Completing Robot Maps by Predicting the Layout of Rooms Behind Closed Doors

Abstract

The availability of maps of indoor environments is often fundamental for autonomous mobile robots to efficiently operate in industrial, office, and domestic applications. When robots build such maps, some areas of interest could be inaccessible, for instance, due to closed doors. As a consequence, these areas are not represented in the maps, possibly limiting the activities robots can perform. In this paper, we provide a method that completes 2D grid maps by adding the predicted layout of the rooms behind closed doors. The main idea of our approach is to exploit the underlying geometrical structure of indoor environments to estimate the shape of unobserved rooms. Results show that our method is accurate in completing maps also when large portions of environments cannot be accessed by the robot during map building.

Mapping beyond what you can see: Predicting the layout of rooms behind closed doors

“…The planning is performed by solving a Traveling Salesperson Problem (TSP) on a graph derived using a Voronoi segmentation of the completed map of the environment, after the inpaiting of the predicted layout of closed rooms. This paper extends our previous work [8]. Specifically, we add the application of our approach to planning a coverage path (Section 5).…”

Reconstruction and prediction of the layout of indoor environments from two-dimensional metric maps

“…An example of a possible online use of our method is provided in Section 4.5 and detailed in [8]. Moreover, an adaptation of our method to predict the layout of completely unexplored rooms that are behind closed doors is presented in [46]. Finally, in [47] our method is embedded in a framework for detecting robust features in indoor maps in order to evaluate map quality.…”

Robust Structure Identification and Room Segmentation of Cluttered Indoor Environments from Occupancy Grid Maps

“…Nevertheless, we show in [6] how such predicted structural knowledge could be used online to improve performance in exploration for map building. Another application is in [30], that shows how the identification of the structure of the environment can help to predict the shape of multiple closed rooms that are behind closed doors, then performing inpainting on the map of the predicted shape of unobserved closed rooms. III.…”