An active Lorentz gas: walking droplets in periodic media

arXiv:2609.02419 · nlin.CD · Submitted 2026-09-02 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Nonlinear Sciences. The day in nonlinear science: chaos, patterns, networks and adaptation.

Kai: We're looking at the paper: "An active Lorentz gas: walking droplets in periodic media".

Mira: Here's the gist of the paper. The active Lorentz gas is introduced by modeling a walking droplet as an inertial active particle moving through a square soft Lorentz gas because this minimal framework enables a systematic investigation of how activity and dissipation modify the classical conservative Lorentz gas paradigm.

Kai: The single most informative sentence about the paper's subject and finding is that the active Lorentz gas is a rich framework for studying transport in dissipative dynamical systems and provides a bridge between active matter and classical models of chaotic transport.

Mira: Model Description The model starts by presenting a widely used theoretical model for the wave–particle entity (WPE) that materializes as a droplet bouncing and walking on a vibrated bath of the same liquid.

Kai: The equation of motion for the WPE is given by mx¨d + Dt x˙ d = −mg∇h(xd, t) − ∇V˜PP(xd).

Paper summary: Kai: So looking at the title "An active Lorentz gas: walking droplets in periodic media" and what we just talked about, it seems like this paper is establishing a core model for how activity influences transport in these structured environments.

Mira: It's setting up that framework to study transport in dissipative dynamical systems and connecting those classical chaotic transport models with the realm of active matter.

Lev: From an engineering standpoint, what they've done is define the basic mathematical description of this walking droplet inside a square soft Lorentz gas so you have something concrete to try and simulate on.

Kai: The paper shows that by including activity and dissipation, you can systematically investigate how those forces modify the classical conservative Lorentz gas paradigm.

Mira: And the big implication for us is that we've got a tool now for looking at how activity and dissipation interact when modeling these kinds of systems.

Lev: It gives us a starting point to see what happens when we push into those more complex, non-conservative regimes that you mentioned later.

Kai: So ultimately, this work lays the groundwork for exploring hydrodynamic quantum analogs in periodic media, which is where it points next.

Mira: It really shows how these classical ideas can be extended into looking at things that might have quantum-like features when you consider the coupling between the active particle and its medium.

Conclusion: Kai: So we're wrapping up this look at the paper "An active Lorentz gas: walking droplets in periodic media." Basically, they've taken a walking droplet bouncing around inside this soft, vibrated gas and modeled it as an inertial active particle moving through a square environment.

Mira: Right. And the main thing they want us to see is how that activity and the dissipation—the friction—change the standard picture of how particles move in those classical Lorentz gas models.

Kai: Yeah, it's about bridging that gap between active matter, which we know is super messy and non-equilibrium, and these more traditional models of chaotic transport.

Lev: From where I sit building the actual hardware for these kinds of simulations, what this means is they've given us a blueprint for how to set up those complex dynamics.

Mira: Exactly. They’re showing that this active Lorentz gas framework is a rich setup for studying transport in systems where things are losing energy, and it connects active matter directly to classical chaotic transport ideas.

Kai: It seems like the authors really want us to see how this specific model provides a way to look at things that are usually too complicated because of both the activity and the dissipation working together.

Mira: And they're pointing toward using this as a stepping stone for looking at hydrodynamic quantum analogs in periodic media, which is where their ideas get really interesting.

Computational Physics Laboratory, Tampere University · Rudolf Peierls Centre for Theoretical Physics, University of Oxford · Centre for Complex Systems, School of Mathematical Sciences, Queen Mary University of London · London Mathematical Laboratory

nlin.CD

Submitted: 2026-09-02

Updated: 2026-10-19

Comments: 26 pages, 12 figures, supplementary video

Journal ref: New J. Phys. 28 114401 (2026)

DOI: 10.1088/1367-2630/aebdbe

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 94/100

The gist: The active Lorentz gas is introduced by modeling a walking droplet as an inertial active particle moving through a square soft Lorentz gas because this minimal framework enables a systematic…

Key concepts

Active Lorentz Gas
This is a theoretical model where an active particle, like a walking droplet, moves within a soft periodic medium (the Lorentz gas). The activity introduces non-conservative forces that modify the standard classical physics of how particles move and interact in such environments.
Mean Squared Displacement (MSD)
MSD measures how far a particle has moved from its starting point over time. In this paper, it is used to characterize transport behavior; if MSD scales as $t^eta$, the exponent $eta$ tells us if the movement is normal diffusion or anomalous diffusion, indicating unusual transport patterns.
Phase Space Contraction
Phase space contraction refers to the overall shrinking of the volume occupied by a system's possible states over time. The analysis shows that introducing non-conservative forces (activity/dissipation) causes this volume to contract, suggesting the presence of attracting dynamics in the system.
Bifurcation Diagrams
These are plots used to visualize how the system's behavior changes as a parameter is varied. In this context, they reveal complex structures showing how different types of trajectories (like periodic or chaotic) appear and disappear as activity levels change.

Terminology

Summary

The active Lorentz gas is introduced by modeling a walking droplet as an inertial active particle moving through a square soft Lorentz gas because this minimal framework enables a systematic investigation of how activity and dissipation modify the classical conservative Lorentz gas paradigm. p. 1

Where the paper says it

In this work, we investigate the dynamics of a walking droplet in a square soft Lorentz gas using a reduced droplet model derived from the low-memory limit. The resulting deterministic dynamical system describes an active particle subject to a velocity-dependent dissipative force and a periodic external potential.

The single most informative sentence about the paper's subject and finding is that the active Lorentz gas is a rich framework for studying transport in dissipative dynamical systems and provides a bridge between active matter and classical models of chaotic transport.

Model Description

The model starts by presenting a widely used theoretical model for the wave–particle entity (WPE) that materializes as a droplet bouncing and walking on a vibrated bath of the same liquid. The equation of motion for the WPE is given by mx¨d + Dt x˙ d = −mg∇h(xd, t) − ∇V˜PP(xd). p. 2

Where the paper says it

We consider a WPE at position xd = (xd, yd) and moving with velocity x˙ d in an external periodic potential V˜ PP(x) whilst generating axisymmetric waves with a spatial profile defined by W(|x|). The self-interaction with the waves produces memory effects, which lead to the following equation of motion for the WPE [37] mx¨d + Dtx˙ d = −mg∇h(xd, t) − ∇V˜ PP(xd), (1) where the overdot notation is used for time derivatives, m denotes the mass of the particle, and Dt is an effective time-averaged drag coefficient.

In the low-memory limit, the self-force Fself on the particle is approximated as Fself ≈ β e − 1/Me f (x˙ d) / x˙ d x˙ d. p. 2
Where the paper says it

To obtain a tractable model of the WPE, we consider the low-memory limit, where the particle is only influenced by its most recent impact at time t − TF, or t − 1 in dimensionless coordinates. This enables us to approximate xd(t)−xd(t−1) ≈ x˙ d(t), simplifying the self-force Fself on the particle from its self-generated waves to Fself ≈ β e − 1 Me f (|x˙ d(t)|) x˙ d(t) |x˙ d(t)| , resulting in the simplified equation of motion x¨d + µx˙ d = γf (|x˙ d|) x˙ d |x˙ d| + Fext (5) 3

This results in the simplified equation of motion x¨d + µx˙ d = γf (x˙ d) / x˙ d x˙ d + Fext. p. 2
Where the paper says it

To obtain a tractable model of the WPE, we consider the low-memory limit, where the particle is only influenced by its most recent impact at time t − TF, or t − 1 in dimensionless coordinates. This enables us to approximate xd(t)−xd(t−1) ≈ x˙ d(t), simplifying the self-force Fself on the particle from its self-generated waves to Fself ≈ β e − 1 Me f (|x˙ d(t)|) x˙ d(t) |x˙ d(t)| , resulting in the simplified equation of motion x¨d + µx˙ d = γf (|x˙ d|) x˙ d |x˙ d| + Fext (5) 3

Phase Space Analysis

The authors introduce an energyvariance filtering method to distinguish between localized periodic, quasi-ballistic periodic, and chaotic trajectories. This method is used to construct bifurcation diagrams in a non-conservative setting. p. 0

Where the paper says it

We find that islands of stability collapse into attracting and repelling sets. To characterize these structures, we introduce an energyvariance filtering method that distinguishes localized periodic, quasi-ballistic periodic, and chaotic trajectories, enabling the construction of bifurcation diagrams in a non-conservative setting.

The analysis of the average phase space contraction rate kpt shows that it remains negative, indicating an overall contraction of phase-space volume and the presence of attracting dynamics. p. 6
Where the paper says it

2 for σ = 0. Throughout the examined range, kpt remains negative, indicating an overall contraction of phase-space volume and, consequently, the presence of attracting dynamics.

The qualitative effects on KAM islands are illustrated by showing that they collapse into repelling and attracting sets under the introduction of the non-conservative force FNC. p. 7
Where the paper says it

This strengthening may be related to the decreasing density of scatterers with increasing w, which allows the walking droplet to explore the lattice more freely and results in less constrained dynamics as the system evolves toward the attracting set. The qualitative effects of the dissipative force on the KAM islands of the conservative square soft Lorentz gas are illustrated in Fig.

Anomalous Transport Characterization

The system's transport behavior is characterized by the mean squared displacement (MSD), which scales as MSD ∼ t α, where α is the diffusion exponent. The authors investigate regimes exhibiting strong bifurcation cascades, anomalous diffusion, and significant phase-space contraction. p. 0

Where the paper says it

To characterize these structures, we introduce an energyvariance filtering method that distinguishes localized periodic, quasi-ballistic periodic, and chaotic trajectories, enabling the construction of bifurcation diagrams in a non-conservative setting. We identify parameter regimes exhibiting strong bifurcation cascades, anomalous diffusion, and significant phase-space contraction.

The energy variance of PSP crossings is used to obtain further signs of anomalous transport. p. 6
Where the paper says it

We further investigate the energy variance of PSP crossings as a function of system parameters in Sec. 4, obtaining further signs of anomalous transport.

Bifurcation and Diffusion

The bifurcation diagrams reveal a rich variety of bifurcation structures across the investigated range of w, with thin branches associated with attracting sets. The diffusion coefficient D exhibits an overall increase with w, superimposed by complex variations on finer parameter scales. p. 6

Where the paper says it

Conservative soft-wall Lorentz gases are known to exhibit KAM islands [106, 107], leading to the emergence of localized and quasi-ballistic periodic orbits, fundamentally changing the properties of the system through anomalous diffusion [70, 71, 81]. The diffusion coefficient of a softwall Lorentz gas exhibits a high degree of irregularity as a function of its control parameters, and in some parameter regions is completely undefined due to quasi-ballistic trajectories [81, 83, 84].

Finally, the diffusion exponent α shifts toward larger values within parameter regions containing phase-space attractors. p. 9
Where the paper says it

The correspondence between these qualitative changes in phase-space structure and the variations in kpt suggests that the non-monotonic behaviour of the average contraction rate is associated with the formation, coexistence, and disappearance of attracting and repelling sets. In particular, the emergence of attracting sets is consistent with the more negative values of kpt observed within this interval, whereas repelling regions may provide a competing contribution through local phase-space expansion.

Discussion and Future Directions

The work provides a first step toward understanding transport in active Lorentz gases and lays the groundwork for exploring hydrodynamic quantum analogs in periodic media. The model can be realized experimentally using magnetic forcing to create an effective soft Lorentz gas for walking droplets. p. 3

Where the paper says it

We note that, in our model, the periodic potential acts directly on the particle without explicitly modifying the particle-generated wavefield. This scenario can be realized experimentally in walking-droplet systems using magnetic forcing [24], in contrast to implementations based on subsurface topographies.

Future work should explore the high-memory regime where long-lived wavefields introduce strong non-Markovian effects. p. 0
Where the paper says it

There are three key features that make this system unique: (i) the walking droplet and its underlying wavefield coexist as a wave-particle entity; without the droplet, the liquid surface becomes flat, while the gradient of the accompanying waves drives the droplet motion. (ii) In certain regimes where the wave field decays slowly, the droplet retains long-term memory of its past motion, rendering the dynamics intrinsically non-Markovian.

The interplay between conservative and dissipative dynamics is a key area for future theoretical embedding.

Acknowledgments

The authors acknowledge CSC–Finnish IT centre for Science for computational resources. The work was supported by Research Council of Finland, ManyBody2D Project No. p. 0

Where the paper says it

IOP Publishing Journal vv (yyyy) aaaaaa Author et al Journal Name PAPER Crossmark RECEIVED dd Month yyyy REVISED dd Month yyyy An active Lorentz gas: walking droplets in periodic media Aleksi Majaniemi1 , Esko Toivonen1,∗ , Rahil N Valani2 , Rainer Klages3,4 , and Esa R¨as¨anen1 1Computational Physics Laboratory, Tampere University, Finland 2Rudolf Peierls Centre for Theoretical Physics, Parks Road, University of Oxford, OX1 3PU, United Kingdom 3Centre for Complex Systems, School of Mathematical Sciences, Queen Mary University of London, Mile End Road, London E1 4NS, United Kingdom 4London Mathematical Laboratory, 8 Margravine Gardens, London W6 8RH, United Kingdom ∗Author to whom any correspondence should be addressed. E-mail: esko.

349956. R.V. acknowledges the support of the Leverhulme Trust [Grant No LIP-2020-014] and the ERC Advanced Grant ActBio (funded as UKRI Frontier Research Grant EP/Y033981/1). The simulation data are available upon reasonable request. p. 7
Where the paper says it

IOP Publishing Journal vv (yyyy) aaaaaa Author et al 0. 20 kpt Simulation data kpt(w) ¼ 0:3031e ¡2:8259w 0:4555 ¡ 0:3047 Figure 2.

Supplementary video is provided for the evolution of the phase-space projection and trajectories corresponding to attractors. p. 7
Where the paper says it

We find that a stretched exponential form fits the data well with fit parameters shown in the Figure, with the asymptotic value almost matching the free-flight value of k free pt ≈ −0. Evolution of the phase-space projection (see text) and trajectories corresponding to attractors in the w range of the inset are shown in the Supplementary video.

The references listed cover various aspects of active matter, Lorentz gas, and related dynamical systems. p. 0
Where the paper says it

We identify parameter regimes exhibiting strong bifurcation cascades, anomalous diffusion, and significant phase-space contraction. Our results establish the active Lorentz gas as a rich framework for studying transport in dissipative dynamical systems and provide a bridge between active matter and classical models of chaotic transport, with potential implications for hydrodynamic quantum analogs in periodic media.

References

[1] Couder Y, Fort E, Gautier C H and Boudaoud A 2005 Phys. Rev. Lett. 94 177801

[2] Couder Y, Proti‘ere S, Fort E and Boudaoud A 2005 Nature 437 208–208

[3] Valani R N, Slim A C and Simula T 2019 Phys. Rev. Lett. 123(2) 024503

[4] Pismen L M 2021 Active Matter Within and Around Us: From Self-Propelled Particles to Flocks and Living Forms (Cham: Springer)

[5] Löwen H 2020 J. Chem. Phys. 152 040901

[6] Valani R N and Paganin D M 2023 Chaos 33 ISSN 1054-1500 023125

[7] Ramaswamy S 2010 Annu. Rev. Condens. Matter Phys. p. 0

Where the paper says it

We identify parameter regimes exhibiting strong bifurcation cascades, anomalous diffusion, and significant phase-space contraction. Our results establish the active Lorentz gas as a rich framework for studying transport in dissipative dynamical systems and provide a bridge between active matter and classical models of chaotic transport, with potential implications for hydrodynamic quantum analogs in periodic media.

p. 0
Where the paper says it

We identify parameter regimes exhibiting strong bifurcation cascades, anomalous diffusion, and significant phase-space contraction. Our results establish the active Lorentz gas as a rich framework for studying transport in dissipative dynamical systems and provide a bridge between active matter and classical models of chaotic transport, with potential implications for hydrodynamic quantum analogs in periodic media.

1 323–345

[8] Marchetti M C, Joanny J F, Ramaswamy S, Liverpool T B, Prost J, Rao M and Simha R A 2013 Rev. Mod. Phys. 85 1143–1189

[9] Bernheim-Groswasser A, Gov N S, Safran S A and Tzlil S 2018 Advanced Materials 30 1707028

[10] Schweitzer F 2003 Brownian agents and active particles (Berlin: Springer)

[11] Romanczuk P, Bär M, Ebeling W, Lindner B and Schimansky-Geier L 2012 Eur. Phys. J. Spec. Top. 202 1–162 ISSN 1951-6401

[12] Bechinger C, Di Leonardo R, Löwen H, Reichhardt C, Volpe G and Volpe G 2016 Rev. Mod. Phys. 88(4) 045006

[13] Klages R, Lucero-Azuara N, Kumar V, Gentili A, Torres P A R, Eizaguirre C, Volpe G and Palyulin V 2026 Modelling the movements of organisms: From cell migration to bumblebee flights to foraging sea turtles The Mathematics of Movement: An Interdisciplinary Approach to Mutual Challenges in Animal Ecology and Cell Biology ed Giuggioli L and Maini P K (Berlin: Springer) pp 223–249

[14] Paramanick S, Pardhi U, Soni H and Kumar N 2025 Phys. Rev. Lett. 135(16) 168301

[15] Howse J R, Jones R A L, Ryan A J, Gough T, Vafabakhsh R and Golestanian R 2007 Phys. Rev. Lett. 99(4) 048102

[16] Vicsek T and Zafeiris A 2012 Phys. Rep. 517 71–140 ISSN 0370-1573

[17] Doostmohammadi A, Ign´es-Mullol J, Yeomans J M and Sagu´es F 2018 Nature Communications 9 3246 ISSN 2041-1723

[18] Cates M E and Tailleur J 2015 Annu. Rev. Condens. Matter Phys. 6 219–244

[19] S´aenz P J, Cristea-Platon T and Bush J W M 2018 Chaos 28 096116

[20] Harris D M, Moukhtar J, Fort E, Couder Y and Bush J W M 2013 Phys. Rev. E 88(1) 011001

[21] Durey M, Milewski P A and Wang Z 2020 J. Fluid Mech. p. 0

Where the paper says it

In the periodic setting, a point particle scatters elastically off disks arranged on a two-dimensional lattice. Here we replace the disks by smooth potentials and the particle by the widely studied walking droplet, which propels itself on a vertically vibrating fluid.

p. 0
Where the paper says it

In the periodic setting, a point particle scatters elastically off disks arranged on a two-dimensional lattice. Here we replace the disks by smooth potentials and the particle by the widely studied walking droplet, which propels itself on a vertically vibrating fluid.

p. 0
Where the paper says it

In the periodic setting, a point particle scatters elastically off disks arranged on a two-dimensional lattice. Here we replace the disks by smooth potentials and the particle by the widely studied walking droplet, which propels itself on a vertically vibrating fluid.

p. 0
Where the paper says it

In the periodic setting, a point particle scatters elastically off disks arranged on a two-dimensional lattice. Here we replace the disks by smooth potentials and the particle by the widely studied walking droplet, which propels itself on a vertically vibrating fluid.

891 A3

[22] Cristea-Platon T, S´aenz P J and Bush J W M 2018 Chaos 28 096116

[33] Bush J W M 2015 Annu. Rev. Fluid Mech. 47 269–292

[34] Bush J W M and Oza A U 2020 Rep. Prog. Phys. 84 017001

[35] Bush J W M, Frumkin V and S´aenz P J 2024 Applied Physics Letters 125 030503 ISSN 6951

[36] Bush J W M, Papatryfonos K and Frumkin V 2024 The State of Play in Hydrodynamic Quantum Analogs (Cham: Springer International Publishing) pp 7–34 ISBN 978-3-031-49861-9

[37] Oza A U, Rosales R R and Bush J W M 2013 J. Fluid Mech. 737 552–570

[38] Proti‘ere S, Boudaoud A and Couder Y 2006 J. Fluid Mech. 554 85–108

[39] Bush J W M, Oza A U and Mol´aˇcek J 2014 Journal of Fluid Mechanics 755 R7

[40] Hubert M, Labousse M and Perrard S 2017 Phys. Rev. E 95(6) 062607

[41] López A G and Valani R N 2025 Phys. Rev. E 111(2) L022201

[43] Schweitzer F, Ebeling W and Tilch B 1998 Phys. Rev. Lett. 80 5044–5047

[73] Dettmann C 2014 Commun. Theor. Phys. 62 521

[83] Toivonen E, Kaipainen J, Molkkari M, Keski-Rahkonen J, Klages R and R¨as¨anen E 2025 Phys. Rev. E 111(1) 014216

[84] Toivonen E, Majaniemi A, Klages R and R¨as¨anen E 2025 Physical Review E 112 044217

[85] Pikovsky A 2023 Chaos 33 113114

[86] Heller E J, Fleischmann R and Kramer T 2021 Physics Today 74 44–51

[87] Heller E J and Shaw S 2003 Int. J. Mod. Phys. B 17 3977–3987 ISSN 0217-9792

[88] Topinka M A, LeRoy B J, Westervelt R M, Shaw S E J, Fleischmann R, Heller E J, Maranowski K D and Gossard A C 2001 Nature 410 183–186 ISSN 1476-4687

[89] Patsyk A, Sivan U, Segev M and Bandres M A 2020 Nature 583 60–65 ISSN 1476-4687

[90] Daza A, Heller E J, Graf A M and R¨as¨anen E 2021 Proceedings of the National Academy of Sciences 118 e21102851

[93] Chalangari F, Varma A V, Keski-Rahkonen J and R¨as¨anen E 2026 Phys. Rev. B 113(19) 195401

[94] Damiano A P, Brun P T, Harris D M, Galeano-Rios C A and Bush J W M 2016 Experiments in Fluids 57 163 ISSN 1432-1114

[95] Valani R N, Slim A, Paganin D, Simula T and Vo T 2021 Phys. Rev. E 104(1) 015106

[96] Labousse M and Perrard S 2014 Phys. Rev. E 90(2) 022913

[103] Solanp¨a¨a J, Luukko P and R¨as¨anen E 2016 Computer Physics Communications 199 133–138 ISSN 0010-4655

[104] Kibble T W B and Berkshire F H 2004 Classical Mechanics (Imperial College Press)

[113] Bacot V, Perrard S, Labousse M, Couder Y and Fort E 2019 Phys. Rev. Lett. 122(10) 104303

[114] Hubert M, Perrard S, Labousse M, Vandewalle N and Couder Y 2019 Phys. Rev. E 100(3) 032201

[15] Howse J R, Jones R A L, Ryan A J, Gough T, Vafabakhsh R and Golestanian R 2007 Phys. Rev. Lett. 99(4) 048102

[16] Vicsek T and Zafeiris A 2012 Phys. Rep. 517 71–140 ISSN 0370-1573

[83] Toivonen E, Kaipainen J, Molkkari M, Keski-Rahkonen J, Klages R and R¨as¨anen E 2025 Phys. Rev. E 111(1) 014216

[84] Toivonen E, Majaniemi A, Klages R and R¨as¨anen E 2025 Physical Review E 112 044217

[85] Pikovsky A 2023 Chaos 33 113114

[86] Heller E J, Fleischmann R and Kramer T 2021 Physics Today 74 44–51

[87] Heller E J and Shaw S 2003 Int. J. Mod. Phys. B 17 3977–3987 ISSN 0217-9792

[88] Topinka M A, LeRoy B J, Westervelt R M, Shaw S E J, Fleischmann R, Heller E J, Maranowski K D and Gossard A C 2001 Nature 410 183–186 ISSN 1476-4687

[89] Patsyk A, Sivan U, Segev M and Bandres M A 2020 Nature 583 60–65 ISSN 1476-4687

[90] Daza A, Heller E J, Graf A M and R¨as¨anen E 2021 Proceedings of the National Academy of Sciences 118 e21102851 <ref:2609.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Improved AI can perform phase-space structure analysis in non-conservative systems by using an energyvariance filtering method to distinguish between localized periodic, quasi-ballistic periodic, and chaotic trajectories. This enables the construction of bifurcation diagrams in a non-conservative setting that reveal islands of stability collapse into attracting and repelling sets.

  2. Improved AI can predict the emergence and disappearance of attracting and repelling sets by correlating the nonmonotonic behaviour of the average contraction rate with specific parameter regimes, as these coincide with the formation, coexistence, and disappearance of attracting and repelling sets.

  3. Improved AI can characterize transport properties in active Lorentz gases by calculating an average phase space contraction rate (kpt) to identify regimes where dynamics are governed by an attracting set versus a repeller-dominated region. This allows the system to distinguish between attractor-driven motion, long-lived repeller-induced transients, and anomalous diffusion.

  4. Improved AI can predict the diffusion exponent alpha for individual trajectories in both dissipative and conservative systems by analyzing the distribution of exponents as a function of parameter w. This helps in distinguishing between transport regimes, such as normal diffusion (alpha = 1) and superdiffusive behavior (alpha > 1).

Abstract

The Lorentz gas is a paradigmatic model in dynamical systems theory for understanding the origin of nonequilibrium transport in terms of microscopic deterministic chaos. In the periodic setting, a point particle scatters elastically off disks arranged on a two-dimensional lattice. Here we replace the disks by smooth potentials and the particle by the widely studied walking droplet, which propels itself on a vertically vibrating fluid. In the low-memory limit, this droplet reduces to a particle with nonlinear active friction. We call this system an active Lorentz gas. Using extensive numerical simulations, we analyze how dissipation generated by the active deterministic dynamics alters the phase-space structure of the corresponding conservative Lorentz gas. We find that islands of stability collapse into attracting and repelling sets. To characterize these structures, we introduce an energy-variance filtering method that distinguishes localized periodic, quasi-ballistic periodic, and chaotic trajectories, enabling the construction of bifurcation diagrams in a non-conservative setting. We identify parameter regimes exhibiting strong bifurcation cascades, anomalous diffusion, and significant phase-space contraction. Our results establish the active Lorentz gas as a rich framework for studying transport in dissipative dynamical systems and provide a bridge between active matter and classical models of chaotic transport, with potential implications for hydrodynamic quantum analogs in periodic media.

Sources

Related papers