Stairs and Doors Recognition as Natural Landmarks Based on Clouds of 3D Edge-Points from RGB-D Sensors for Mobile Robot Localization

Abstract

Natural landmarks are the main features in the next step of the research in localization of mobile robot platforms. The identification and recognition of these landmarks are crucial to better localize a robot. To help solving this problem, this work proposes an approach for the identification and recognition of natural marks included in the environment using images from RGB-D (Red, Green, Blue, Depth) sensors. In the identification step, a structural analysis of the natural landmarks that are present in the environment is performed. The extraction of edge points of these landmarks is done using the 3D point cloud obtained from the RGB-D sensor. These edge points are smoothed through the Sl0 algorithm, which minimizes the standard deviation of the normals at each point. Then, the second step of the proposed algorithm begins, which is the proper recognition of the natural landmarks. This recognition step is done as a real-time algorithm that extracts the points referring to the filtered edges and determines to which structure they belong to in the current scenario: stairs or doors. Finally, the geometrical characteristics that are intrinsic to the doors and stairs are identified. The approach proposed here has been validated with real robot experiments. The performed tests verify the efficacy of our proposed approach.

A Versatile Method for Depth Data Error Estimation in RGB-D Sensors

“…Because no measurement device is entirely accurate, errors in depth measurements of a sensed object are prone to increase with the distance to the object. Ultimately, this leads to general errors affecting the captured geometry of the scene deteriorating, thus, the performance of computer vision applications, such as visual odometry [ 43 , 44 ] and object recognition [ 16 ].…”

A software tool for the semi-automatic segmentation of architectural 3D models with semantic annotation and Web fruition

“…Stairs detection is important in several different fields, such as multi-storey path finding for explorer robots venturing into buildings [13][14][15], as an aid for the visually impaired [16][17][18], and last but not least in the semantic segmentation of 3D models of Architectural Heritage buildings [19,20]. Algorithms developed for the first two cases are usually not directly applicable to semantic segmentation for CH, [15] because they generally work on organized point clouds derived from RGB-D data, while CH applications use unorganized clouds coming from photogrammetry or laser scanners.…”

“…Some start from edge point extraction. For example in [13] edge points are detected in RGB-D images by the Concave Hull algorithm available in the PCL framework, then classification relies on depth and geometric information. In paper [14], a staircase extraction algorithm based on super-voxels is proposed.…”

A Maximum Feasible Subsystem for Globally Optimal 3D Point Cloud Registration

“…Despite significant research efforts in recent decades, solving the 3D Euclidean registration between two point clouds remain challenging when the rough transformation between point cloud sets is unknown or there is severe noise and outliers in the point cloud sets. This issue is closely related to several areas, including 3D scene reconstruction [ 1 ], object instance recognition [ 2 ], 3D pose estimation [ 3 ], simultaneous location and mapping (SLAM) [ 4 ], digital cultural heritage [ 5 ], and robot grasping systems [ 6 ].…”