Optimal control of the discrete-time fractional-order Cucker-Smale model (Q1678313)

From MaRDI portal
Revision as of 05:12, 1 February 2024 by Import240129110113 (talk | contribs) (Added link to MaRDI item.)
scientific article
Language Label Description Also known as
English
Optimal control of the discrete-time fractional-order Cucker-Smale model
scientific article

    Statements

    Optimal control of the discrete-time fractional-order Cucker-Smale model (English)
    0 references
    14 November 2017
    0 references
    The discrete time system \[ \begin{aligned} x_i(t + h) - x_i(t) & = hv_i(t) \cr v_i(t + h) - v_i(t) & = h\sum_{j=1}^N a_{ij} (v_j(t) - v_i(t)) \quad i = 1, \dots, N \end{aligned} \eqno(1) \] where \(x_i(t), v_i(t) \in {\mathbb R}^d\) and \[ a_{ij} = {c \over (1 + \|x_i - x_j\|^2)^\beta} \] \((c >0, \beta \geq 0, \|\cdot\|\) the \(\ell^2\) norm of \({\mathbb R}^d)\) has been proposed by F. Cucker and S. Smale to model emerging collective behavior in interacting groups. Its continuous time counterpart is the system \[ \begin{aligned} x'_i(t) & = v_i(t) \cr v'_i(t) & = \sum_{j=1}^N a_{ij}(v_j(t) - v_i(t)) \quad i = 1, \dots, N \, . \end{aligned}\eqno(2) \] Cucker and Smale proved that if \(\beta <1/2\) the \(v_i(t)\) converge as \(t \to \infty\) to a common limit \(\bar v \in {\mathbb R}^d\) independent of the initial conditions (the consensus value) while if \(\beta \geq 1/2\) additional assumptions on the initial conditions are needed. In the continuous case (2), M. Caponigro, M. Fornasier, B. Piccoli and E. Trelat considered the case where the system does not converge to a consensus and used an external control \(u_i(t)\) added to the second line of (2); the trajectory minimizes \[ \int_0^T \bigg(\sum_{i=1}^N \Big\|v_i(t) - {1 \over N}\sum_{j=1}^N v_j(t) \Big \|^2 + \gamma \sum_{i=1}^N\|u_i(t)\| \bigg) dt \eqno(3) \] under a boundedness condition on the control. In this paper, the authors apply the external control strategy to the a version of the discrete system (1) where fractional order differences are used to model memory effects; the cost functional is a discrete analogue of (3). The main result is a version of Pontryagin's maximum principle for the discrete problem.
    0 references
    Cucker-Smale model
    0 references
    emergent collective behavior
    0 references
    interacting groups
    0 references
    consensus value
    0 references
    fractional difference
    0 references
    optimal control
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references