With the continuous development of China's economy, the sales volume of automobiles across the country continues to grow. The vehicle driving cycle describes the vehicle speed time curve, which is used to determine vehicle pollutant emissions, fuel consumption, new model development and evaluation. It is a key basic technology in the automotive industry. Therefore, it is of great significance to carry out research on the driving cycle of vehicles on urban roads in China. Markov model is used to model the vehicle driving cycle in a given city. The average speed, average driving speed, average acceleration, average deceleration, idling time ratio, acceleration time ratio, deceleration time ratio, standard deviation of speed, standard deviation of acceleration and other index values are extracted for each kinematics segment, and these index values are input into the Markov chain to obtain the corresponding vehicle driving condition data. At the same time, the comprehensive evaluation system is constructed based on the calculated index values. We use the entropy weight method combined with the gray evaluation model to evaluate the vehicle driving cycle model. The evaluation results show that the Markov model of vehicle driving cycle proposed by us has a good effect.