Minimal time functions and the smallest intersecting ball problem with unbounded dynamics (Q1937003): Difference between revisions

The paper studies some generalizations of the smallest intersecting ball problem, which is an extension of the smallest enclosing circle problem introduced by Sylvester in 1857. Given a normed space \(X\) and a nonempty, closed, convex subset \(F\) of \(X,\) the minimal time function associated to a nonempty subset \(Q\) of \(X\) is defined as \[ {\mathcal T}_Q^F(x):=\inf\{t\geq 0:\,(x+tF)\cap Q\neq \emptyset\}. \] For target sets \(\Omega_i\) (\(i=1,2,\dots, m\)) and a constraint set \(\Omega,\) it turns out that the smallest intersecting ball problem generated by the dynamic \(F\) can be modeled as follows: \[ {\text{minimize } } {\mathcal T}(x)\quad {\text{subject to}}\quad x\in \Omega, \] where \({\mathcal T}(x):=\max\{{\mathcal T}^F_{\Omega_i}(x):\,i=1,\dots,m\}.\) The authors focus on the study of the minimal time function \({\mathcal T}_Q^F\) with \(F\) unbounded; the case where \(F\) is bounded has already been investigated. Sections 3--5 present preliminary and of independ interest results; in particular, Section 4 is devoted to the investigation of some properties of \({\mathcal T}_Q^F\). In Section 6, mathematical modeling and numerical algorithms for solving the smallest intersecting ball problem and its further generalizations are studied.