Dynamic Human-Robot Collaboration for Industry 5.0

Wang, Xiaodan 2024. Dynamic Human-Robot Collaboration for Industry 5.0. PhD Thesis, Cardiff University. Item availability restricted.

	PDF (Thesis) - Accepted Post-Print Version Restricted to Repository staff only until 11 June 2026 due to copyright restrictions. Download (8MB)
	PDF (Cardiff University Electronic Publication Form) Restricted to Repository staff only until 11 June 2026 due to copyright restrictions. Download (175kB)

Abstract

Industry 5.0 is a human-centred manufacturing approach, which promotes Human-Robot Collaboration (HRC) to benefit production by combining human flexibility with robotic precision and endurance. While HRC has the potential to enhance productivity, it also introduces challenges due to the inherent unpredictability of human behaviour. This uncertainty, coupled with increasingly complex and dynamic manufacturing environments, introduce additional constraints to task planning and optimisation. A key challenge is ensuring HRC's efficiency and robustness by detecting and responding effectively to dynamic changes in real time while evolving human and robot models together. To address this challenge, this research aims to develop a novel general-purpose framework of dynamic human-robot collaboration that considers human and robot factors to improve manufacturing efficiency and ergonomics. In particular, the dynamic task planning problem in manufacturing systems is addressed by considering the whole process, from collaboration-centred task modelling, task allocation and scheduling, through task monitoring with abnormality detection, to task re-allocation and re-scheduling. A novel collaboration-centred task model is developed based on characteristics of HRC manufacturing and a hierarchical task analysis approach. Trust models are developed based on essential human and robot factors. An optimisation model for task allocation and scheduling is then developed, which considers the task sequence and the collaboration of humans and robots. Next, a novel hybrid heuristic algorithm named improved-NSGA-II-SA-AHP is developed to minimise the cycle time and ergonomic risks, and to improve the trust levels of the HRC. The task execution states and agents’ factors are monitored to detect any abnormalities. The impact of uncertainties and dynamic factors on HRC is analysed through mathematical modelling and II simulations, providing insights into human factors in collaborative settings. Additionally, the research introduces a digital twin-based task monitoring approach, incorporating the abnormality detection methods to enable task re-allocation and ensure system adaptability. Finally, the application of the proposed framework is described and evaluated through an industrial assembly case study. Computational results are presented to show the performance and feasibility of the proposed framework. The significance of this research lies in its contribution to the development of a more efficient, and adaptive HRC through the use of Artificial Intelligence and Machine Learning techniques. Keywords: Industry 5.0; human-robot collaboration; human factors; trust

Item Type:	Thesis (PhD)
Date Type:	Completion
Status:	Unpublished
Schools:	Schools > Engineering
Uncontrolled Keywords:	1. Industry 5.0 2. Human-robot collaboration 3. Human factors 4. Multi-objective optimisation 5. Task allocation and scheduling 6. Digital twin
Date of First Compliant Deposit:	11 June 2025
Last Modified:	11 Jun 2025 15:06
URI:	https://orca.cardiff.ac.uk/id/eprint/179007

Actions (repository staff only)

Edit Item