Design of a Biomechanical Model and a Set of Neural Networks for Monitoring of Weightlifting

Abstract

Nowadays, young athletes have injuries that leave them out the competitions, this injuries may occur due to the lack of the basic principles of weight training, muscle imbalance, fatigue, overhead, among others. The proposed weightlifting monitoring system is intended to coaches in order to provide them better control of the athletes during the training session, through continuous monitoring of exercise's measurements. This allows to detect certain risks of injury that may result if an exercise is poorly executed. To handle the uncertainty generated by the comparison of data obtained from athlete with the data of the biomechanical model for weightlifting, a set of neural networks to classify the athletes' movements during training is used. This paper describes the biomechanical model and the proposed neural network design for such monitoring system.