SearcharxivSearch

arXiv subjects

Rodrigo G. Schaefer

Publications and source records attributed to Rodrigo G. Schaefer.

4 recordsLinked to original sources

No infinite spin for partial collisions converging to isolated central configurations on the plane

In the $n$-body problem, when a~cluster of bodies tends to a collision, then its normalized shape curve converges to the set of normalized central configurations, which has $SO(2)$ symmetry in the planar case. This leaves a possibility that the normalized shape curve tends to the circle obtained by rotation of some central configuration instead of a particular point on it. This is the \emph{infinite spin problem} which concerns the rotational behavior of total collision orbits in the $n$-body problem. The question also makes sense for partial collision. We show that the infinite spin is not possible if the limiting circle is isolated from other connected components of the set of normalized central configurations. Our approach extends the method from recent work for total collision by Moeckel and Montgomery, which was based on a combination of the center manifold theorem with {\L}ojasiewicz inequality. To that we add a shadowing result for pseudo-orbits near normally hyperbolic manifold and careful estimates on the influence of other bodies on the cluster of colliding bodies.

math.DS

Arnold diffusion for an a priori unstable Hamiltonian system with 3 $+$ 1/2 degrees of freedom

In the present paper we apply the geometrical mechanism of diffusion in an \emph{a priori} unstable Hamiltonian system with 3 $+$ 1/2 degrees of freedom. This mechanism consists of combining iterations of the \emph{inner} and \emph{outer} dynamics associated to a \emph{Normallly Hyperbolic Invariant Manifold} (NHIM), to construct diffusing \emph{pseudo-orbits} and subsequently apply shadowing results to prove the existence of diffusing orbits of the system. In addition to proving the existence of diffusion for a wide range of the parameters of the system, an important part of our study focuses on the search for \emph{Highways}, a particular family of orbits of the outer map (the so-called \emph{scattering} map), whose existence is sufficient to ensure a very large drift of the action variables, with a diffusion time near them that agrees with the optimal estimates in the literature. Moreover, this optimal diffusion time is calculated, with an explicit calculation of the constants involved. All these properties are proved by analytical methods and, where necessary, supplemented by numerical calculations.

math.DS

Arnold diffusion for a complete family of perturbations with two independent harmonics

We prove that for any non-trivial perturbation depending on any two independent harmonics of a pendulum and a rotor there is global instability. The proof is based on the geometrical method and relies on the concrete computation of several scattering maps. A complete description of the different kinds of scattering maps taking place as well as the existence of piecewise smooth global scattering maps is also provided.

math.DS

Arnold diffusion for a complete family of perturbations

In this work we illustrate the Arnold diffusion in a concrete example---the \emph{a priori} unstable Hamiltonian system of $2+1/2$ degrees of freedom $H(p,q,I,φ,s) = p^{2}/2+\cos q -1 +I^{2}/2 + h(q,φ,s;\varepsilon)$---proving that for \emph{any} small periodic perturbation of the form $h(q,φ,s;\varepsilon) = \varepsilon\cos q\left( a_{00} + a_{10}\cosφ+ a_{01}\cos s \right)$ ($a_{10}a_{01} \neq 0$) there is global instability for the action. For the proof we apply a geometrical mechanism based in the so-called Scattering map. This work has the following structure: In a first stage, for a more restricted case ($I^*\thicksimπ/2μ$, $μ= a_{10}/a_{01}$), we use only one scattering map, with a special property: the existence of simple paths of diffusion called highways. Later, in the general case we combine a scattering map with the inner map (inner dynamics) to prove the more general result (the existence of the instability for any $μ$). The bifurcations of the scattering map are also studied as a function of $μ$. Finally, we give an estimate for the time of diffusion, and we show that this time is primarily the time spent under the scattering map.

math.DS