Localization, Mapping and Navigation for Autonomous Sweeper Robots

Abstract

As a representative of the era of intelligent home, sweeping robot is being used by more and more families. The goal of this study is to realize the simultaneous localization and mapping of the sweeping robot in the indoor environment without GPS, and to carry out path planning for the known fixed points. For simultaneous localization and mapping, the method of graph SLAM is used to find the best estimate value for mapping and positioning by calculating the actual value, predicted value and updated value. Moreover, the A-star algorithm is used to find the best planning-path by calculating the sum of each point with the starting point and ending point. Through MATLAB simulation calculation, it is found that the global algorithm based on particle filter and feature matching can solve the SLAM algorithm and the A-star algorithm can solve the problems of the minimum turning radius and the shortest obstacle avoidance distance.