Treadmill-based Exercise Fitness Guidance Method For College Students

Exercising on indoor treadmill can avoid the influence of seasonal weather,and is so simple and efficient.It has become a major fitness method for college students.The existing treadmills in the market partly provide a lot of optional exercise programs based on domain knowledge,However,the current personalized guidance is still lack of enough consideration of physical features for college students’ fitness running.Consequently,it is an important issue to improve the effect of under the premise of ensuring exercise safety on the treadmill.To solve this problem,this paper proposes a treadmill-based exercise fitness guidance method for college students.The paper consists of 4 work packages as follows.(1)Propose a college students’ fitness exercise data collection method based on domain knowledge and treadmill.This method caters for the data-driven fitness running modeling requirements,and fully considers the domain knowledge of fitness running and the adjustable range of speed and slope of most treadmills.Consequently,a personalized fitness running data collection method is suitable for different college students’ personal physical characteristics and can keep their safety in exercise.(2)Design a data-driven approach to modeling college students’ personalized fitness running based on the domain knowledge.On the basis of data collection method,we construct a data-driven auto-regressive fitness running model.Then,we use three methods(Neural Network,Support Vector Regression and Random Forest)to integrate the auto-regressive fitness running model with the domain model in order to improve the accuracy of the model.(3)Design a hybrid coded-based ant colony algorithm to generate a personalized exercise program on treadmills for college students and.Based on the proposed modeling method.we abstract the generation of personalized fitness running program into a constrained optimization problem involving discrete variables and continuous variables.Further,we propose a constrained fitness evaluation method and a method for updating solution in order to design an efficient hybrid code-based ant colony algorithm.The exercise program generated by this algorithm can guild college students to safely and effectively exercise on the treadmill.(4)Carry out contrast experiments.We recruit ten male and female college students for experiments.The results show that the data collection method not only can avoid the excessive motion intensity,but also can ensure the uniform distribution of sample points.The integrated model method in this paper is better than the domain model and the data-driven auto-regressive model.The personalized exercise program is enough to help college students improve the effectiveness of exercise and fitness on the treadmill.