Intelligent technologies have great potential for advancing physical education (PE), thus finding an appropriate independent PE plan (IPEP) is a key step in improving undergraduates' physical fitness (PF) level. In this study, an adaptive fitness enhancement model (AFEM) was designed based on senseless exercise behaviors monitoring (EBM) technology and an intelligent PE platform to explore the changes in PF levels of undergraduates with different fitness levels after receiving different IPEP. A total of 400 undergraduates participated in this study, and they were randomly assigned to four experimental groups and one control group. This study not only considered the historical performance of the undergraduates (fixed effect) but also delved into the individual differences of the undergraduates (random effect). The findings indicated that high-frequency aerobic exercise promoted endurance qualities more than low-frequency anaerobic exercise in an EBM environment, while low-frequency anaerobic exercise promoted strength qualities more than low-frequency anaerobic exercise. In this study, the initial decision-making mechanism of the AFEM model was developed based on the results of linear mixed model data analysis. The results showed that the AFEM model was able to maximize the effect of exercise, which in turn effectively improved the PF of undergraduates. At the same time, the AFEM model also adjusts the control variables according to the actual needs of the users, thus enriching the diversity of IPEP and further exploring the potential of the application of intelligent technology in personalized PE.