3D Mapping of an Unknown Environment by Cooperatively Mobile Robots

摘要

Abstract — Some of the important future applications of mobile robot systems are autonomous indoor and outdoor measurements of building, factories and objects in three dimensional view. A goal of such measurement is to provide us with a detailed map of the environment with the interesting characteristics. This map can then be transferred into a model, which represents the measuring objects. By following an automated proceeding, an autonomous measurement robot could be useful in order to extract independently the map and the model of the environment. The environmental model can then be directly used by the autonomous mobile robots for navigation. A further increase of the effectiveness in map production can be achieved, if the environment is mapped by several robots, or whole robot fleets. Such an approach can be DGYDQWDJHRXV for example, by reducing the exploration time. paper develops the framework for 3D mapping using multi-mobile robots. Two mobile robots equipped with different sensors capabilities are used. The main contribution of the paper is to autonomously build a 3D map of indoor environments within a good exploration time by using multi-mobile robots. Employing multiple autonomous robots with different types of sensors, two different algorithms are presented in this paper. The first is an algorithm for natural feature extraction using stereo camera in order to build a 3D feature-based map. The second is an algorithm to extract geometrical features from range images in order to build a 3D model of the environment. The algorithms and the framework are demonstrated on an experimental testbed that involves a team of two mobile robots. One of them is working as a master equipped with stereo camera. Whereas the second is involved as a slave equipped with a rotated laser scanner sensor.