LiDAR-Based Augmented Reality for the Development of Test Scenarios on Safety for Autonomous Operation of a Shunting Locomotive

  • As public interest in autonomous driving systems grows, safety is becoming a critical issue. Extensive testing is therefore required before these systems can be deployed in real-world traffic. Simulation-based testing has proven to be a valuable tool for evaluating autonomous systems, but there remains a gap between simulation and real-world testing. A system may be well tested in simulation, but in real-world testing it may encounter sudden events that result in unpredictable behavior, and equipment may be damaged in the event of a malfunction. This paper describes a novel augmentation interface for Light Detection And Ranging (LiDAR) sensor data that aims to bridge this gap. The interface is designed to generate realistic test scenarios on live data streams, making it ideal for investigating special and borderline cases. The interface has been developed for the Robot Operating System (ROS) using the Point Cloud Library and is intended to be used for testing an autonomous shunting locomotive. With this interface, a data stream from aAs public interest in autonomous driving systems grows, safety is becoming a critical issue. Extensive testing is therefore required before these systems can be deployed in real-world traffic. Simulation-based testing has proven to be a valuable tool for evaluating autonomous systems, but there remains a gap between simulation and real-world testing. A system may be well tested in simulation, but in real-world testing it may encounter sudden events that result in unpredictable behavior, and equipment may be damaged in the event of a malfunction. This paper describes a novel augmentation interface for Light Detection And Ranging (LiDAR) sensor data that aims to bridge this gap. The interface is designed to generate realistic test scenarios on live data streams, making it ideal for investigating special and borderline cases. The interface has been developed for the Robot Operating System (ROS) using the Point Cloud Library and is intended to be used for testing an autonomous shunting locomotive. With this interface, a data stream from a 3D LiDAR can be augmented with any given object represented by a point cloud, allowing for the use of data from both simulation and real-world environments.show moreshow less

Export metadata

Additional Services

Search Google Scholar
Metadaten
Author:Niklas Kohlisch, Philipp Koch, Stefan MayORCiD
DOI:https://doi.org/10.1109/ICARSC58346.2023.10129540
Parent Title (English):2023 IEEE International Conference on Autonomous Robot Systems and Competitions (ICARSC)
Publisher:IEEE
Document Type:conference proceeding (article)
Language:English
Reviewed:Begutachtet/Reviewed
Release Date:2025/02/18
Pagenumber:6
First Page:23
Last Page:28
institutes:Institut für Angewandte Wasserstoffforschung, Elektro- und Thermochemische Energiesysteme (H2Ohm)
Research Themes:Digitalisierung & Künstliche Intelligenz
Verstanden ✔
Diese Webseite verwendet technisch erforderliche Session-Cookies. Durch die weitere Nutzung der Webseite stimmen Sie diesem zu. Unsere Datenschutzerklärung finden Sie hier.