Building 3D models is important in many applications, ranging from virtual visits of historical buildings, game and entertainment, to risk analysis in partially collapsed buildings. This task is performed at different scales: city, buildings, indoor environments, objects and using different sensors: cameras, 2D and 3D laser, etc. Moreover, different map representation have been considered: metric (or feature based) 3D maps represented as a set of 3D points (plus color information), in contrast with maps represented as a set of semantic structural elements (i.e., floors, walls, steps, stairs, etc.). In this paper we describe an approach to generate visually realistic 3D maps formed by semantic structural elements. The approach is suitable for indoor environments and integrates three different sensors: a 2D laser range finder, a stereo camera, an inertial measurement unit (IMU). Data acquisition is automatically performed by an autonomous mobile robot mounting such sensors on board. Model building is then achieved by using 2D SLAM techniques for building 2D consistent maps, stereo vision and inertial navigation system to detect semantic elements. Stereo vision is also used to extract textures of such elements. While the main objective of our research is to represent a 3D map as a set of 3D elements with their appropriate texture, and to use a generative model to build the map starting from such primitives, in this paper we outline a system doing this task and present some experiments to evaluate the impact of human interaction in the modelling process on increasing the semantic modelling and the visual realism of the maps. 1

The need for improving the robustness, as well as the ability to adapt to different operational conditions, is a key requirement for a wider deployment of robots in many application domains. In this paper, we present an approach to the design of robotic systems, that is based on the explicit representation of knowledge about context. The goal of the approach is to improve the system performance, by dynamically tailoring the functionalities of the robot to the specific features of the situation at hand. While the idea of using contextual knowledge is not new, the proposed approach generalizes previous work and its advantages are discussed through a case study including several experiments. In particular, we identify many attempts to use contextual knowledge in several basic functionalities of a mobile robot such as: behaviors, navigation, exploration, localization, mapping and perception. We then show how re-designing our mobile platform with a common representation of contextual knowledge, leads to interesting improvements in many of the above mentioned components, thus achieving greater flexibility and robustness in the face of different situations. Moreover, a clear separation of contextual knowledge leads to a design methodology, which supports the design of small specialized system components instead of complex self-contained subsystems. Key words: contextual knowledge and reasoning, cognitive robotics, system architecture PACS: