Gait disorders are significant indicators of neurological diseases such as Parkinson’s disease and reduce the quality of life of patients. Although the detection and classification of gait disorders is essential for treatment and diagnosis, there is currently no single standardized gait analysis system. Wearable sensors offer a promising solution, providing accessible gait analysis by capturing periodic movements during walking. In addition to analysis, soft body robotic exoskeletons improve walking by applying controlled robotic forces to correct abnormal gait patterns. However, for optimal therapeutic effects, exoskeletons must be controlled according to the disorder characteristics and real-time feedback.

This study presents the design of a real-time gait analysis system using wearable sensors. This analysis system can be used to both diagnose and control soft body exoskeletons in Parkinson's patients. Wearable sensors consist of three low-cost electromyography (EMG) circuits and four 6-axis inertial measurement units (IMUs), positioned on the primary muscle groups involved in gait. Load cells are placed under the feet to capture dynamic force data. All sensor data is acquired and wirelessly transmitted to the server for signal processing by a central microcontroller.

The data is processed to extract both physiological and kinematic parameters from gait cycles. Using the dataset of gait cycle parameters, a machine learning model facilitates a quantitative assessment of the spectrum of gait disorders. This analysis will generate real-time feedback by evaluating kinematic and physiological parameters. The objective of the feedback is to provide an adaptive control mechanism for therapeutic devices such as soft body exoskeletons suitable for gait disorders. The machine learning model is also used to iteratively improve the control model at each step. In this way, our study will offer low-cost adaptive physiologic control for traditional therapeutic exoskeletons especially for Parkinson's patients.