EXTRACTING TOPOLOGICAL INFORMATION FROM GRID MAPS FOR ROBOT NAVIGATION

Abstract

In many robotic navigation-related tasks, abstracting the environment where mobile robots carry out some mission can be of a great benefit. In particular, extracting a simple topological graph-like representation from a more complex and detailed metric map is often required for path-planning and navigation. In this work, an approach to perform such extraction in grid maps is presented. The focus is not only on obtaining a diagram or visual representation of possible paths, but also to propose a new way to obtain graph data information related to the connectivity of the environment that can be passed to robots or to a centralized planner, in order to assist the navigation task. The approach proposed is based on image processing techniques. Simulation results prove its simplicity, accuracy and efficiency.