A hybrid human motion tracking system for virtual rehabilitation

The main objective of this paper is to design and develop a novel hybrid human motion tracking system for gait and dynamic balance training program based on virtual reality games for children with motor impairments. The whole training program comprises of a novel human motion tracking system and a VR game system that enables objective monitoring of a child's progress, improves consistency between and within therapists and increases the ease of setting up an appropriate environment for different rehabilitation activities. This program will also be able to engage a child's attention and increase his/her motivation during rehabilitation. The precise determination of human motion is an important requirement to be satisfied in order to develop human interface rehabilitation programm. Human motion tracking provides critical information for intelligent human-computer collaboration. This paper evaluates a camera motion capture system which registers foot movements of an user. However, the occlusion problem returned by this system has led to the need of using another localization system to enlarge the area which the child can play, based on electro-magnet technology. A Kalman filter fusion algorithm which combines the measurements of these systems is developed. This algorithm unifies the advantages of both systems: high localization accuracy from the opti-track motion capture system and no occlusion problem from the electro-magnet system. The developed hybrid system not only tracks the movements of lower limbs of the user as opti-track motion capture systems, but is also able to identify the user in the environment.