Ankle exoskeleton torque controllers based on soleus muscle models.

Area of Science:

• Biomechanics
• Robotics
• Human-Machine Interaction

Background:

• Current powered exoskeletons are often limited to specific tasks.
• Wider adoption requires more generalized controller designs.
• Ankle exoskeletons are crucial for mobility assistance and augmentation.

Purpose of the Study:

• To present and evaluate two novel controller designs for ankle exoskeletons.
• To develop controllers capable of generalizing across different locomotion tasks (walking, running).
• To move beyond task-specific parameter optimization for exoskeleton control.

Main Methods:

• Developed controllers based on soleus fascicle and Achilles tendon biomechanical models.
• Utilized estimates of adenosine triphosphate hydrolysis rate based on fascicle velocity.
• Evaluated models using established muscle dynamics from literature, validated with ultrasound data.
• Compared simulated controller performance against human-in-the-loop optimized torque profiles.

Main Results:

• Both controllers generated distinct torque profiles for walking and running, adapting to speed variations.
• One controller showed better performance for walking, while the other produced profiles similar to literature for both gaits.
• Proposed methods demonstrated potential to generate generalized profiles without extensive per-task optimization.

Conclusions:

• The developed controllers offer a generalized approach for ankle exoskeleton control, applicable to both walking and running.
• These methods reduce the need for lengthy, individual-specific parameter tuning for each task.
• Future work should investigate the impact of these generalized controllers on human motor behavior and adaptation.