Abstract
In recent times, the use of autonomous mobile robots (AMRs) has increased in many industries. As more AMRs roam the same area, traffic management becomes essential to prevent congestion and deadlocks. In related work, traffic management is often achieved using sophisticated, centralised planning approaches, albeit this often suffers from scalability issues. This paper therefore explores an approach where multi-AMR traffic management is achieved with layered costmaps and a modified Dijkstra's algorithm. This keeps the global path planner in the individual AMRs, thus not suffering scalability issues. To achieve multi-AMR coordination, traffic lanes and restricted areas are added to the AMRs' global costmaps. Furthermore, the AMRs also use a modified Dijkstra's algorithm that supports implmentation of traffic directions. A proof-of-concept solution is implemented in Robot Operating System 2 with Nav2 and Gazebo. The implemented solution was tested against a standard solution without any traffic management in three scenarios designed to provoke collisions. The results indicate that the implemented solution can prevent a set of collisions better than one without traffic management.
Originalsprog | Engelsk |
---|---|
Tidsskrift | Procedia Computer Science |
Vol/bind | 232 |
Sider (fra-til) | 53-63 |
Antal sider | 11 |
ISSN | 1877-0509 |
DOI | |
Status | Udgivet - 2024 |
Begivenhed | 5th International Conference on Industry 4.0 and Smart Manufacturing, ISM 2023 - Lisbon, Portugal Varighed: 22 nov. 2023 → 24 nov. 2023 |
Konference
Konference | 5th International Conference on Industry 4.0 and Smart Manufacturing, ISM 2023 |
---|---|
Land/Område | Portugal |
By | Lisbon |
Periode | 22/11/2023 → 24/11/2023 |
Bibliografisk note
Publisher Copyright:© 2024 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0)