JMIR Publications

ABSTRACT

Background:

The integration of collaborative robots (cobots) in industrial settings has the potential to enhance worker safety and efficiency by improving postural control and reducing the biomechanical risk. Understanding the specific impacts of varying levels of human-robot collaboration (HRC) on these factors is crucial for optimizing cobot use.

Objective:

This study aims to investigate the biomechanical effects of different levels of HRC on postural stability and control during simulated working tasks.

Methods:

Fourteen participants performed simulated cashier working activities under four different collaboration modalities, with increasing levels of cobot assistance: Full (Fu), Half Robot Touch (HRT), Half Robot (HRb), and Full Robot (FRb). Center of Pressure (CoP) trajectories were extracted from two force plates data to calculate four posturography parameters: Mean Distance (MDIST), Mean Velocity (MVELO), 95% confidence ellipse area (AREA-CE), and Sway Area (AREA-SW) were analysed to assess the impact of cobot intervention on postural control.

Results:

Non-parametric tests showed significance (P < .01) on the effect of the collaboration modalities on the four analyzed parameters. Post-hoc tests revealed that FRb modality led to the greatest enhancement in postural stability,, with a reduction in MDIST (4.2 ± 1.3 cm in Fu vs. 1.6 ± 0.5 cm in FRb, P < .01) and MVELO (16.3 ± 5.2 cm/s in Fu vs. 7.9 ± 1.1 cm/s in FRb, P < .01). AREA-CE and AREA-SW also decreased significantly with higher levels of cobot assistance (AREA-CE: 134 ± 91 cm² in Fu vs. 22 ± 12 cm² in FRb, P < .01; AREA-SW: 16.2 ± 8.4 cm²/s in Fu vs. 4.0 ± 1.6 cm²/s in FRb, P < .01). Complete assistance of the cobot significantly reduced inter-individual variability of all CoP parameters: FRb modality, as compared to all other conditions, by removing the weight of the object during loading/unloading phases, causes a significant decrease in all parameter values.

Conclusions:

Increased cobot assistance significantly enhances postural stability and reduces biomechanical load on workers during simulated tasks. Full assistance from cobot, in particular, minimizes postural displacements, indicating more consistent postural control improvements across individuals. However, high levels of cobot intervention also reduced the natural variation in how people balanced themselves. This could potentially lead to discomfort in the long run. Mid-level cobot assistance modalities can thus be considered as a good compromise in reducing biomechanical risks associated with postural stability at the same time granting a satisfactory level of user control.