Non-Euclidean visibility problems (Q998215): Difference between revisions

Visibility problems have been studied since a century. Perhaps the most celebrated problems are the visible version of Gauss' circle problem and the so-called orchard problem. In contrast with Gauss' circle problem, orchard problem can be considered as solved in a wide sense [see \textit{Allen} (1986)] by elementary methods. In this paper the author considers the analog of visibility problems in hyperbolic plane represented by Poincaré's plane \(\mathbb{H}\), i.e., the upper half plane with the metric \(ds^2= y^{-2}\,dx^2+ y^{-2}\,dy^2\), the origin \(i\in\mathbb{H}\) and replacing the standard lattice \(\mathbb{Z}\times\mathbb{Z}\) by the orbit \({\mathfrak L}\) of \(z= i\) under the full modular group \(\text{SL}_2(\mathbb{Z})\). A point \(z\in{\mathfrak L}\), \(z\neq i\) is said to be visible if the arc of geodesic connecting \(i\) and \(z\) does not contain any other point in \({\mathfrak L}\). The study of the visibility problem in the hyperbolic case differs from the Euclidean case due to the following algebraic and geometric facts: (i) In the Euclidean case the group of integral translations is Abelian whereas in hyperbolic case the underlying group \(\text{SL}_2(\mathbb{Z})\) is not Abelian. (ii) the Euclidean isoperimetric inequality, which is sharp for circles, is qualitatively different from its hyperbolic analog for large areas. In this paper, a hyperbolic criterion for visibility of points and the structure of lattice points in rays have been investigated. The visible Gauss circle problem and the orchard problem have also been studied and some numerical data are given to illustrate the results.