Control of a wrist joint motion simulator: A phantom study.

Area of Science:

• Biomechanics
• Robotics
• Human-Machine Interface

Background:

• Muscle redundancy and co-activation necessitate optimized load distribution in physiologic joint simulators.
• Current control strategies rely on position/force feedback, with limitations in replicating complex movements.

Purpose of the Study:

• To develop and refine control strategies for a muscle-activated physiologic wrist simulator.
• To evaluate novel hybrid and cascade control strategies against standard position and force control methods.

Main Methods:

• A functional human arm replica with a muscle-activated wrist simulator was used.
• Electromechanical actuators applied tensile loads, with load cells monitoring muscle forces and optical motion capture tracking joint angles.
• Four control strategies were assessed for kinematic error, repeatability, and co-contraction variation.

Main Results:

• Novel hybrid and cascade control strategies achieved kinematic errors under 1.5° and demonstrated superior co-contraction control.
• These strategies did not require predefined antagonistic forces or muscle force ratios.
• Muscle forces from novel strategies closely matched in vivo EMG and literature data.

Conclusions:

• Hybrid and cascade control strategies offer accurate, repeatable complex wrist motions (e.g., dart thrower's motion, circumduction).
• These novel strategies provide physiologically relevant muscle forces and robust performance for joint simulators.