Advancing HRCA: Physical fatigue alleviation and proactive robotic assistance

You, Yingchao 2024. Advancing HRCA: Physical fatigue alleviation and proactive robotic assistance. PhD Thesis, Cardiff University. Item availability restricted.

Preview	PDF (Thesis) - Accepted Post-Print Version Download (25MB) | Preview
	PDF (Cardiff University Electronic Publication Form) Restricted to Repository staff only Download (1MB)

Abstract

Human-robot collaborative assembly (HRCA) is a critical paradigm in modern manufacturing, as it leverages the complementary strengths of humans and robots. As manufacturing systems evolve toward a more human-centric direction, there is growing emphasis not only on efficiency but also on factors such as ergonomics. However, research improving worker’s well-being in HRCA remains insufficient. On the one hand, human fatigue is rarely factored into decision-making in HRCA task planning systems, and neglecting physical fatigue can adversely impact worker health, potentially leading to musculoskeletal disorders in severe cases. On the other hand, proactive robot assistance significantly improves operational efficiency and worker ergonomics in HRCA. Nevertheless, state-of-the-art knowledge-based task planners in HRCA struggle to rapidly update their knowledge and adapt to the demands of new products. Therefore, this thesis focuses on two critical aspects of HRCA systems: mitigating physical fatigue and learning proactive robotic assistance, both achieved through task planning methodologies. To address the human physical fatigue estimation and alleviation challenges, we introduce a human-digital twin method for physical fatigue assessment in HRC assembly tasks. The methodology encompasses an IK-BiLSTM-AM-based surrogate model for real-time muscle force and muscle fatigue assessment. The effectiveness of i this approach has been validated through proof-of-concept assembly experiments. The results show that the IK-BiLSTM-AM model achieves a minimum of 8% greater accuracy in muscle force estimation than the baseline methods. To realize the human fatigue alleviation in HRCA, we study a task planning method for physical exertion alleviation of workers in HRCA by leveraging the reinforcement learning (RL) method to train a policy, ready to be used in HRCA tasks. The policy is trained in a DuelingDQN-AM framework, utilising a carefully designed reward function informed by the estimated physical exertion of workers. The effectiveness of this approach has been validated through a simulation experiment and a proof-of-concept real assembly experiment. This work studies learning proactive assistance by leveraging RL to train a policy, ready to be used for robot proactive assistance planning in HRCA. To solve the limitation therein, we propose an offline RL framework where a policy for proactive assistance is trained using the dataset visually extracted from human demonstration. In particular, an RL algorithm with a conservative Q-value is utilised to train a planning policy in an actor-critic framework with carefully designed state and reward functions. The experimental results show that the assistance task provided by the robot can fully meet the task requirement and improve human assembly preference satisfaction by 47.06% compared to a static strategy.

Item Type:	Thesis (PhD)
Date Type:	Completion
Status:	Unpublished
Schools:	Schools > Engineering
Uncontrolled Keywords:	1. Human robot collaboration 2. Reinforcement learning 3. Human-centric manufacturing 4. Ergonomics 5. Human digital twin 6. Proactive assistance
Date of First Compliant Deposit:	11 June 2025
Last Modified:	11 Jun 2025 14:57
URI:	https://orca.cardiff.ac.uk/id/eprint/178999

Actions (repository staff only)

Edit Item