Stationary solutions of Keller-Segel-type crowd motion and herding models: Multiplicity and dynamical stability

Jean Dolbeault, Gaspard Jankowiak, Peter Markowich

Research output: Contribution to journalArticlepeer-review

4 Scopus citations


In this paper we study two models for crowd motion and herding. Each of the models is of Keller-Segel type and involves two parabolic equations, one for the evolution of the density and one for the evolution of a mean field potential. We classify all radial stationary solutions, prove multiplicity results, and establish some qualitative properties of these solutions, which are characterized as critical points of an energy functional. A notion of variational stability is associated with such solutions. Dynamical stability in the neighborhood of a stationary solution is also studied in terms of the spectral properties of the linearized evolution operator. For one of the two models, we exhibit a Lyapunov functional which allows us to make the link between the two notions of stability. Even in that case, for certain values of the mass parameter, with all other parameters taken in an appropriate range, we find that two dynamically stable stationary solutions exist. We further discuss the qualitative properties of the solutions using theoretical methods and numerical computations.

Original languageEnglish (US)
Pages (from-to)211-242
Number of pages32
JournalMathematics and Mechanics of Complex Systems
Issue number3
StatePublished - 2015

Bibliographical note

Publisher Copyright:
© 2015 Mathematical Sciences Publishers.


  • Continuum model
  • Crowd motion
  • Dynamical stability
  • Herding
  • Lyapunov functional
  • Non-self-adjoint evolution operators
  • Variational methods

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Numerical Analysis
  • Computational Mathematics


Dive into the research topics of 'Stationary solutions of Keller-Segel-type crowd motion and herding models: Multiplicity and dynamical stability'. Together they form a unique fingerprint.

Cite this