Research on intelligent analysis of running sports training performance based on artificial intelligence

Abstract

With the increasing awareness of human beings in the pursuit of human health, running in sports has become a fashionable and healthy first choice. The research uses artificial intelligence technology to conduct intelligent analysis when running training, and aims at the use of artificial intelligence technology. Artificial intelligence technology can accurately analyze and predict the application requirements of sports training postures. We proposed a set of sports posture analysis and predict system to design in this paper. It uses the running training record data in the watch heart rate and GPS smart sports watch, and uses Recurrent Neural Network (RNN), Long and short-term memory (LSTM) and Gate recursive unit (GRU) that three types of neural network models to predict whether the road race can be in the conference. And confirm it will be completed within the scheduled closing time, and it will also perform intelligent analysis of physical fitness (heart rate, pace) and running technology (cadence, pace). The training and test data for this study are the running training records from (Running distance, time, heart rate, stride frequency, stride length, pace, calories, altitude and other characteristic values) as input parameters to test and compare the running completion time trend of RNN, LSTM, GRU neural network models in the exercise table predictive power. The results show predict accuracy is the best is GRU method, and the worst is LSTM method. After the hidden layers are added to the three predict models, RNN is slightly regressive, LSTM has a significant trend of improvement, and GRU is less obvious.

Analysis of MEMS sensor application and prospects

“…Chemical MEMS sensors mainly refer to gas sensors that measure the gas composition in automobile systems. Bio-MEMS sensors are widely used in the field of vehicle driving safety, such as predicting driving fatigue [8].…”