arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1909.01957cs.DCcs.DScs.RO
收藏

移动机器人在全局通信模型中的离散性

Dispersion of Mobile Robots in the Global Communication Model

University of Illinois at Chicago(伊利诺伊大学芝加哥分校);Indian Statistical Institute(印度统计研究所);Kent State University(肯特州立大学)

Ajay D. Kshemkalyani, Anisur Rahaman Molla, Gokarna Sharma

摘要:

图上的分散问题要求最初任意放置在$n$ -节点匿名图的节点上的$k\leq n$机器人自主地重新定位以达到每个机器人在图的不同节点上的配置。这个问题是显着的兴趣,由于它的关系,其他基本的机器人协调问题,如探索,分散,负载平衡等。在本文中,我们考虑分散在{ \em全球通信}模型中的机器人可以与任何其他机器人在图(但图形是未知的机器人)。我们提供了三种新的确定性算法,两个任意图形和任意树,在一个同步设置,所有机器人在每个时间步执行他们的行动。对于任意图形,我们的第一个算法是基于DFS遍历,并保证$O(\min(m,kΔ))$步骤运行时使用$Θ(\log (\max(k,Δ)))$位在每个机器人,其中$m$是边缘的数量和$Δ$是图的最大程度。第二种算法是基于BFS遍历,并保证$O( \max(D,k) Δ(D+Δ))$步骤运行时使用$O(\max(D,Δ\log k))$位在每个机器人,其中$D$是图的直径。用于任意树的算法也基于BFS遍历,并在每个机器人处使用$O(\max(D,Δ\log k))$位来保证$O(D\max(D,k))$步骤运行时。我们的研究结果是显着的改进相比,现有的结果建立在{ \em本地通信}模型,其中一个机器人只能与其他机器人在同一个节点上进行通信。特别是,基于DFS的算法是最佳的内存和时间在常数度任意图。当$k\leq O(D)$ .

英文摘要:

The dispersion problem on graphs asks $k\leq n$ robots placed initially arbitrarily on the nodes of an $n$-node anonymous graph to reposition autonomously to reach a configuration in which each robot is on a distinct node of the graph. This problem is of significant interest due to its relationship to other fundamental robot coordination problems, such as exploration, scattering, load balancing etc. In this paper, we consider dispersion in the {\em global communication} model where a robot can communicate with any other robot in the graph (but the graph is unknown to robots). We provide three novel deterministic algorithms, two for arbitrary graphs and one for arbitrary trees, in a synchronous setting where all robots perform their actions in every time step. For arbitrary graphs, our first algorithm is based on a DFS traversal and guarantees $O(\min(m,kΔ))$ steps runtime using $Θ(\log (\max(k,Δ)))$ bits at each robot, where $m$ is the number of edges and $Δ$ is the maximum degree of the graph. The second algorithm for arbitrary graphs is based on a BFS traversal and guarantees $O( \max(D,k) Δ(D+Δ))$ steps runtime using $O(\max(D,Δ\log k))$ bits at each robot, where $D$ is the diameter of the graph. The algorithm for arbitrary trees is also based on a BFS travesal and guarantees $O(D\max(D,k))$ steps runtime using $O(\max(D,Δ\log k))$ bits at each robot. Our results are significant improvements compared to the existing results established in the {\em local communication} model where a robot can communication only with other robots present at the same node. Particularly, the DFS-based algorithm is optimal for both memory and time in constant-degree arbitrary graphs. The BFS-based algorithm for arbitrary trees is optimal with respect to runtime when $k\leq O(D)$.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑