The particulate matter emissions of vehicles exhaust has become the main source of urban air pollution, which seriously affects ambient air quality and human health. Based on the total particulate matter (TPM) emissions data of municipal vehicles in China from 2011 to 2015, this paper explores the spatio-temporal evolution characteristics of TPM emissions of vehicles in China from emissions and emission increments by using spatial analysis method, and then quantitatively evaluates the influence intensity of main driving factors by using geographical detector model. The results show that the TPM emissions of vehicles in China have been decreasing year by year, and provincial capitals and municipalities are the main contributors to the reduction of TPM emissions of vehicles in China. The TPM emissions and emission reductions of vehicles in provincial capitals and municipalities are the largest, and their annual averages are more than twice the national municipal average. Both the TPM emissions and emission reductions of vehicles in China showed a decreasing trend from the eastern coast to the western inland, and the spatial distribution characteristics of "low emission, and high increase; high emission, and high reduction". The "low emission and high increase" region is mainly concentrated in the central and western of China, especially in the southwest provinces. The "high emission and high reduction" area is the pan-North China Plain with Beijing-Tianjin-Hebei region as the core. The number and spatial distribution range of the High-High cluster and Low-Low cluster areas of TPM emissions of vehicles have been reduced year by year, and the spatial agglomeration has declined. On the contrary, the trend of spatial random distribution has strengthened. The Low-High outlier areas have separated the concentrated and contiguous distribution of High-High cluster areas, so the spatial distribution pattern of fragmentation in the High-High cluster areas has become increasingly prominent. The analysis of geographical detection shows that the TPM emissions of vehicles are the most affected by the number of vehicles, followed by the use intensity of vehicles, and the least affected by natural environmental factors such as average annual temperature and altitude. The natural environmental factors such as annual average temperature and altitude significantly enhance the driving explanatory power of TPM emissions of vehicles through the interaction with human activities. Exploring the spatio-temporal heterogeneity and driving factors of vehicle particulate emissions is of great significance to improve the accuracy of vehicle exhaust control in China.