Journal of Frontiers of Computer Science and Technology ›› 2020, Vol. 14 ›› Issue (9): 1571-1579.DOI: 10.3778/j.issn.1673-9418.1909007

Previous Articles     Next Articles

Online Two-Robot Coordination Algorithm for Searching in Streets with Limited

WEI Qi, LINWeida, WU Tong, REN Yonggong   

  1. School of Computer and Information Technology, Liaoning Normal University, Dalian, Liaoning 116081, China
  • Online:2020-09-01 Published:2020-09-07

双弱感知能力机器人在线协作街道搜索算法

魏琦林韦达吴彤任永功   

  1. 辽宁师范大学 计算机与信息技术学院,辽宁 大连 116081

Abstract:

Searching for a target in an unknown environment is a basic task scenario of robots. It has a wide range of applications in real life. The related problem has received much attention in the study of computational geometry and robotics. This paper considers an online problem of searching in a street by two limited sensing robots. The street is defined as a simple polygon P with LR-visibility. Two robots equipped with limited sensing sensor start from the starting point s of P, and they can communicate with each other but have no previous knowledge of P. The goal of them is to search the ending point t of P with cooperation. The limited sensing sensor can only detect the discontinuities of the boundary of P in its visibility region, but cannot infer any other geometric information, such as distances or angles. Although the sensing capability is limited, the robot can provide low-cost solutions to challenging problems while achieving greater reliability in the face of uncertainties. Based on the geometry properties of the online searching model, this paper presents a 3-competitive online coordination algorithm and proves that it is optimal by showing a matching lower bound.

Key words: computational geometry, online coordination searching algorithm, competitive analysis, robot, limited sensing

摘要:

在未知环境中搜索一个目标,是机器人的基本任务场景之一,在现实中有着广泛的应用。相关问题在计算几何学科和机器人学科的研究中得到了广泛的关注。研究了一个双弱感知能力机器人在线协作搜索街道的问题。街道被定义为具有LR可视性的简单多边形P ,两个可相互通信的弱感知能力机器人从街道起点s出发,在预先不知道街道几何信息的前提下,协作完成街道终点t 的搜索。弱感知能力机器人携带的感应器仅能探测可视区域内街道边界的不连续情况,除此之外不能获取任何其他几何信息,如距离和角度。弱感知能力机器人虽然感知能力有限,却具有性价比高、面对不确定因素时可靠性强等优点。基于在线搜索模型的几何特征,提出了竞争比为3的在线协作搜索算法,并通过给出相匹配的竞争比下界,证明了算法的最优性。

关键词: 计算几何, 在线协作搜索算法, 竞争分析, 机器人, 弱感知