The Quantitative Prediction of Auxiliary Sliding Distance of Freestyle Skiing Based on MLP Neural Network

Abstract

Freestyle skiing is a highly concerned air skill project. At present, the selection of auxiliary sliding distance is mainly based on the on-site experience of athletes and coaches. In this paper, the researchers design a quantitative prediction system of auxiliary sliding distance of freestyle skiing composed of data acquisition robot and Multilayer Perception (MLP) neural network prediction model. The athlete's own parameters and the natural parameters collected by the robot are taken as the input of the prediction model, and finally a predicted value of auxiliary sliding distance is obtained. The results show that the root mean square error of the prediction is 0.4384 meters (the skis are about 1.7 meters in length), which indicated that the prediction model can be used as a reasonable reference for athletes and coaches.