Evaluation of kinematic similarity between non-assisted and assisted walking with a hip exoskeleton using a vector coding technique.

Flexible wearable walking-assist robots, unlike rigid rehabilitation robots, offer lightweight designs and ease of wear. This study focused on quantifying differences in joint coordination patterns between robotic and human motion during gait. Eleven healthy participants completed 2-min treadmill walking trials in three conditions: normal walking (None), walking with the robot worn but inactive (Off), and walking with the robot worn and active (On). Using a vector coding technique, coupling angles and their variability were analyzed across four gait phases. An increase in the coupling angle was detected in the mid stance and late stance phases in the On condition, which was driven by changes in the knee joint angles. Coupling angle variability was significantly reduced in the terminal swing phase in the On condition compared to None, suggesting enhanced consistency in the movement. These findings suggest that a vector coding technique can detect differences in hip-knee joint coordination patterns in the mid and late stance phases between non-assisted and assisted walking with a hip assist robot and that coupling angle can be used as a measure for evaluating kinematics of assisted walking.Clinical Relevance- A vector coding technique can be used to evaluate the effect of walking assist robot on movement patterns for a better applicability in daily life.