Biologically Inspired Optimal Terminal Iterative Learning Control for the Swing Phase of Gait in a Hybrid...

Area of Science:

• Biomedical Engineering
• Neuroscience
• Robotics

Background:

• Development of a "Muscle First" Motor-Assisted Hybrid Neuroprosthesis (MAHNP) for individuals with paraplegia.
• MAHNP integrates exoskeletal bracing with neural stimulation for upright ambulation.
• Spinal cord injury (SCI) necessitates advanced assistive technologies for mobility.

Purpose of the Study:

• To develop and evaluate an iterative learning control (ILC) strategy for the MAHNP.
• To enable biologically inspired, ballistic leg swing motions.
• To optimize the balance between muscular recruitment and exoskeletal assistance.

Main Methods:

• An ILC strategy was designed to maximize muscle activation and motor assistance.
• A 3D musculoskeletal model of the lower leg and pelvis was utilized.
• The OpenSim biomechanical modeling suite was employed for simulations, incorporating exoskeletal inertia.

Main Results:

• Preliminary simulations demonstrated the controller's efficacy in swing-leg movements.
• The ILC strategy successfully learned to balance muscular and motor contributions.
• Consistent stepping was achieved with a 0.3% error after 15 iterations.

Conclusions:

• The developed ILC strategy shows promise for enhancing the performance of the MAHNP.
• This approach facilitates learning and adaptation for improved assistive walking in SCI individuals.
• The findings support the potential of hybrid neuroprosthetics for restoring ambulation.