Deterministic byzantine-resilient robot gathering
Loading...
Date
2020-08
Authors
Saha, Ullash
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Gathering is a fundamental problem in the field of mobile robot algorithms. The goal is to gather the robots at a single point, which can be useful for sharing information or coordinating future actions. In this thesis, we consider models in which m robots are placed at nodes in a graph of size n, and the goal is to gather them at a single node of the graph. Here, each robot has an identifier (ID) and runs its own deterministic algorithm, i.e., there is no centralized coordinator. We consider a particularly challenging scenario: there are f maliciously faulty robots in the team that can behave arbitrarily, and even have the ability to change their IDs to any value at any time. There is no way to distinguish these robots from non-faulty robots, other than perhaps observing strange or unexpected behaviour. In this thesis, we seek to design a deterministic algorithm that is run by each robot to eventually have all the non-faulty robots located at the same node in the same round. It is known that no algorithm can solve this task unless there at least f+1 non-faulty robots in the team. In this thesis, we design an algorithm that runs in polynomial time with respect to n and m that matches this bound, i.e., it works in a team that has exactly f+1 non-faulty robots. In our model, we have equipped the robots with sensors that enable each robot to see the subgraph (including robots) within some distance H of its current node. We also prove that the gathering task is solvable if this visibility range H is at least the radius of the graph, and not solvable if H is any fixed constant.
Description
Keywords
Robot gathering, Byzantine faults, Visibility, Graphs, Distributed algorithms
Citation
Miller, A., Saha, U. Fast Byzantine Gathering with Visibility in Graphs. In press of the 16th International Symposium on Algorithms and Experiments for Wireless Sensor Networks, Algosensors 2020.