Abstract
This paper presents a Wearable Prototype for indoor mapping developed by the University of Vigo. The system is based on a Velodyne LiDAR, acquiring points with 16 rays for a simplistic or low-density 3D representation of reality. With this, a Simultaneous Localization and Mapping (3D-SLAM) method is developed for the mapping and generation of 3D point clouds of scenarios deprived from GNSS signal. The quality of the system presented is validated through the comparison with a commercial indoor mapping system, Zeb-Revo, from the company GeoSLAM and with a terrestrial LiDAR, Faro Focus3D X330. The first is considered as a relative reference with other mobile systems and is chosen due to its use of the same principle for mapping: SLAM techniques based on Robot Operating System (ROS), while the second is taken as ground-truth for the determination of the final accuracy of the system regarding reality. Results show that the accuracy of the system is mainly determined by the accuracy of the sensor, with little increment in the error introduced by the mapping algorithm.
Highlights
The generation of 3D models and maps in building interiors is a task of increasing interest in different fields, from accessibility to energy efficiency and assistance in navigation through big buildings such as hospitals and commercial centers
We present a wearable backpack-based indoor mapping prototype developed at the University of Vigo
The features are submitted to an iterative matching process, which is performed per line of the scan, as it is an extrapolation of the procedure performed for 2D LiDAR sensors that only measure one line per acquisition through repetition of the procedure 16 times, one per ray or channel of the Velodyne sensor integrated in the mobile platform
Summary
The generation of 3D models and maps in building interiors is a task of increasing interest in different fields, from accessibility to energy efficiency and assistance in navigation through big buildings such as hospitals and commercial centers. The first robots were dedicated to movement indoors, medical applications appeared soon, incorporating the requirements of reduction of sizes for sensors and electronics [3] and adaptability to the human body In this way, wearable devices were developed, with the main aim of contributing to mobility of humans with disabilities and providing information on their positioning [4]. The subset approach is applied for the indirect evaluation of the trajectory through the analysis of the shape of the point cloud in areas where the trajectory deviations could have an important effect, such as in the entrances in new spaces involving turns With this aim, the wearable prototype is compared to both mobile and static commercial systems, Zeb-Revo [17] and Faro Focus3D X330 [18].
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
