Abstract

When the era of artificial intelligence comes in full swing, it is more important for primary and secondary schools to grasp the focus of teaching reform in the process of physical education reform and combine it with the actual situation to achieve teaching innovation. In this paper, we combine the characteristics of chaotic system such as randomness and ergodicity to improve the population diversity, thus forming a scheduling algorithm based on chaotic genetic algorithm. The improved genetic algorithm can standardize the physical education curriculum and complete the scheduling of it. Based on the calculation of the fitness function, conflicting lesson planning schemes can be eliminated. This ensures both the population size and the diversity of the population. Then some physical education courses in a university in Guiyang for the academic year 2021-2022 are used as experimental data, and the population size is now set to 100, and the number of iterations are 50, 100, 150, 200, 250 and 300, and the simulation experiments are analyzed. The results show that the chaotic genetic algorithm, however, reaches equilibrium at 150, converges quickly, and its fitness value is higher than that of the genetic algorithm, so the application of chaotic genetic algorithm in physical education innovation is feasible and effective. This study perfectly achieves the goal of physical education innovation by developing a reasonable curriculum plan.