当前位置: X-MOL 学术Comput. Geom. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
On the approximation of shortest escape paths
Computational Geometry ( IF 0.4 ) Pub Date : 2020-09-02 , DOI: 10.1016/j.comgeo.2020.101709
David Kübel , Elmar Langetepe

A hiker is lost in a forest of unknown shape. What is a good path for the hiker to follow in order to escape from the forest within a reasonable amount of time? The hiker's dilemma clearly is: Should one start exploring the area close-by and expand the search radii gradually? Or should one rather pick some direction and run straight on?

We employ a competitive analysis to prove that a certain spiral strategy achieves a reasonable competitive factor for the case where the forest has a non-empty kernel; moreover, if the hiker's unknown starting position lies in the kernel of the forest, this strategy is (almost) optimal w.r.t. the competitive factor. As a basis for our competitive analysis, we introduce a new measure of intrinsic complexity for instances of this escape problem, which we compare to several known shortest escape paths.



中文翻译:

关于最短逃生路径的近似

徒步旅行者迷失在未知形状的森林中。远足者在合理的时间内逃离森林的最佳途径是什么?徒步旅行者的困境显然是:一个人应该开始探索附近区域并逐渐扩大搜索半径吗?还是应该选择某个方向并继续前进?

我们通过竞争分析来证明对于森林具有非空核的情况,某种螺旋策略可以实现合理的竞争因素。此外,如果徒步旅行者的未知起始位置位于森林的内核中,则该策略(几乎)是竞争因素的最佳选择。作为我们竞争分析的基础,我们针对这种逃生问题的实例引入了一种新的内在复杂性度量,将其与几种已知的最短逃生路径进行了比较。

更新日期:2020-09-21
down
wechat
bug