AutoNav in C-L-U-E: A Baseline Autonomous Software Stack for Autonomous Navigation in Closed Low-Speed Unstructured Environments

Mohamed Sabry, Amr Farag, Bassem Magued, Ahmed Mazhr, Amr El Mougy, Slim Abdennadher

2024

Abstract

The development of Autonomous systems modules has been growing exponentially within the past few years with various complex approaches. Most of these systems have some restrictions or dependencies on numerous inputs. There are two main categories of these systems, Highway and Urban Road-Vehicle autonomous systems, and short-distance autonomous platforms. The short-distance category includes minipods and golfcars that operate in closed environments such as residential compounds or university campuses. Various challenges have been identified in both categories. A challenge example for Highways / Urban areas is controlling the vehicle’s motion on high and moderate speeds. However, for closed campuses, the challenge is mainly in maneuvering around high density pedestrians moving with low speeds and being able to avoid low pavements and obstacles that may damage the platform, such as potholes. For this matter and given the increasing complexity of modules-in-development, this paper proposes a low-complexity baseline map-less autonomous software stack with a perception module capable of navigating closed campuses within unstructured environments. The system is a simple one that requires 1 - 2 LiDARs as well as an input route to follow, which is inserted by the user from offline Open Street Maps (OSM) data. The system runs fully on-board on a consumer grade PC without the need for internet connectivity and has been tested successfully in various scenarios on campus at the German University in Cairo (GUC), Egypt. The tests included pedestrian and obstacle avoidance as well as emergency stopping with the capability of resuming and the following the preset global path before departure. The proposed system is based on the golf-car platform at the GUC.

Paper Citation

in Harvard Style

Sabry M., Farag A., Magued B., Mazhr A., El Mougy A. and Abdennadher S. (2024). AutoNav in C-L-U-E: A Baseline Autonomous Software Stack for Autonomous Navigation in Closed Low-Speed Unstructured Environments. In Proceedings of the 16th International Conference on Agents and Artificial Intelligence - Volume 1: ICAART; ISBN 978-989-758-680-4, SciTePress, pages 189-197. DOI: 10.5220/0012295500003636

in Bibtex Style

@conference{icaart24,
author={Mohamed Sabry and Amr Farag and Bassem Magued and Ahmed Mazhr and Amr El Mougy and Slim Abdennadher},
title={AutoNav in C-L-U-E: A Baseline Autonomous Software Stack for Autonomous Navigation in Closed Low-Speed Unstructured Environments},
booktitle={Proceedings of the 16th International Conference on Agents and Artificial Intelligence - Volume 1: ICAART},
year={2024},
pages={189-197},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012295500003636},
isbn={978-989-758-680-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 16th International Conference on Agents and Artificial Intelligence - Volume 1: ICAART
TI - AutoNav in C-L-U-E: A Baseline Autonomous Software Stack for Autonomous Navigation in Closed Low-Speed Unstructured Environments
SN - 978-989-758-680-4
AU - Sabry M.
AU - Farag A.
AU - Magued B.
AU - Mazhr A.
AU - El Mougy A.
AU - Abdennadher S.
PY - 2024
SP - 189
EP - 197
DO - 10.5220/0012295500003636
PB - SciTePress