Application of multiple regression models in the analysis of kinematic parameters in competitive gymnastics

Abstract

Abstract Nowadays, the progress of competitive gymnasts needs more and more help from sports science, and it is from the mechanics perspective that human movement science analyzes this complex system of the human body. In this paper, the back handsprings in gymnastics were tested and recorded with a motion capture system to evaluate the back handsprings’ technical movements quantitatively. A multiple regression model was used to model the dynamics of the human body and carry out simulation calculations to analyze and study the movement patterns and lower limb stresses. The results showed that the maximum load rate through the back handspring experimental stage was (379.91±101.23) BW/s, and the time of appearance of the maximum load rate was (0.025±0.003) s. The maximum load decay rate was (-321.61±107.21) BW/s, and the time of appearance of the maximum load decay rate was (0.043±0.005) s. The difference between the results and the actual test value difference between the results and the actual test values was less than 0.005, indicating the feasibility of applying the multiple regression model to the kinematic parameters analysis of competitive gymnastics. The multiple regression-based kinematic parameter analysis models can effectively solve the problem of force on the bones and joints of difficult gymnastic movements, which can better assist trainers in learning gymnastic items and provide theoretical support and reference basis for the development and training of difficult skills in competitive gymnastics.