Ground reaction force prediction for rehabilitation robotics and biomechanics

Biomechanics is rapidly increasing research area which aims to increase the fidelity and complexity of the models to better reflect real world occurrences accurately. As such, the use of motion capture devices to drive the position of a human model have become extremely popular. This information can be used for a variety of applications including rehabilitation, human movement studies and virtual reality. Motion capture systems are only capable of recording kinematics from human motion and subsequently are insufficient for detailed analysis involving dynamics for uses in fields such as rehabilitation robotics. Therefore, this paper looks at prediction of ground reaction forces based upon kinematics data acquired through motion capture and musculo-skeletal mass-inertia properties. For this purpose, the dynamic effects of the human motion on the ground reaction force are studied using the sum of forces acting upon each segment of the body. The results of these calculations is then compared to measured ground reaction forces showing a high level of accuracy from the estimated values ground reaction forces. Such results could be used in subsequent research for rehabilitation robotics for better understanding of human performance in a cost-effective manner.