Application of the analytic hierarchy process on automatic exploration method for a mobile robot working environment map building

In this paper, an automatic Simultaneous Localization and Mapping (SLAM) method is studied. This method is based on Grid-based FastSLAM and frontier-based exploration. Firstly, the Grid-based FastSLAM generates an occupied grid map that is intuitive and that easily generates the frontiers that are required for frontier-based exploration. To execute a SLAM, the robot must move to a frontier to explore the unknown area in order to update the map. Before the robot selects a frontier as a sub-target, it has to select a frontier for efficient map building. In this research, a method for selecting a frontier among several candidates is suggested by means of the Analytic Hierarchy Process, which is a multi-objective decision making method. In the simulations, several cases are studied to demonstrate the performances of the proposed method and the results are compared with conventional frontier selection method.