Buffon's needle landing near Besicovitch irregular self-similar sets (Q2874071)

From the abstract of the authors: In this paper, the authors got an upper estimate of the Favard length (sometimes called Buffon needle probability) of an arbitrary neighborhood of a big class of self-similar Cantor set of dimension 1. Considering \(L\) disjoint closed discs of radius \(1/L\) inside the unit disc, by using linear maps of the disc onto the smaller discs, the authors could generate a self-similar Cantor set \(\mathcal G\). Let \({\mathcal G}_n\) be the union of all possible images of the unit disc under \(n\)-fold compositions of the similarity maps. The question one may ask is at which rate the Favard length -- the average over all directions of the length of the orthogonal projection onto a line in that direction -- of these sets \({\mathcal G}_n\) decays to zero as a function of \(n\). In this paper, the authors proved the estimate Fav\(({\mathcal G}_n)\leq e^{-c\sqrt{\log n}}\). This estimate is vastly improved compared to the ones given in [\textit{Y. Peres} and \textit{B. Solomyak}, Pac. J. Math. 204, No. 2, 473--496 (2002; Zbl 1046.28006); \textit{T. Tao}, Proc. Lond. Math. Soc. (3) 98, No. 3, 559--584 (2009; Zbl 1173.28001)]. But it is worse than the power estimate Fav\(({\mathcal G}_n)\leq C/n^p\) proved for specific sets \({\mathcal G}_n\) with additional product structures in [\textit{F. Nazarov} et al., St. Petersbg. Math. J. 22, No. 1, 61--72 (2011) and Algebra Anal. 22, No. 1, 82--97 (2010; Zbl 1213.28006); \textit{I. Łaba} and \textit{K. Zhai}, Bull. Lond. Math. Soc. 42, No. 6, 997--1009 (2010; Zbl 1206.28004); \textit{M. Bond} et al., Am. J. Math. 136, No. 2, 357--391 (2014; Zbl 1318.28016)].