Abstract
With the advent of an aging society, the lack of physical activity has become a major concern, leading to various age-related diseases. To prevent such issues, research on wearable robots aimed at improving gait has been actively pursued. Among them, exoskeleton robots, a widely used approach, require an accurate understanding of the user's gait cycle for effective control. Various studies have explored gait cycle detection and prediction methods depending on the type of gait robot platform and the use of sensors. However, a major challenge in gait cycle prediction algorithms remains the issue of nonlinear predictive trajectories. In the study, a robotic hip exoskeleton (RHE) was utilized to implement an enhanced gait phase estimation (GPE) algorithm integrated with a muscle support system. Participants were divided into two groups (Group A and Group B) based on their initial gait performance, and the effectiveness of gait rehabilitation training was evaluated. The results showed that in the 10-meter walk test (10MWT), walking time decreased by approximately 5% in Group A and 27% in Group B. In the 6-minute walk test (6MinWT), walking distance increased by approximately 1% in Group A and 14% in Group B. Group B, which had lower initial gait performance, showed a greater gait performance improvement rate compared to Group A, which had higher initial gait performance. Through the gait performance results of the two groups, the applicability of the GPE based muscle support algorithm was confirmed.