Intelligent Hybrid Approaches for Mobile Robots Path Planning

Lin, Shiwei

Intelligent Hybrid Approaches for Mobile Robots Path Planning

Lin, Shiwei

Permalink

Publication Type:: Thesis
Issue Date:: 2023

Open Access

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is open access.

Adobe PDF

Download thesisAdobe PDF (6.89 MB)

View statistics

Full metadata record

Field	Value	Language
dc.contributor.author	Lin, Shiwei
dc.date.accessioned	2023-09-06T04:27:38Z
dc.date.available	2023-09-06T04:27:38Z
dc.date.issued	2023
dc.identifier.uri	http://hdl.handle.net/10453/171955
dc.description	University of Technology Sydney. Faculty of Engineering and Information Technology.	en_US.UTF-8
dc.description.abstract	The practical applications of mobile robots are widely implemented in various areas, such as education, industry, environment, and civil applications. The requirements of robots' navigation are one of the primary considerations for autonomous operation. Path planning is essential for the successful application of mobile robots. Considering all available information, it aims to generate the robot's optimal path from the start to the target location. Depending on the operation scenarios, various factors are counted in the cost function for path planning. Meeting the flexibility, robustness, and efficiency requirements for real-time mobile robot path planning implementation is challenging. A review of multi-robot path planning is published to compare the path planning approaches and decision-making strategies, listing the challenges. This thesis aims to tackle major challenges by developing intelligent hybrid approaches, including 1) trapping in local optimal, 2) slow convergence of path generation, and 3) robots' fault tolerance. It also provides the path planning algorithms for single-robot and multi-robot systems in three-dimensional and two-dimensional space. For single mobile robot path planning, the bio-inspired approaches have gained more attention recently with high robustness and flexibility. In contrast, it is highly possible to trap in a local optimal. The proposed Harmony-particle swarm optimization algorithm significantly reduces the iterations during planning to solve the aerial path planning problem in a multi-building environment. Also, a hybrid approach of particle swarm optimization and simulated annealing is proposed for single-vehicle path planning in the industrial warehouse scenario. It updates the personal best value to jump out of the locally optimal. Compared with other evolutionary approaches, it shows excellent performance. Moreover, fast convergence is a significant challenge for multiple robots' path planning. A dual-layer Weight-Leader-Vicsek-Model is proposed that generates the path for the virtual leaders first for each group of robots, and then all the robots will move by following their leaders. This dual-layer approach can achieve fast convergence, generating vehicle paths in one calculation step. Fault tolerance is also an essential issue for the real-time implementation of path planning, but it is lacking in previous studies. The Cultural-Particle Swarm Optimization algorithm is proposed to offer a backup plan in case of system failures. It updates the inertial weight to enhance the search abilities, balancing the global and local search abilities. The experiments and validated results are presented for each proposed approach.	en_US.UTF-8
dc.format	Thesis (PhD)
dc.language.iso	en_US	en_US.UTF-8
dc.relation	https://opus.lib.uts.edu.au/bitstream/10453/171955/1/thesis.pdf
dc.rights	info:eu-repo/semantics/openAccess
dc.rights	The author owns the copyright in this thesis including all reproduction and reuse rights for the work. The work may not be altered without the permission of the copyright owner. Attribution is essential when quoting or paraphrasing from this thesis.
dc.rights	© 2023 Shiwei Lin
dc.rights	au.edu.uts.lib/nph
dc.title	Intelligent Hybrid Approaches for Mobile Robots Path Planning	en_US.UTF-8
dc.type	Thesis
utslib.copyright.status	open_access	*

Abstract:

The practical applications of mobile robots are widely implemented in various areas, such as education, industry, environment, and civil applications. The requirements of robots' navigation are one of the primary considerations for autonomous operation. Path planning is essential for the successful application of mobile robots. Considering all available information, it aims to generate the robot's optimal path from the start to the target location. Depending on the operation scenarios, various factors are counted in the cost function for path planning. Meeting the flexibility, robustness, and efficiency requirements for real-time mobile robot path planning implementation is challenging. A review of multi-robot path planning is published to compare the path planning approaches and decision-making strategies, listing the challenges. This thesis aims to tackle major challenges by developing intelligent hybrid approaches, including 1) trapping in local optimal, 2) slow convergence of path generation, and 3) robots' fault tolerance. It also provides the path planning algorithms for single-robot and multi-robot systems in three-dimensional and two-dimensional space. For single mobile robot path planning, the bio-inspired approaches have gained more attention recently with high robustness and flexibility. In contrast, it is highly possible to trap in a local optimal. The proposed Harmony-particle swarm optimization algorithm significantly reduces the iterations during planning to solve the aerial path planning problem in a multi-building environment. Also, a hybrid approach of particle swarm optimization and simulated annealing is proposed for single-vehicle path planning in the industrial warehouse scenario. It updates the personal best value to jump out of the locally optimal. Compared with other evolutionary approaches, it shows excellent performance. Moreover, fast convergence is a significant challenge for multiple robots' path planning. A dual-layer Weight-Leader-Vicsek-Model is proposed that generates the path for the virtual leaders first for each group of robots, and then all the robots will move by following their leaders. This dual-layer approach can achieve fast convergence, generating vehicle paths in one calculation step. Fault tolerance is also an essential issue for the real-time implementation of path planning, but it is lacking in previous studies. The Cultural-Particle Swarm Optimization algorithm is proposed to offer a backup plan in case of system failures. It updates the inertial weight to enhance the search abilities, balancing the global and local search abilities. The experiments and validated results are presented for each proposed approach.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/171955