An Intuitive Formulation of the Human Arm Active Endpoint Stiffness

Area of Science:

• Biomechanics
• Robotics
• Human-Computer Interaction

Background:

• Understanding human arm active endpoint stiffness is crucial for effective human-robot interaction and ergonomic assessments.
• Previous models for human arm stiffness lacked real-time personalization and computational efficiency.

Purpose of the Study:

• To propose an intuitive and real-time model for human arm active endpoint stiffness.
• To develop a model that leverages geometric information from a simplified human arm skeleton and muscle synergy assumptions.
• To enable fast identification and personalization of the stiffness model for individual users.

Main Methods:

• Constructed a symmetric and positive-definite stiffness matrix (Kc) using eigendecomposition (Kc=VDVT).
• Directly computed the eigenvector (V) and eigenvalue (D) matrices by utilizing 3D skeleton geometry and muscle co-contraction assumptions.
• Validated the model through perturbation experiments across multiple subjects with varying arm configurations and muscle activation states.

Main Results:

• The proposed model accurately predicts human active arm endpoint stiffness.
• The model demonstrates significant advantages in fast identification and personalization compared to previous models.
• Perturbation experiments confirmed the model's accuracy across different conditions.

Conclusions:

• The developed model provides a principled and simple approach to real-time human arm stiffness prediction.
• The model's speed and personalization capabilities make it suitable for applications like teleoperation, human-robot collaboration, and ergonomic assessments.
• This work advances the development of personalizable human kinodynamic models for interactive systems.