Force-velocity coupling limits human adaptation in physical human-robot interaction

Area of Science:

• Robotics
• Human-Computer Interaction
• Biomechanics

Background:

• Physical human-robot interaction requires mutual adaptation for synergistic behavior.
• Human movement naturally scales velocity with trajectory curvature.
• Understanding human responses to robot motion is crucial for safe and efficient collaboration.

Purpose of the Study:

• To investigate human responses to robots employing different velocity profiles during physical interaction.
• To determine how robot velocity impacts interaction forces and human adaptation.
• To explore the role of biomechanical constraints and learning in human-robot force modulation.

Main Methods:

• Two experiments involved humans tracking a robot moving along an elliptical path with varying velocity profiles.
• Participants were instructed to minimize interaction forces.
• Data analysis focused on involuntary forces, angular velocities, and adaptation over practice sessions with and without visual feedback.

Main Results:

• Higher involuntary forces were observed when robots moved at constant velocity or exaggerated biological velocity-curvature scaling.
• Increased robot angular velocity correlated with greater tangential and normal interaction forces.
• Interaction forces decreased for non-biological profiles with real-time visual feedback, indicating learned compensation for inertial forces.

Conclusions:

• Robot velocity profiles significantly influence human interaction forces and adaptation.
• Biomechanical constraints play a minor role; human motor learning and prediction are key factors.
• Designing robots with adaptable velocity profiles considering human motor capabilities is essential for effective physical human-robot interaction, especially in collaborative and wearable robotics.