The constrained degree and fixed-point index theory for set-valued maps (Q2495371): Difference between revisions

The article is a continuation of the previous article of the authors [Nonlinear Anal., Theory Methods Appl. 65, 175--209 (2006; Zbl 1107.54019)]. It contains some modifications and generalizations of arguments and constructions presented there. First, the degree theory for so-called admissible (in general, non-convex-valued) upper semicontinuous strongly tangent vector fields \(\Phi: \;U \to E\) is considered; here \(E\) is a Banach space, \(U\) is an open set in an \({\mathcal L}\)-retract in \(E\); \(\Phi\) is strongly tangent if, for \(x \in M\), \(\Phi(x) \subset T_M(x)\), \(T_M(x)\) is the Clarke tangent cone to \(M\) at a point \(x \in M\); the admissibility property of \(\Phi\) means that \(\Phi = F \circ \Psi\), \(\Psi: \;U \to Y\) (\(Y\) is a metric space) is an upper semicontinuous map with values being proximally \(\infty\)-connected in \(Y\) and \(F: \;Y \to E\) is a continuous map. Secondly, the authors discuss how the assumption concerning the local compactness property of the \({\mathcal L}\)-retract \(M\) may be relaxed; here the authors consider the case \(M = \{x \in E: \;f(x) \leq 0\}\) where \(f: \;E \to {\mathbb R}\) is a locally Lipschitzian function and describe additional conditions for \(\Phi\) under which the degree theory can be constructed. Finally, the problem to define the fixed-point index for weakly inward maps defined on arbitrary closed convex sets (\(\Phi\) is weakly inward if, for all \(x \in U\), \(\Phi(x) \cap (x + T_M(x)) \neq \emptyset)\) is considered.