Vision based gait analysis on robotic walking stabilization system for patients with Parkinson´s Disease

In this paper, we propose a gait analysis based walking stabilization system on an active robotic walker for Parkinson´s Disease patient. We developed a vision based gait analysis that lower limbs from the frontal view which is captured by single RGB-D camera on the robotic walker. Our system constructs the lower limbs skeleton and kinematic model for calculating the walking gait features such as average walking step length, velocity, width and joint angles. When the user suffer in abnormal gait, the robotic walker assists user to stabilize their gaits through the adjustable control motion of walker according to the user´s walking gait and prompting user with auditory cues. In experiment, we had invited Parkinson´s Disease patients to test our system in senior care unit, the experimental results show that when the patients have relatively stable gait than without using this system.