Human-arm-and-hand-dynamic model with variability analyses for a stylus-based haptic interface

Area of Science:

• Robotics
• Human-Computer Interaction
• Biomechanics

Background:

• Accurate human arm dynamic models are essential for effective haptic interface control and system design.
• Existing literature lacks detailed analyses of human arm variability, leading to conservative and suboptimal controller designs.
• Quantifying this variability is key to developing more precise and adaptive haptic systems.

Purpose of the Study:

• To develop inter- and intrasubject variability models for human arm dynamics in the context of a stylus-based haptic device.
• To identify grip-force-dependent dynamic models for human arm interaction.
• To model human variability as both structured and unstructured uncertainties.

Main Methods:

• System identification techniques were applied to data collected from 15 human subjects using a Phantom Premium 1.5a haptic device.
• Grip-force-dependent models were identified for grip forces ranging from 1 to 3 Newtons across three spatial axes.
• Variability was modeled using frequency-domain analysis, with force as input and position as output, examining confidence intervals.

Main Results:

• Grip-force-dependent human arm dynamic models were successfully identified.
• Inter- and intrasubject variability were quantified and modeled as structured and unstructured uncertainties.
• Maximum variation, 95%, and 67% confidence interval limits for human arm dynamics were determined.

Conclusions:

• The developed models provide a more accurate representation of human arm dynamics and their variability.
• These precise models enable the design of targeted controllers for specific subsets of human operator dynamics.
• This research advances the development of more sophisticated and effective haptic interfaces.