ProbXiv
Explorersign in
Problem archiveProblem record

Statement

For a compact convex set P⊂R2P \subset \mathbb{R}^2 with nonempty interior, and a point xx in the exterior R2∖P\mathbb{R}^2 \setminus P, call a point m∈∂Pm \in \partial P a right support point for xx if the line through xx and mm is a support line of PP and PP lies in the closed half-plane to the right of the ray directed from xx through mm. If this right support point mm is unique, define the outer billiards map by

F(x)=2m−x,F(x) = 2m - x,

so that mm is the midpoint of the segment from xx to F(x)F(x); otherwise F(x)F(x) is undefined. The map FF is injective where defined, and F−1F^{-1} is obtained by replacing "right" with "left" in the definition. A point x∈R2∖Px \in \mathbb{R}^2 \setminus P has a well-defined orbit if Fn(x)F^n(x) is defined for all n∈Zn \in \mathbb{Z}, and in that case its orbit is O(x)={Fn(x):n∈Z}O(x) = \{F^n(x) : n \in \mathbb{Z}\}. The orbit is periodic if Fn(x)=xF^n(x) = x for some n≥1n \geq 1, and bounded if O(x)O(x) is contained in some bounded subset of R2\mathbb{R}^2, and unbounded otherwise. When PP is a convex polygon, the right support point of xx is unique unless the support line through xx contains an edge of PP; the set of points of R2∖P\mathbb{R}^2 \setminus P without well-defined orbit has planar Lebesgue measure zero.

For n≥3n \geq 3, identify a tuple v=(v1,…,vn)∈(R2)n≅R2nv = (v_1, \dots, v_n) \in (\mathbb{R}^2)^n \cong \mathbb{R}^{2n} with the polygon P(v)P(v) having vertices v1,…,vnv_1, \dots, v_n, and let Cn⊂R2nC_n \subset \mathbb{R}^{2n} be the set of tuples whose points are in convex position and listed in cyclic order, so that P(v)P(v) is a convex nn-gon.

Then the following two propositions hold.

  1. For every n≥3n \geq 3 and Lebesgue-almost every v∈Cnv \in C_n, there exists a point x∈R2∖P(v)x \in \mathbb{R}^2 \setminus P(v) with a well-defined orbit that is unbounded.

  2. For every n≥3n \geq 3, if PP is a regular nn-gon in R2\mathbb{R}^2, then Lebesgue-almost every point x∈R2∖Px \in \mathbb{R}^2 \setminus P has a well-defined orbit that is periodic.

Record

Added
Links

Comments

Nothing has been published against this problem yet, and nothing has been said about it.

Sign in with an institutional address to take part in the discussion. Reading every thread stays open to everyone.

Sign in

Solve with an agent

Open the statement in a chat, with the problem and the ground rules already written into the prompt.

This opens a third-party site. Nothing is posted back to ProbXiv and nothing you write there is recorded here — what a model gives you is an attempt, which a person still has to check.