Skip to main content
SpringerLink
Log in

Cellular Automata Modeling of Physical Systems

  • Reference work entry
Encyclopedia of Complexity and Systems Science

Definition of the Subject

The computational science community has always been faced with thechallenge of bringing efficient numerical tools to solve problems ofincreasing difficulty. Nowadays, the investigation and understandingof the so‐called complex systems, and the simulation of all kinds ofphenomena originating from the interaction of many components are ofcentral importance in many area of science.

Cellular automata turn out to be a very fruitful approach to addressmany scientific problems by providing an efficient way to model andsimulate specific phenomena for which more traditional computationaltechniques are hardly applicable.

The goal of this article is to provide the reader with the foundationof this approach, as well as a selection of simple applications ofthe cellular automata approach to the modeling of physical systems.We invite the reader to consult the web sitehttp://cui.unige.ch/%7Echopard/CA/Animations/img-root.htmlin orderto view short movies about several of the...

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Log in via an institution

Institutional subscriptions

Abbreviations

BGK models:

Lattice Boltzmann models where the collision termΩ is expressed as a deviation from a given local equilibriumdistribution \( { f^{(0)} } \), namely \( { \Omega=(f^{(0)}-f)/\tau } \), where fis the unknown particle distribution and τ a relaxation time(which is a parameter of the model). BGK stands for Bhatnager, Grossand Krook who first considered such a collision term, but notspecifically in the context of lattice systems.

CA:

Abbreviation for cellular automata or cellular automaton.

Cell:

The elementary spatial component of a CA. The cell ischaracterized by a state whose value evolves in time according to theCA rule.

Cellular automaton:

System composed of adjacent cells or sites(usually organized as a regular lattice) which evolves in discrete timesteps. Each cell is characterized by an internal state whose value belongsto a finite set. The updating of these states is made in parallel accordingto a local rule involving only a neighborhood of each cell.

Conservation law:

A property of a physicalsystem in which some quantity (such as mass, momentum or energy) is locallyconserved during the time evolution.These conservation laws should be included in the microdynamicsof a CA model because they are essential ingredients governing themacroscopic behavior of any physical system.

Collision:

The process by which the particles of a LGAchange their direction of motion.

Continuity equation:

An equation of the form \( \partial_t\rho+\operatorname{div}\rho \mathbf{u} = 0 \) expressing the mass (or particle number) conservation law.The quantity ρ is the local density of particles and \( { \mathbf{u} } \) the localvelocity field.

Critical phenomena:

The phenomena which occur in the vicinity ofa continuous phase transition, and are characterized by very long correlationlength.

Diffusion:

A physical process described by the equation \( { \partial_t\rho=D\nabla^2\rho } \), where ρ is the density of a diffusing substance.Microscopically, diffusion can be viewed as a random motion of particles.

DLA:

Abbreviation of Diffusion Limited Aggregation. Model ofa physical growth process in which diffusing particles stick on an existingcluster when they hit it. Initially, the cluster is reduced to a singleseed particle and grows as more and more particles arrive. A DLA cluster isa fractal object whose dimension is typically 1.72 if the experiment isconducted in a two‐dimensional space.

Dynamical system:

A system of equations (differential equations or discretized equations) modeling the dynamical behavior of a physical system.

Equilibrium states:

States characterizing a closed system ora system in thermal equilibrium with a heat bath.

Ergodicity:

Property of a system or process for which thetime‐averages of the observables converge, in a probabilistic sense, totheir ensemble averages.

Exclusion principle:

A restriction which is imposed on LGA or CAmodels to limit the number of particles per site and/or lattice directions.This ensures that the dynamics can be described with a cellular automatarule with a given maximum number of bits. The consequence of this exclusionprinciple is that the equilibrium distribution of the particle numbersfollows a Fermi–Dirac-like distribution in LGA dynamics.

FHP model:

Abbreviation for the Frisch, Hasslacher and Pomeaulattice gas model which was the first serious candidate to simulatetwo‐dimensional hydrodynamics on a hexagonal lattice.

Fractal:

Mathematical object usually having a geometricalrepresentation and whose spatial dimension is not an integer. The relationbetween the size of the object and its “mass” does not obey that of usualgeometrical objects. A DLA cluster is an example of a fractal.

Front:

The region where some physical process occurs. Usuallythe front includes the locations in space that are first affected by thephenomena. For instance, in a reaction process between two spatiallyseparated reactants, the front describes the region where the reactiontakes place.

HPP model:

Abbreviation for the Hardy, de Pazzis and Pomeaumodel. The first two‐dimensional LGA aimed at modeling the behavior ofparticles colliding on a square lattice with mass and momentumconservation. The HPP model has several physical drawbacks that have beenovercome with the FHP model.

Invariant:

A quantity which is conserved during the evolution ofa dynamical system. Some invariants are imposed by the physical laws (mass,momentum, energy) and others result from the model used to describephysical situations (spurious, staggered invariants). Collisionalinvariants are constant vectors in the space where the Chapman–Enskogexpansion is performed, associated to each quantity conserved by thecollision term.

Ising model:

Hamiltonian model describing the ferromagneticparamagnetic transition. Each local classical spin variables \( { s_i=\pm 1 } \)interacts with its neighbors.

Isotropy:

The property of continuous systems to beinvariant under any rotations of the spatial coordinate system. Physicalquantities defined on a lattice and obtained by an averaging procedure mayor may not be isotropic, in the continuous limit. It depends on the type oflattice and the nature of the quantity. Second‐order tensors areisotropic on a 2D square lattice but fourth‐order tensors need a hexagonallattice.

Lattice:

The set of cells (or sites) making up the spatialarea covered by a CA.

Lattice Boltzmann model:

A physical model defined on a latticewhere the variables associated to each site represent an averagenumber of particles or the probability of the presence of a particlewith a given velocity. Lattice Boltzmann models can be derived fromcellular automata dynamics by an averaging and factorizationprocedure, or be defined per se, independently of a specificrealization.

Lattice gas:

A system defined on a lattice where particlesare present and follow given dynamics. Lattice gas automata (LGA)are a particular class of such a system where the dynamics areperformed in parallel over all the sites and can be decomposed intwo stages: (i) propagation: the particles jump toa nearest‐neighbor site, according to their direction of motionand (ii) collision: the particles entering the same site at thesame iteration interact so as to produce a new particledistribution. HPP and FHP are well-known LGA.

Lattice spacing:

The separation between two adjacent sites ofa regular lattice. Throughout this book, it is denoted by the symbol \( { \Delta_\mathrm{r} } \).

LB:

Abbreviation for Lattice Boltzmann.

LGA:

Abbreviation for Lattice Gas Automaton. See lattice gas modelfor a definition.

Local equilibrium:

Situation in which a large system can bedecomposed into subsystems, very small on a macroscopic scale but large ona microscopic scale such that each sub‐system can be assumed to be in thermalequilibrium. The local equilibrium distribution is the function which makesthe collision term of a Boltzmann equation vanish.

Lookup table:

A table in which all possible outcomes ofa cellular automata rule are pre‐computed. The use of a lookup table yieldsa fast implementation of a cellular automata dynamics since howevercomplicated a rule is, the evolution of any configuration of a site and itsneighbors is directly obtained through a memory access. The size of a lookuptable grows exponentially with the number of bits involved in the rule.

Margolus neighborhood:

A neighborhoodmade of two-by-two blocks of cells, typically in a two‐dimensional squarelattice. Each cell is updated according to the values of theother cells in the same block. A different rule may possibly be assigneddependent on whetherthe cell is at the upper left, upper right, lower left or lower rightlocation. After each iteration, the lattice partition definingthe Margolus blocs is shifted one cell right and one cell down so that atevery other step, information can be exchanged across the lattice.Can be generalized to higher dimensions.

Microdynamics:

The Boolean equation governing the time evolutionof a LGA model or a cellular automata system.

Moore neighborhood:

A neighborhood composed of the central celland all eight nearest and next‐nearest neighbors in a two‐dimensionalsquare lattice. Can be generalized to higher dimensions.

Multiparticle models:

Discrete dynamics modeling a physicalsystem in which an arbitrary number of particles is allowed ateach site. This is an extension of an LGA where no exclusionprinciple is imposed.

Navier–Stokes equation:

The equation describing the velocityfield \( { \mathbf{u} } \) in a fluid flow. For an incompressible fluid (\( { \partial_t\rho=0 } \)), itreads

$$ \partial_t \mathbf{u} + (\mathbf{u}\cdot\nabla)\mathbf{u}=-{1\over\rho}\nabla P +\nu\nabla^2\mathbf{u} $$

where ρ is the density and P the pressure. The Navier–Stokesequation expresses the local momentum conservation in the fluid and, asopposed to the Euler equation, includes the dissipative effects witha viscosity term \( { \nu\nabla^2\mathbf{u} } \). Together with the continuity equation,this is the fundamental equation of fluid dynamics.

Neighborhood:

The set of all cells necessary to computea cellular automaton rule. A neighborhood is usually composed of severaladjacent cells organized in a simple geometrical structure. Moore, vonNeumann and Margolus neighborhoods are typical examples.

Occupation numbers:

Boolean quantities indicating the presenceor absence of a particle in a given physical state.

Open system:

A system communicating with the environment byexchange of energy or matter.

Parallel:

Refers to an action which is performed simultaneouslyat several places. A parallel updating rule corresponds to the updating ofall cells at the same time as if resulting from the computations of severalindependent processors.

Partitioning:

A technique consisting of dividing space inadjacent domains (through a partition) so that the evolution ofeach block is uniquely determined by the states of the elementswithin the block.

Phase transition:

Change of state obtained when varyinga control parameter such as the one occurring in the boiling or freezing ofa liquid, or in the change between ferromagnetic and paramagnetic states ofa magnetic solid.

Propagation:

This is the process by which the particles ofa LGA are moved to a nearest neighbor, according to the direction oftheir velocity vector \( { \mathbf{v_i} } \). In one time step Δ t the particletravel from cell \( { \mathbf{r} } \) to cell \( { \mathbf{r}+\mathbf{v_i}\Delta_t } \) where \( { \mathbf{r}+\mathbf{v_i}\Delta_t } \) is thenearest neighbor in lattice direction i.

Random walk:

A series of uncorrelated steps of length unitydescribing a random path with zero average displacement but characteristicsize proportional to the square root of the number of steps.

Reaction‐diffusion systems:

Systems made of one or severalspecies of particles which diffuse and react among themselves to producesome new species.

Scaling hypothesis:

A hypothesis concerning the analyticalproperties of the thermodynamic potentials and the correlation functions ina problem invariant under a change of scale.

Scaling law:

Relations among the critical exponents describing the power law behaviors of physical quantities in systems invariant under a change of scale.

Self‐organized criticality:

Concept aimed at describing a classof dynamical systems which naturally drive themselves to a state whereinteresting physics occurs at all scales.

Site:

Same as a cell, but preferred terminology in LGA and LBmodels.

Spatially extended systems:

Physical systems involving manyspatial degrees of freedom and which, usually, have rich dynamicsand show complex behaviors. Coupled map lattices and cellularautomata provides a way to model spatially extended systems.

Spin:

Internal degree of freedom associated to particles inorder to describe their magnetic state. A widely used case is the one ofclassical Ising spins. To each particle, one associates an “arrow” whichis allowed to take only two different orientations, up or down.

Time step:

Interval of time separating two consecutiveiterations in the evolution of a discrete time process, like a CAor a LB model. Throughout this work the time step is denoted bythe symbol Δ t .

Universality:

The phenomenon whereby many microscopicallydifferent systems exhibit a critical behavior with quantitatively identicalproperties such as the critical exponents.

Updating:

operation consisting of assigning a new value toa set of variables, for instance those describing the states ofa cellular automata system. The updating can be done in paralleland synchronously as is the case in CA dynamics or sequentially,one variable after another, as is usually the case forMonte–Carlo dynamics. Parallel, asynchronous updating is lesscommon but can be envisaged too. Sequential and parallel updatingschemes may yield different results since the interdependenciesbetween variables are treated differently.

Viscosity:

A property of a fluid indicating how much momentum“diffuses” through the fluid in a inhomogeneous flow pattern.Equivalently, it describes the stress occurring between two fluid layersmoving with different velocities. A high viscosity means that the resultingdrag force is important and low viscosity means that this force is weak.Kinematic viscosity is usually denoted by ν and dynamic viscosity isdenoted by \( { \eta=\nu\rho } \) where ρ is the fluid density.

von Neumann neighborhood:

On a two‐dimensional squarelattice, the neighborhood including a central cell and its nearestneighbors north, south, east and west.

Ziff model:

A simple model describing adsorption–dissociation–desorption on a catalytic surface.This model is based upon some of the known steps of the reaction A–B 2 on a catalyst surface (for example CO–O2).

Bibliography

Primary Literature

  1. Alemani D, Chopard B, Buffle J, Galceran J (2006) Two grid refinement methods in theLattice Boltzmann framework for reaction‐diffusion processes. Phys Chem Chem Phys 8:35

    Google Scholar 

  2. Ansumali S, Karlin I, Arcidiacono S, Abbas A, Prasianakis N (2007) Hydrodynamics beyondNavier–Stokes: Exact solution to the lattice boltzmann hierarchy. Phys Rev Lett 98:124502

    ADS  Google Scholar 

  3. Banks E (1971) Information processing and transmission in cellular automata.Tech rep., MIT, MAC TR-81

    Google Scholar 

  4. Boon JP (ed) (1992) Advanced Research Workshop on Lattice Gas Automata Theory,Implementations, and Simulation. J Stat Phys 68(3/4):347–672

    Google Scholar 

  5. Boon JP, Dab D, Kapral R, Lawniczak A (1996) Lattice gas automata for reactive systems.Phys Rep 273:55–148

    MathSciNet  ADS  Google Scholar 

  6. Burks A (1970) Von Neumann’s self‐reproducing automata. In: Burks A (ed) Essays on Cellular Automata. University of Illinois Press, Chicago, pp 3–64

    Google Scholar 

  7. Burstedde C, Klauck K, Schadschneider A, Zittartz J (2001) Simulation of pedestriandynamics using a two‐dimensional cellular automaton. Physica A 295:506–525

    ADS  Google Scholar 

  8. Chen S, Doolen G (1998) Lattice Boltzmann methods for fluid flows. Annu Rev FluidMech 30:329

    MathSciNet  ADS  Google Scholar 

  9. Chopard B, Droz M (1987) Cellular automata approach to non equilibrium phase transitionsin a surface reaction model: static and dynamic properties. J Phys A 21:205

    ADS  Google Scholar 

  10. Chopard B, Droz M (1998) Cellular Automata Modeling of Physical Systems. CambridgeUniversity Press, Cambridge

    Google Scholar 

  11. Chopard B, Dupuis A (2003) Cellular automata simulations of traffic: a model for thecity of geneva. Netw Spat Econ 3:9–21

    Google Scholar 

  12. Chopard B, Luthi P, Droz M (1994) Reaction‐diffusion cellular automata model for theformation of Liesegang patterns. Phys Rev Lett 72(9):1384–1387

    ADS  Google Scholar 

  13. Chopard B, Luthi PO, Queloz PA (1996) Cellular automata model of car traffic intwo‐dimensional street networks. J Phys A 29:2325–2336

    MathSciNet  ADS  Google Scholar 

  14. Doolen G (ed) (1990) Lattice Gas Method for Partial DifferentialEquations. Addison‐Wesley, Redwood City

    Google Scholar 

  15. Drossel B, Schwabl F (1992) Self‐organized critical forest‐fire model. Phys Rev Lett 69:1629

    ADS  Google Scholar 

  16. Farmer D, Toffoli T, Wolfram S (eds) (1984) Cellular Automata. Proceedings of anInterdisciplinary Workshop, Los Alamos. Physica D, vol 10. North‐Holland, Amsterdam

    Google Scholar 

  17. Fisch R, Gravner J, Griffeath D (1991) Threshold‐range scaling of excitable cellularautomata. Stat Comput 1:23

    Google Scholar 

  18. Frisch U, Hasslacher B, Pomeau Y (1986)Lattice‐gas automata for the navier–stokes equation.Phys Rev Lett 56:1505

    ADS  Google Scholar 

  19. Gardner M (1970) The fantastic combinations of john conway’s new solitaire game life.Sci Am 220(4):120

    Google Scholar 

  20. Gravner J, Griffeath D (1993) Threshold grouse dynamics.Trans Amer Math Soc 340:837

    MathSciNet  Google Scholar 

  21. Gunton J, Droz M (1983) Introduction to the Theory of Metastable and Unstable States.Springer, Berlin

    Google Scholar 

  22. Henisch H K (1988) Crystals in Gels and Liesegang Rings. Cambridge University Press, Cambridge

    Google Scholar 

  23. Hoekstra A, Lorenz E, Falcone JL, Chopard B (2007) Towards a complex automataframework for multi‐scale modeling: Formalism and the scale separation map. In: Shi Y et al (ed) ComputationalSciences ICCS 2007. LNCS, vol 4487. Springer, Berlin, pp 922–939

    Google Scholar 

  24. Kanai M, Nishinari K, Tokihiro T (2005) Stochastic optimal velocity model and itslong-lived metastability. Phys Rev E 72:035102(R)

    ADS  Google Scholar 

  25. Kanai M, Nishinari K, Tokihiro T (2006) Stochastic cellular automaton model for trafficflow. In: Yacoubi SE, Chopard B, Bandini S (eds) Cellular Automata: 7th ACRI conference. LNCS,vol 4173. Springer, Berlin, pp 538–547

    Google Scholar 

  26. Kapral R, Showalter K (eds) (1995) Chemical Waves and Patterns. Kluwer, Dordrecht

    Google Scholar 

  27. Keener J, Tyson J (1992) The dynamics of scroll waves in excitable media. SIAM Rev. 34:1–39

    MathSciNet  Google Scholar 

  28. Luthi P O, Preiss A, Ramsden JJ, Chopard B (1998) A cellular automaton model forneurogenesis in drosophila. Physica D 118:151–160

    ADS  Google Scholar 

  29. MarconiS, Chopard B (2002) A multiparticle lattice gas automata fora crowd. In: Bardini S et al (ed) Proceedings of ACRI 2002 Geneva, Oct, 2002. Lecture notes in computer science, vol 2493. Springer, Berlin, p 230

    Google Scholar 

  30. Marconi S, Chopard B (2006) Discrete physics, cellular automata and cryptography. In:Yacoubi SE, Chopard B, Bandini S (eds) Cellular Automata: 7th ACRI conference. LNCS, vol 4173.Springer, Berlin, pp 617–626

    Google Scholar 

  31. Muray J (1990) Mathematical Biology. Springer, Berlin

    Google Scholar 

  32. Nagel K, Herrmann H (1993) Deterministic models for traffic jams. Physica A 199:254

    ADS  Google Scholar 

  33. Nagel K, Schreckenberg M (1992) Cellular automaton model for freeway traffic.J Physique I 2:2221

    ADS  Google Scholar 

  34. Pearson JE (1993) Complex patterns in a simple system. Science 261:189–192

    ADS  Google Scholar 

  35. Rothman D, Zaleski S (1997) Lattice‐Gas Cellular Automata: Simple Models of ComplexHydrodynamics. Collection Aléa. Cambridge University Press, Cambridge

    Google Scholar 

  36. Schadschneider A, Schreckenberg M (1993) Cellular automaton models and traffic flow.J Phys A 26:L679

    ADS  Google Scholar 

  37. Schreckenberg M, Schadschneider A, Nagel K, Ito N (1995) Discrete stochastic models fortraffic flow. Phys. Rev. E 51:2939

    ADS  Google Scholar 

  38. Selkov E (1968) Self‐oscillation in glycolysis: A simple kinetic model. Eur J Biochem4:79

    Google Scholar 

  39. Shannon C (1949) Communication theory of secrecy systems. Bell Syst Tech J 28:656–715

    MathSciNet  Google Scholar 

  40. Succi S (2001) The Lattice Boltzmann Equation, For Fluid Dynamics and Beyond. OxfordUniversity Press, Oxford

    Google Scholar 

  41. Sukop M, Thorne D (2005) Lattice Boltzmann Modeling: an Introduction for Geoscientistsand Engineers. Springer, Berlin

    Google Scholar 

  42. Toffoli T, Margolus N (1987) Cellular Automata Machines: a New Environment forModeling. MIT Press, Cambridge

    Google Scholar 

  43. Tolman S, Meakin P (1989) Off‐lattice and hypercubic‐lattice models fordiffusion‐limited aggregation in dimension 2–8. Phys Rev A 40:428–37

    ADS  Google Scholar 

  44. Vichniac G (1984) Simulating physics with cellular automata. Physica D 10:96–115

    MathSciNet  ADS  Google Scholar 

  45. Vicsek T (1989) Fractal Growth Phenomena. World Scientific, Singapore

    Google Scholar 

  46. Witten T, Sander L (1983) Diffusion‐limited aggregation. Phys Rev B 27:5686

    MathSciNet  ADS  Google Scholar 

  47. Wolf D, Schreckenberg M et al (eds) Traffic and Granular Flow ’97. Springer, Singapore

    Google Scholar 

  48. Wolf D, Schreckenberg M, Bachem A (eds) (1996)Traffic and Granular Flow. World Scientific, Singapore

    Google Scholar 

  49. Wolf‐Gladrow D A (2000) Lattice‐Gas Cellular Automata and Lattice Boltzmann Models:an Introduction. Lecture Notes in Mathematics, vol 1725. Springer, Berlin

    Google Scholar 

  50. Wolfram S (1986) Theory and Application of Cellular Automata. World Scientific, Singapore

    Google Scholar 

  51. WolframS (1994) Cellular Automata and Complexity. Addison‐Wesley, Reading

    Google Scholar 

  52. Yukawa S, Kikuchi M, Tadaki S (1994) Dynamical phase transition in one‐dimensionaltraffic flow model with blockage. J Phys Soc Jpn 63(10):3609–3618

    ADS  Google Scholar 

  53. Ziff R, Fichthorn K, Gulari E (1991) Cellular automaton version of the ab 2 reactionmodel obeying proper stoichiometry. J Phys. A 24:3727

    Google Scholar 

  54. Ziff R, Gulari E, Barshad Y (1986) Kinetic phase transitions in an irreversiblesurface‐reaction model. Phys Rev Lett 56:2553

    ADS  Google Scholar 

Books and Reviews

  1. Chopard B, Droz M (1998) Cellular AutomataModeling of Physical Systems. Cambridge University Press, Cambridge

    Google Scholar 

  2. Deutsch A, Dormann S (2005) Cellular Automaton Modeling of Biological Pattern Formation. Birkhäuser, Basel

    Google Scholar 

  3. Gaylord RJ, Nishidate K (1996) Modeling Nature with Cellular Automata usingMathematica. Springer, Berlin

    Google Scholar 

  4. Ilachinski A (2001) Cellular Automata: a discrete universe. World Scientific, Singapore

    Google Scholar 

  5. Rivet JP, Boon JP (2001) Lattice Gas Hydrodynamics. Cambridge University Press, Cambridge

    Google Scholar 

  6. Rothman D, Zaleski S (1994) Lattice‐gas models of phase separation: interface, phasetransition and multiphase flows. Rev Mod Phys 66:1417–1479

    ADS  Google Scholar 

  7. Weimar JR (1998) Simulation with Cellular Automata. Logos, Berlin

    Google Scholar 

  8. Wolfram S (2002) A new kind of science. Wolfram Sciences, Champaign

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Computer Science Department, University of Geneva, Geneva, Switzerland

    Bastien Chopard

Authors
  1. Bastien Chopard
    View author publications

    You can also search for this author in PubMed Google Scholar

Editor information

Editors and Affiliations

  1. RAMTECH LIMITED, 122 Escalle Lane, Larkspur, CA, 94939, USA

    Robert A. Meyers Ph. D. (Editor-in-Chief) (Editor-in-Chief)

Rights and permissions

Reprints and permissions

Copyright information

© 2009 Springer-Verlag

About this entry

Cite this entry

Chopard, B. (2009). Cellular Automata Modeling of Physical Systems. In: Meyers, R. (eds) Encyclopedia of Complexity and Systems Science. Springer, New York, NY. https://doi.org/10.1007/978-0-387-30440-3_57

Download citation

Publish with us

Policies and ethics

Search

Navigation