannian metric defined over the fluid domain, which are time-optimal under a maximum-power constraint. Subject to random boundary forcing, particle paths exhibit metric-governed anisotropic diffusion. Our geometric concepts governing optimal control, though developed explicitly for circulation-driven flows, generalize to generic driven Stokes flows and so elucidate recent observations in a three-dimensional context.

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.

Referee Report

2 major / 2 minor

Summary. The manuscript develops a geometric framework for optimal control of passive particles in driven Stokes flows, focusing on a Hele-Shaw cell where Lorentz forces induce tunable circulations. It claims that energy-optimal control paths correspond exactly to geodesics of an emergent Riemannian metric on the fluid domain; these paths are also time-optimal under a maximum-power constraint. The work further shows that particle trajectories under random boundary forcing exhibit anisotropic diffusion governed by the same metric and asserts that the geometric concepts extend to generic driven Stokes flows, offering insight into recent three-dimensional observations.

Significance. If the claimed equivalence is derived without post-hoc adjustment, the result supplies a parameter-free geometric method for computing optimal steering in linear Stokes systems, directly connecting the dissipation-based control cost to the geodesic equation on the induced metric. This could streamline analysis of particle manipulation in microfluidics and provide a unifying explanation for control observations across flow regimes. The generalization claim, if substantiated with explicit reductions, would be a notable contribution to the intersection of geometric control theory and low-Reynolds-number fluid mechanics.

major comments (2)
  1. [§3 and Eq. (12)] §3 (Metric construction) and the paragraph following Eq. (12): the manuscript must supply an explicit variational derivation showing that the Pontryagin maximum principle applied to the particle advection equation dx/dt = u(x; controls), with quadratic control cost given by the Stokes dissipation, reduces to the geodesic equation on the metric obtained by inverting the Gram matrix of the basis velocity fields. Without this step, the correspondence between energy-optimal paths and geodesics remains at risk of being by construction rather than a direct consequence of the driven Stokes operator.
  2. [§4.1] §4.1 (Generalization statement): the claim that the same metric construction applies to arbitrary driven Stokes flows is load-bearing for the broader significance yet is supported only by a brief analogy; a concrete reduction for at least one additional driving mechanism (e.g., pressure-driven or electrokinetic) is required to substantiate the generalization.
minor comments (2)
  1. [Figure 2] Figure 2 caption: the color scale for the metric tensor components should be labeled with the explicit normalization used (e.g., relative to the uncontrolled dissipation).
  2. [§2 and §3] Notation: the symbol for the control vector field basis is introduced in §2 but reused with a different index convention in §3; a single consistent definition would improve readability.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and constructive comments on our manuscript. We address each major comment below and have revised the manuscript to incorporate the requested clarifications and derivations.

read point-by-point responses
  1. Referee: [§3 and Eq. (12)] §3 (Metric construction) and the paragraph following Eq. (12): the manuscript must supply an explicit variational derivation showing that the Pontryagin maximum principle applied to the particle advection equation dx/dt = u(x; controls), with quadratic control cost given by the Stokes dissipation, reduces to the geodesic equation on the metric obtained by inverting the Gram matrix of the basis velocity fields. Without this step, the correspondence between energy-optimal paths and geodesics remains at risk of being by construction rather than a direct consequence of the driven Stokes operator.

    Authors: We agree that an explicit derivation strengthens the presentation. The original manuscript defined the metric via the dissipation inner product and asserted the geodesic correspondence, but omitted the full PMP steps for brevity. In the revised version we have added a detailed derivation in §3: the control problem minimizes ∫ dissipation dt subject to dx/dt = ∑ c_i(t) u_i(x), the PMP yields a Hamiltonian linear in the controls, and the resulting co-state equations are shown to be equivalent to the geodesic equation on the inverse Gram matrix metric. This confirms the result follows directly from the linear Stokes operator and quadratic cost, rather than being imposed by construction. revision: yes

  2. Referee: [§4.1] §4.1 (Generalization statement): the claim that the same metric construction applies to arbitrary driven Stokes flows is load-bearing for the broader significance yet is supported only by a brief analogy; a concrete reduction for at least one additional driving mechanism (e.g., pressure-driven or electrokinetic) is required to substantiate the generalization.

    Authors: We concur that a concrete example bolsters the generalization claim. While the construction is general for any linear Stokes superposition with quadratic dissipation, we have now added in §4.1 an explicit reduction for pressure-driven Poiseuille flow in a 2D channel. The single parabolic basis field yields a scalar Gram matrix, the induced metric is flat Euclidean, and optimal paths reduce to straight lines, recovering the expected isotropic control. This explicit case illustrates applicability beyond the magnetic Hele-Shaw setting. revision: yes

Circularity Check

0 steps flagged

No significant circularity; metric presented as emergent from Stokes equations

full rationale

The abstract states that energy-optimal paths correspond to geodesics of an emergent Riemannian metric defined over the fluid domain from the driven Stokes equations. No equations or sections are provided that define the metric via the target optimality condition or via a fitted Gram matrix that is then renamed as a prediction. The derivation chain is not shown to reduce to self-definition or self-citation load-bearing; the claim of emergence from the linear Stokes operator stands as an independent step pending full text inspection. This is the expected honest non-finding when no explicit reduction is quotable.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 1 invented entities

The claim rests on the standard Stokes equations plus the existence of an emergent metric whose geodesics coincide with optimal paths; no free parameters or invented entities beyond the metric itself are visible in the abstract.

axioms (1)
  • domain assumption Stokes equations govern the incompressible viscous flow in the Hele-Shaw geometry under Lorentz driving.
    Standard low-Reynolds-number assumption invoked to set up the control problem.
invented entities (1)
  • emergent Riemannian metric on the fluid domain no independent evidence
    purpose: To define the distance whose geodesics are the energy-optimal control paths
    Introduced as arising naturally from the driven flow; no independent falsifiable prediction outside the control result is stated in the abstract.

pith-pipeline@v0.9.0 · 5382 in / 1313 out tokens · 54852 ms · 2026-05-17T20:44:22.249138+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

46 extracted references · 46 canonical work pages

  1. [1]

    author author Howard A \ Stone , author Abraham D \ Stroock , \ and\ author Armand \ Ajdari ,\ title title Engineering flows in small devices: microfluidics toward a lab-on-a-chip , \ @noop journal journal Annu. Rev. Fluid Mech. \ volume 36 ,\ pages 381--411 ( year 2004 ) NoStop

  2. [2]

    author author David J \ Beebe , author Glennys A \ Mensing , \ and\ author Glenn M \ Walker ,\ title title Physics and applications of microfluidics in biology , \ @noop journal journal Annual review of biomedical engineering \ volume 4 ,\ pages 261--286 ( year 2002 ) NoStop

  3. [3]

    author author George M \ Whitesides ,\ title title The origins and the future of microfluidics , \ @noop journal journal nature \ volume 442 ,\ pages 368--373 ( year 2006 ) NoStop

  4. [4]

    author author Bhairab \ Mondal , author N Bhavanashri , author SP Mounika , author Deepika \ Tuteja , author Kunti \ Tandi , \ and\ author H Soniya ,\ title title Microfluidics application for detection of biological warfare agents , \ in\ @noop booktitle Handbook on biological warfare preparedness \ ( publisher Elsevier ,\ year 2020 )\ pp.\ pages 103--131 NoStop

  5. [5]

    author author Manu \ Prakash \ and\ author Neil \ Gershenfeld ,\ title title Microfluidic bubble logic , \ @noop journal journal Science \ volume 315 ,\ pages 832--835 ( year 2007 ) NoStop

  6. [6]

    author author Abraham D \ Stroock , author Stephan KW \ Dertinger , author Armand \ Ajdari , author Igor \ Mezic , author Howard A \ Stone , \ and\ author George M \ Whitesides ,\ title title Chaotic mixer for microchannels , \ @noop journal journal Science \ volume 295 ,\ pages 647--651 ( year 2002 ) NoStop

  7. [7]

    author author Haim H \ Bau , author Jihua \ Zhong , \ and\ author Mingqiang \ Yi ,\ title title A minute magneto hydro dynamic (mhd) mixer , \ @noop journal journal Sensors and Actuators B: Chemical \ volume 79 ,\ pages 207--215 ( year 2001 ) NoStop

  8. [8]

    author author Hassan \ Aref ,\ title title Stirring by chaotic advection , \ 10.1017/S0022112084001233 journal journal Journal of Fluid Mechanics \ volume 143 ,\ pages 1–21 ( year 1984 ) NoStop

  9. [9]

    IEEE MEMS Workshop ,\ Vol

    author author John \ Evans , author Dorian \ Liepmann , \ and\ author Albert P \ Pisano ,\ title title Planar laminar mixer , \ in\ @noop booktitle Proc. IEEE MEMS Workshop ,\ Vol. volume 10 \ ( year 1997 )\ pp.\ pages 96--101 NoStop

  10. [10]

    volume 3 \ ( publisher Cambridge university press ,\ year 1989 ) NoStop

    author author Julio M \ Ottino ,\ @noop title The kinematics of mixing: stretching, chaos, and transport ,\ Vol. volume 3 \ ( publisher Cambridge university press ,\ year 1989 ) NoStop

  11. [11]

    author author Daniel J \ Booth \ and\ author Thomas D \ Montenegro-Johnson ,\ title title Controllable microfluidics through active droplets , \ @noop journal journal Physical Review Fluids \ volume 10 ,\ pages 094203 ( year 2025 ) NoStop

  12. [12]

    author author Tobias M \ Schneider , author Shreyas \ Mandre , \ and\ author Michael P \ Brenner ,\ title title Algorithm for a microfluidic assembly line , \ @noop journal journal Physical review letters \ volume 106 ,\ pages 094503 ( year 2011 ) NoStop

  13. [13]

    author author Lucas \ Amoudruz , author Petr \ Karnakov , \ and\ author Petros \ Koumoutsakos ,\ title title Contactless precision steering of particles in a fluid inside a cube with rotating walls , \ @noop journal journal Journal of Fluid Mechanics \ volume 1014 ,\ pages A15 ( year 2025 ) NoStop

  14. [14]

    author author Ankur \ Kislaya , author Aniket Ashwin \ Samant , author Peter \ Veenstra , author Daniel SW \ Tam , \ and\ author Jerry \ Westerweel ,\ title title Particle manipulation in H ele-- S haw flow with programmable hydrodynamics , \ @noop journal journal Physics of Fluids \ volume 37 ( year 2025 ) NoStop

  15. [15]

    author author A Karimi , author S Yazdi , \ and\ author AM Ardekani ,\ title title Hydrodynamic mechanisms of cell and particle trapping in microfluidics , \ @noop journal journal Biomicrofluidics \ volume 7 ( year 2013 ) NoStop

  16. [16]

    author author Kyle I \ McKee ,\ title title Magnetohydrodynamic flow control in H ele- S haw cells , \ @noop journal journal Journal of Fluid Mechanics \ volume 993 ,\ pages A11 ( year 2024 ) NoStop

  17. [17]

    author author H. S. \ H ele S haw ,\ title title The flow of water , \ @noop journal journal Nature \ volume 58 ,\ pages 34--36 ( year 1898 ) NoStop

  18. [18]

    author author C Pozrikidis ,\ title title The motion of particles in the H ele- S haw cell , \ @noop journal journal Journal of Fluid Mechanics \ volume 261 ,\ pages 199--222 ( year 1994 ) NoStop

  19. [19]

    author author DJ Booth , author IM Griffiths , \ and\ author PD Howell ,\ title title The motion of a bubble in a non-uniform H ele- S haw flow , \ @noop journal journal Proceedings of the Royal Society A \ volume 481 ,\ pages 20240613 ( year 2025 ) NoStop

  20. [20]

    note We consider Hele-Shaw cells with homogeneous electrical properties and cell thickness; cells with heterogeneous properties can support vortical flows mirzadeh2020vortices,boyko2021microscale,ajdari1995electro,ajdari2000pumping . Stop

  21. [21]

    author author Milton \ Van Dyke ,\ @noop title An album of fluid motion \ ( publisher Parabolic Press Stanford ,\ year 1982 ) NoStop

  22. [22]

    author author Kyle I \ McKee \ and\ author John WM \ Bush ,\ title title Potential flows with electromagnetically induced circulation in a H ele- S haw cell , \ @noop journal journal Physical Review Fluids \ volume 10 ,\ pages 054103 ( year 2025 ) NoStop

  23. [23]

    note One interesting setting where potential flow with circulation emerges is in rapidly rotating flow satisfying quasi-geostrophic balance. Away from patches of relative vorticity, the flow is a potential flow with circulation; the circulation magnitude around each patch is prescribed by its net vorticity content pullin1992contour . Stop

  24. [24]

    In canonical geometries, the solution can be written exactly in terms of the Schottky-Klein prime function crowdy2020solving

    note In practice the upper limit of the j -summation is truncated at N_ L terms. In canonical geometries, the solution can be written exactly in terms of the Schottky-Klein prime function crowdy2020solving . Stop

  25. [25]

    note For example, a free stream U C can be incorporated by augmenting W(z) in eq:laurent with the term U z . Stop

  26. [26]

    Noting that the circulation is unity @ _ B_k _k n ds=1 , one finds that _j( B_k)= _k( B_j) or M_ jk =M_ kj

    note Symmetry of M follows from the reciprocity of the Laplace equation, stated as follows, @ _ B_j _k( B_j) _j n ds= @ _ B_k _j( B_k) _k n ds . Noting that the circulation is unity @ _ B_k _k n ds=1 , one finds that _j( B_k)= _k( B_j) or M_ jk =M_ kj . Positive definiteness follows from Appendix Powerderiv . Stop

  27. [27]

    note Re \ W_k'(z) \ =0 for z R and all k \ 1,2, , N-1\ NoStop

  28. [28]

    This phenomena is thus by no means restricted to the case of circular conductors

    note In any geometry wherein all obstacles intersect the real axis, while possessing reflectional symmetry about this axis (i.e., D= D ), _ max is necessarily singular along this line. This phenomena is thus by no means restricted to the case of circular conductors. Stop

  29. [29]

    author author Lev Semenovich \ Pontryagin ,\ @noop title Mathematical theory of optimal processes \ ( publisher Routledge ,\ year 2018 ) NoStop

  30. [30]

    author author Sean M \ Carroll ,\ @noop title Spacetime and geometry \ ( publisher Cambridge University Press ,\ year 2019 ) NoStop

  31. [31]

    Let n denote the normal distance from the boundary and t the tangent distance along it

    note Consider a decomposition of the metric, near a point on the boundary of a conductor, into tangent and normal directions to its surface. Let n denote the normal distance from the boundary and t the tangent distance along it. Since g^ -1 = M^ -1 ^T , impermeability implies that locally the normal component of the metric scales as g^ -1 _ nn n^2/C+ o (n...

  32. [32]

    author author John \ Nash ,\ title title C1 isometric imbeddings , \ @noop journal journal Annals of mathematics \ volume 60 ,\ pages 383--396 ( year 1954 ) NoStop

  33. [33]

    volume 13 \ ( publisher American Mathematical Soc

    author author Qing \ Han , author Jia-Xing \ Hong , \ and\ author Jiaxing \ Hong ,\ @noop title Isometric embedding of Riemannian manifolds in Euclidean spaces ,\ Vol. volume 13 \ ( publisher American Mathematical Soc. ,\ year 2006 ) NoStop

  34. [34]

    o ppel , author Ulrich \ Pinkall , \ and\ author Peter \ Schr \

    author author Albert \ Chern , author Felix \ Kn \"o ppel , author Ulrich \ Pinkall , \ and\ author Peter \ Schr \"o der ,\ title title Shape from metric. \ @noop journal journal ACM Trans. Graph. \ volume 37 ,\ pages 63 ( year 2018 ) NoStop

  35. [35]

    When the distribution has variance ^2 , the electrical power follows a Chi-squared distribution with E (P)= (N-1 ) ^2

    note Sampling M^ 1 2 rather than from a mean-zero Gaussian distribution ensures that equal-power ( P= ^T M ) configurations are explored equally. When the distribution has variance ^2 , the electrical power follows a Chi-squared distribution with E (P)= (N-1 ) ^2 . Stop

  36. [36]

    note Our numerical method utilized the Hamiltonian form of the Euler-Lagrange equations eq:geodesic wherein H= ( p ^Tg^ -1 p )/4 and p =2g x . In this form, we verify convergence of our integrations by monitoring the Hamiltonian conservation along trajectories; for geodesics included in the paper, Hamiltonian fluctuations are less that 1\

  37. [37]

    author author Boyang \ Qin \ and\ author Paulo E \ Arratia ,\ title title Confinement, chaotic transport, and trapping of active swimmers in time-periodic flows , \ @noop journal journal Science Advances \ volume 8 ,\ pages eadd6196 ( year 2022 ) NoStop

  38. [38]

    author author Adam \ Uchytil , author Milan \ Korda , \ and\ author Ji r \' \ Zem \'a nek ,\ title title Data-driven control of a magnetohydrodynamic flow , \ @noop journal journal arXiv preprint arXiv:2507.12479 \ ( year 2025 ) NoStop

  39. [39]

    author author Petr \ Karnakov , author Lucas \ Amoudruz , \ and\ author Petros \ Koumoutsakos ,\ title title Optimal navigation in microfluidics via the optimization of a discrete loss , \ @noop journal journal Physical Review Letters \ volume 134 ,\ pages 044001 ( year 2025 ) NoStop

  40. [40]

    author author Mohammad \ Mirzadeh , author Tingtao \ Zhou , author Mohammad Amin \ Amooie , author Dimitrios \ Fraggedakis , author Todd R \ Ferguson , \ and\ author Martin Z \ Bazant ,\ title title Vortices of electro-osmotic flow in heterogeneous porous media , \ @noop journal journal Physical Review Fluids \ volume 5 ,\ pages 103701 ( year 2020 ) NoStop

  41. [41]

    author author Evgeniy \ Boyko , author Vesna \ Bacheva , author Michael \ Eigenbrod , author Federico \ Paratore , author Amir D \ Gat , author Steffen \ Hardt , \ and\ author Moran \ Bercovici ,\ title title Microscale hydrodynamic cloaking and shielding via electro-osmosis , \ @noop journal journal Physical Review Letters \ volume 126 ,\ pages 184502 ( ...

  42. [42]

    author author Armand \ Ajdari ,\ title title Electro-osmosis on inhomogeneously charged surfaces , \ @noop journal journal Physical Review Letters \ volume 75 ,\ pages 755 ( year 1995 ) NoStop

  43. [43]

    author author Armand \ Ajdari ,\ title title Pumping liquids using asymmetric electrode arrays , \ @noop journal journal Physical review E \ volume 61 ,\ pages R45 ( year 2000 ) NoStop

  44. [44]

    author author DI Pullin ,\ title title Contour dynamics methods , \ @noop journal journal Annual review of fluid mechanics \ volume 24 ,\ pages 89--115 ( year 1992 ) NoStop

  45. [45]

    Crowdy ,\ @noop title Solving problems in multiply connected domains \ ( publisher SIAM ,\ year 2020 ) NoStop

    author author D. Crowdy ,\ @noop title Solving problems in multiply connected domains \ ( publisher SIAM ,\ year 2020 ) NoStop

  46. [46]

    author author Sangtae \ Kim \ and\ author Seppo J \ Karrila ,\ @noop title Microhydrodynamics: principles and selected applications \ ( publisher Butterworth-Heinemann ,\ year 2013 ) NoStop