Map Management Approach for SLAM in Large-Scale Indoor and Outdoor Areas

Abstract

This work presents a map management approach for various environments by creating multiple maps with different SLAM (simultaneous localization and mapping) configurations. A modular map structure allows to add, modify or delete maps without influencing other maps of different areas. The hierarchy level of our algorithm is above the utilized SLAM method, since it is able to automatically trigger new maps (e.g. after the detection of passing a doorway). The appropriate SLAM configuration for the next map is chosen by evaluating laser scan data. Single independent maps are connected by link-points which are located in an overlapping zone of both maps, enabling global navigation over several maps. Loopclosures between maps are detected by an appearance-based method using feature matching and ICP registration between point clouds. The number of possible loop-closure locations is limited to the number of link-points. Based on the arrangement of maps and link-points, a topological graph is extracted for navigation purposes and tracking the global robot's position over several maps. Our approach is evaluated by mapping a university campus with multiple indoor and outdoor areas and abstracting a metrical-topological graph. It is also compared to a single map running with different SLAM configurations. Our approach enhances the overall map quality compared to the single map approaches by automatically choosing appropriate SLAM configurations for different environmental setups.

Machine Learning Based Indoor Localization Using Wi-Fi RSSI Fingerprints: An Overview

“…Due to its high processing cost, it may not be appropriate to run on resource-constrained smart devices such as IoT devices and mini drones. Map management methods (MMMs) [96], such as pedestrian deadreckoning and map filtering, are often used to locate the device's starting point and optimize its output accordingly. However, due to high computational costs, their implementation in resource-constrained smart devices is limited.…”

Have I been here before? Learning to Close the Loop with LiDAR Data in Graph-Based SLAM

“…IV-B). The mobile service robot Sobi [23] is used for both validations, which is a ROS-based information and guidance system equipped with a differential drive base (Neobotix MP-500), two RGBD cameras (Intel Realsense D435, front and back) and a 3D LiDAR (Velodyne VLP-16). An extended Kalman filter [24] is used to fuse the wheel odometry with the IMU data (XSens MTi-30).…”

Sobi: An Interactive Social Service Robot for Long-Term Autonomy in Open Environments

“…Since the different types of environment place different demands on the SLAM system, we developed a method that automatically selects predefined SLAM configurations for different environments depending on various criteria, such as distances to walls. For example, for outdoor or large-scale environments the depth information of the 3D Lidar is used instead of the cameras, and for indoor environments the maximum range for map generation is reduced [21].…”