Tele-Impedance Control Approach Using Wearable Sensors

Area of Science:

• Robotics
• Human-Computer Interaction
• Biomechanics

Background:

• Tele-operational tasks face challenges with instability and reliability, especially during unpredictable human-robot interactions.
• Existing methods often require complex setups, limiting their use in real-world, unstructured environments.

Purpose of the Study:

• To develop a real-time control law for a 2-Degrees-of-Freedom (2-DoF) task that accurately reproduces a subject's arm impedance and kinematic behavior on a remote avatar.
• To create a portable tele-impedance controller utilizing only wearable sensors for enhanced usability in diverse environments.

Main Methods:

• Human arm impedance and kinematics were estimated using electromyography (EMG) and motion inertial measurement unit (M-IMU) data.
• These estimated human arm properties were mapped onto the avatar arm via an impedance control strategy.
• A muscle model was employed to map human arm stiffness to the robot, with performance evaluated using R-squared (R²) values.

Main Results:

• The proposed portable tele-impedance controller demonstrated effective reproduction of human arm impedance and kinematic behavior.
• The muscle model achieved good performance (R² > 0.7) in mapping human stiffness from five healthy subjects.
• The system's reliance on wearable sensors facilitates use in unstructured environments, overcoming limitations of prior literature methods.

Conclusions:

• The developed portable tele-impedance controller offers a stable and reliable solution for tele-operational tasks.
• The successful mapping of human arm properties using wearable sensors indicates the potential for widespread application in tele-impedance control.
• This technology enhances the feasibility of remote avatar control in real-world, unpredictable scenarios.