An Automatic Generation Method For Personalized Fitness Running Scheme Based On Genetic Algorithm

Fitness running is the main exercise of the national fitness program, which can improve people's heart and lung function, aerobic metabolism capacity and the body's immune system. How to formulate a personalized exercise plan to ensure that both improving the effectiveness of exercise and avoiding their physical and psychological damage caused by excessive exercise is the key problem need to be solved in fitness running. To solve this problem, a method for evolutionary modeling is proposed. On this basis, an automatic generation method for personalized fitness running scheme based on genetic algorithm is proposed.The main works of this paper are listed as follows:(1) A method named EAKFA is proposed in this paper, which can effectively eliminate the random noise of MEMS gyro. EAKFA can improve the accuracy of the acceleration and speed. The effectiveness of EAKFA is validated by comparing EAKFA with a general adaptive Kalman filter algorithm. The results show that the error mean and the standard deviation of the MEMS gyroscope are reduced by 2.5% and 55.7% after filtered, respectively.(2) An evolutionary modeling method for fitness running process is proposed. The method uses genetic algorithm to optimize the parameters of the motion model defined by Cheng and Scalzi based on the motion data, and then builds a personalized motion model. The test results of 10 volunteers show that the average value of error variance of the method proposed in this paper is reduced by 63.6%compared with Levenberg-Marquardt algorithm.(3) An automatic generation method for fitness running scheme is proposed in this paper. Based on the personal fitness running model, the specified exercise time, the safe and effective heart rate interval and other parameters, the generation of the optimal fitness running scheme is abstracted into optimization problems. This paper uses genetic algorithm to solve this problem, and then generates the optimal fitness running scheme automatically. The test results of 10 volunteers show that the times of the heart rate drop in the safe and effective interval in the optimal fitness running scheme generated in this paper increased 42.6% compared with the running process without any guidance of running scheme.(4) This paper designs a fitness running system based on Android smartphones and Baidu cloud platform. This system can provide services such as the motion data acquisition, motion modeling and generating the optimal fitness running scheme.The automatic generation method for motion model and personalized fitness running scheme proposed in this paper can provide useful support for improving exercise efficiency and avoiding excessive exercise.