Path-Planning Algorithms for a Levitating Planar Motion System
(2025)Department of Automatic Control
- Abstract
- Multi-agent pathfinding (MAPF) is becoming increasingly interesting because of its role in complex systems, such as warehouse sorting, search-and-rescue missions and autonomous vehicles. Another application for path planning is within production in industry when more advanced solutions than conveyor belts are developed. One such advanced transportation system is the XPlanar system, which is produced by Beckhoff Automation. In cases of unexpected shut downs of the system, a start up algorithm is run which implements MAPF.
This thesis investigates and compares two algorithms through simulations and experiments. The algorithms are TaN-A*, that is A* path planning with a Conflict Based Search (CBS), and a D*-Lite algorithm with local... (More) - Multi-agent pathfinding (MAPF) is becoming increasingly interesting because of its role in complex systems, such as warehouse sorting, search-and-rescue missions and autonomous vehicles. Another application for path planning is within production in industry when more advanced solutions than conveyor belts are developed. One such advanced transportation system is the XPlanar system, which is produced by Beckhoff Automation. In cases of unexpected shut downs of the system, a start up algorithm is run which implements MAPF.
This thesis investigates and compares two algorithms through simulations and experiments. The algorithms are TaN-A*, that is A* path planning with a Conflict Based Search (CBS), and a D*-Lite algorithm with local conflict avoidance. It was found that the D*-Lite implementation has lower computational times than the TaNA* algorithm, but the A* implementation was more robust. The findings led to the development of a final algorithm that was a hybrid of the two algorithms, that generates conflict-free paths for all agents without failures in all of the tested scenarios. Moreover, the final algorithm demonstrates competitive execution times compared to the other methods analysed. This final implementation was suggested to be used on the XPlanar system. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/9186817
- author
- Rehme, Louise
- supervisor
- organization
- year
- 2025
- type
- H3 - Professional qualifications (4 Years - )
- subject
- report number
- TFRT-6270
- other publication id
- 0280-5316
- language
- English
- id
- 9186817
- date added to LUP
- 2025-05-09 08:21:17
- date last changed
- 2025-05-09 08:21:17
@misc{9186817,
abstract = {{Multi-agent pathfinding (MAPF) is becoming increasingly interesting because of its role in complex systems, such as warehouse sorting, search-and-rescue missions and autonomous vehicles. Another application for path planning is within production in industry when more advanced solutions than conveyor belts are developed. One such advanced transportation system is the XPlanar system, which is produced by Beckhoff Automation. In cases of unexpected shut downs of the system, a start up algorithm is run which implements MAPF.
This thesis investigates and compares two algorithms through simulations and experiments. The algorithms are TaN-A*, that is A* path planning with a Conflict Based Search (CBS), and a D*-Lite algorithm with local conflict avoidance. It was found that the D*-Lite implementation has lower computational times than the TaNA* algorithm, but the A* implementation was more robust. The findings led to the development of a final algorithm that was a hybrid of the two algorithms, that generates conflict-free paths for all agents without failures in all of the tested scenarios. Moreover, the final algorithm demonstrates competitive execution times compared to the other methods analysed. This final implementation was suggested to be used on the XPlanar system.}},
author = {{Rehme, Louise}},
language = {{eng}},
note = {{Student Paper}},
title = {{Path-Planning Algorithms for a Levitating Planar Motion System}},
year = {{2025}},
}