Navigation of simulated mobile robots in the Webots Environment

Abstract

Development of various robotic algorithms can be supported with the aid of robot simulation programs. We present the Webots mobile robot simulator and its applicability to handle machine-learning methods. Furthermore, we show advantages and disadvantages of different navigation paradigms. We focus our investigation on metric navigation and especially on the creation and the usage of occupancy grids. We show a working method on how to use occupancy grid to efficiently navigate in Webots. Key words: mobile robotics, simulation, metric/topological navigation, cognitive map, occupancy grid.