Time-based identification of human ankle impedance via Microsoft Kinect Conference Paper uri icon

abstract

  • This paper presents an affordable platform to estimate human ankle mechanical impedance. This platform uses Microsoft Kinect version 2 as the motion capture system of choice and a hybrid algorithm to estimate the biomechanical parameters. The algorithm is based on the combination of an Extended Kalman filter and a Genetic Algorithms. The information provided by Kinect together with the ankle biomechanical parameters can be utilized to estimate the dynamic behavior of the recovery from falls. To prove the precision of the 3D measurements obtained with Kinect a comparison with a visual system, based on two industrial cameras was performed. Both systems were calibrated tracking the end-effector position of an industrial robot. The hold and release (H%26R) experimental paradigm was used to estimate the ankle mechanics on seven subjects. The results show that Kinect v2 is a reliable motion capture device to study the neuro-mechanical response of recovering from falling. © 2015 IEEE.

publication date

  • 2016-01-01