Dynamic Gait Monitoring Mobile Platform

Abstract

Human gait is an important indicator of health. Existing gait analysis systems are either expensive, intrusive, or require structured environments such as a clinic or a laboratory. In this research, a low-cost, non-obtrusive, dynamic gait monitoring platform is presented. By utilizing a mobile robot equipped with a Kinect sensor, comprehensive gait information can be extracted. The mobile platform tracks the skeletal joint movements while following the person. The acquired skeletal joint data is filtered to improve detection. Gait parameters such as step length, cadence and gait cycle time are extracted by processing the filtered data. The proposed approach was validated by using a VICON motion capture system. Results show that the proposed system is able to accurately detect gait parameters but requires a calibration procedure. Even though the camera is moving while tracking, the performance is on par with existing works. Step times can be detected with an average accuracy of around 10 milliseconds. Step length can be detected with an average accuracy of a few centimeters.