Control System of a Lower-Extremity Exoskeleton Based on the Artificial Neural Network

Area of Science:

• Biomedical Engineering
• Rehabilitation Robotics
• Artificial Intelligence in Medicine

Background:

• Lower-extremity exoskeletons offer structural support and strength for rehabilitation.
• Muscular weakness and reduced range of motion hinder recovery after certain complications.
• Advanced control systems are crucial for effective exoskeleton-assisted therapy.

Purpose of the Study:

• To design and implement a functional prototype of a lower extremity exoskeleton brace.
• To develop a wireless communication control system for the exoskeleton.
• To investigate the use of Artificial Neural Networks (ANNs) for controlling exoskeleton functions.

Main Methods:

• Designed a lower extremity exoskeleton brace providing supportive torque and increasing range of motion.
• Implemented a wireless control system for the exoskeleton prototype.
• Utilized an Artificial Neural Network (ANN) as the primary control modality.
• Evaluated prototype functionality using time-angle graphs.

Main Results:

• The prototype successfully provided supportive torque and enhanced range of motion.
• The wireless control system enabled basic mobility functions: leg elevation, standing, and slow walking.
• ANN-based control demonstrated effective operation of the exoskeleton.
• Functional testing validated the prototype's capabilities.

Conclusions:

• The developed lower-extremity exoskeleton prototype shows promise for rehabilitation.
• Artificial Neural Networks (ANNs) are a viable control strategy for exoskeletons in joint impairment therapy.
• The system has the potential to significantly aid patients with reduced muscular strength and limited joint mobility.