Simulating Safe Bite Transfer in Robot-Assisted Feeding with a Soft Head and Articulated Jaw

Area of Science:

• Robotics
• Biomechanics
• Human-Robot Interaction

Background:

• Robot-assisted feeding requires safe and comfortable human-robot interaction.
• Modeling physical interactions is crucial for developing safe feeding robots.

Purpose of the Study:

• To develop a novel simulation approach for modeling physical human-robot interaction in robot-assisted feeding.
• To evaluate the impact of bite transfer parameters on user safety and comfort.

Main Methods:

• Utilized a physics-based simulator (MuJoCo) with soft-body dynamics.
• Integrated a flexible head model with a rigid skeleton, accounting for internal dynamics.
• Incorporated realistic soft-skin contact dynamics for simulation.

Main Results:

• Systematically evaluated bite transfer parameters like insertion depth and entry angle.
• Identified a straight-in-straight-out strategy as optimal for minimizing forces.
• Demonstrated enhanced user comfort with the straight-in-straight-out approach, assuming a static head.

Conclusions:

• Simulation-based modeling provides a safe and controlled method for evaluating robot-assisted feeding parameters.
• The proposed approach enables systematic analysis of human-robot interaction for improved feeding robot design.