Evaluation of Kinect Sensor in Mechanical Horse Simulator for Equine-Assisted Therapy

Abstract

Equine-assisted therapy (EAT) is one of therapy strategy practiced for disabled people to improve physical functions which can be performed using either real horse or mechanical house simulator. In EAT based on horse simulator, there is no automatic tracking tools that are being used to track movement, speed, position or posture of rider on horses during equine-assisted therapy. Hence, this study aims to evaluate the potential application of Kinect as the sensor for automated tracking of rider speed on a mechanical horse simulator. This sensor is used to capture skeletal data of rider riding mechanical horse simulator for three increasing speed and to analyze the skeletal data using the boxplot approach. From 25 human joints that could be detected, there are four joints that are significance and stable joints to represent the speed which are joint 2 (neck), joint 3 (head), joint 4 (left shoulder) and joint 8 (right shoulder).