Home // VEHICULAR 2018, The Seventh International Conference on Advances in Vehicular Systems, Technologies and Applications // View article

LiDAR-based SLAM Algorithm for Indoor Scenarios

Authors:
Felipe Jiménez
Miguel Clavijo
Javier Juana

Keywords: SLAM; LiDAR; Autonomous vehicle; detection; algorithm.

Abstract:
Simultaneous Localization and Mapping (SLAM) algorithms are one of the elements that have great relevance for autonomous driving in order to locate the vehicle, even in areas in which other methods have difficulties, or to improve the positioning given by these other systems. It also offers knowledge of the scenario in which the vehicle moves, information that can have multiple uses. There are several solutions to the SLAM problem using Light Detection and Ranging (LiDAR), but these algorithms require a high computational cost. However, certain environments with a specific structure allow the use of more simplified algorithms. Specifically, this paper shows a SLAM algorithm where only the LiDAR signal is used and vertical planes are taken as reference (perpendicular to the ground plane). This solution is quite effective in some scenarios, such as indoor parking areas. In addition, various alternatives are explored to increase the robustness of the results of positioning and mapping reconstruction. The algorithm has been tested in real scenarios with satisfactory results.

Pages: 47 to 53

Copyright: Copyright (c) IARIA, 2018

Publication date: June 24, 2018

Published in: conference

ISSN: 2327-2058

ISBN: 978-1-61208-643-9

Location: Venice, Italy

Dates: from June 24, 2018 to June 28, 2018