A review led by researchers from Monash University suggests that improving how robots predict human behaviour could help make manufacturing environments safer, more adaptive and efficient, as the sector transitions toward Industry 5.0.
Published in the International Journal of Production Research, the study examines how human-robot collaboration can be enhanced in shared industrial settings, where closer interaction between workers and machines is becoming more common in modern manufacturing systems.
The review notes that Industry 5.0 is driving a shift toward more human-centred production, combining human creativity and decision-making with robotic precision and speed.
However, it also highlights that closer collaboration introduces new risks, particularly when robots are unable to accurately anticipate human actions, potentially leading to collisions, delays and inefficiencies.
Researchers assessed several approaches to predicting human behaviour, including mechanism-based models focused on physical motion, data-driven models using sensors and artificial intelligence, and hybrid systems that combine both.
While each method offers advantages, the study finds that integrated approaches are likely to be more effective in future manufacturing environments.
The paper also identifies ongoing challenges, such as the variability of human behaviour, the lack of standardised multimodal datasets, limitations in physical modelling, and the need to better account for human factors like trust, workload and cognitive state.
To address these gaps, the authors propose a unified framework that brings together multimodal data, physical modelling, behaviour prediction and adaptive control to improve collaboration between humans and robots.
Co-author Yunlong Tang said enhancing how robots interpret and respond to human behaviour will be critical for next-generation manufacturing systems.
“Industry 5.0 is about designing manufacturing systems around people as well as technology. By improving how robots predict human behaviour, we can move towards production environments that are not only more productive, but also safer, more adaptive and more human-centred,” he said.
Read the full paper here: https://doi.org/10.1080/00207543.2026.2639732
