Analysis of Multi-Robot Patrolling Algorithms

Abstract

This article is dedicated to analyzing the common problems of robots when patrolling indoors and the corresponding solution algorithms. In the article, four problems of patrol robots (whether centralized or non-centralized) during operation are listed: the overload of data to be processed when the robotic sensors are transmitting the data, the repetitiveness and simultaneity of the robot work, the incorrect execution of the algorithm and how to deal with the inevitable unknown external factors. The simultaneity of the robots is to be eliminated because robots need to explore independently in the patrol task. Subsequently, four algorithms and functionalities are introduced (Monte Carlo Tree Planning Algorithm, Rapid Exploring Random Tree Planning, Mutual Exclusion Algorithm, and Kalman Filter Algorithm). The question of how the four algorithms address the corresponding factors will be illustrated by an elaboration of the process by which the algorithms solve the problem.