Search results

Aims to make a mobile robot able to build accurate 2D and 3D models of its environment while navigating autonomously.

2D map building is performed using a laser range scanner. The map is used by the robot to both localize itself and recognize places already explored. This is the well‐known simultaneous localization and mapping (SLAM) problem. 3D model reconstruction, instead, uses computer vision techniques based on feature extraction and matching.

The experimental results illustrate the validity and accuracy of the reconstructed maps of the environment and enable the robot to navigate autonomously in indoor environments, such as museums, hospitals, airports, offices and so on. Such a robot can play a major role in different tasks such as surveillance, image‐based rendering, remote fruition of hardly accessible sites, monitoring and maintenance applications, reverse engineering in construction. In these areas accurate 3D models in addition to 2D maps can convey a lot of very useful information.

The main contribution of the paper is an interesting integration of different algorithms in an experimental platform that performs 2D map building using a laser range scanner, autonomous navigation and 3D reconstruction of the areas of particular interest.