Regression model for predicting the maximum load of the movement

In this paper, the motion information is considered to be essential for calculating the corresponding maximum load of the movement. Due to the high cost of force plates and the challenges on practical scenes, recording the body joints' loading in a direct way is not feasible to do outside the laboratory. Therefore, the main purpose of this paper is to investigate the relationship between the motion pattern and the maximum load, using the regression analysis. In our experiments, the dataset contains 22 subjects. The motion information of each subject is recorded using the 3D motion-capture camera system. The knee's angles, hip's angles, and ankle's angles on sagittal and non-sagittal planes are investigated for predicting the maximum loads. Finally, the proposed method can achieve the average relative error about twelve percent for the prediction.