Aiming at the traffic saturation situation, a estimation method based on the maximum queuing length and initial queue of the period based on the maximum queuing length and initial queue of the period is proposed by using the vehicle arrival and parking position information provided by the real-time observation of the connected vehicle trajectory data during the period. Based on the trajectory of connected vehicles, the four critical points of vehicle arrival time, queuing time, start time and departure time are determined according to the arrival rate of each cycle according to the information provided by the critical point, and the initial queue length and maximum queuing length of each cycle are estimated by using shock wave theory, and the model is simulated and verified by the micro traffic simulation software SUMO. The experimental results show that under the permeability of 20%, the average absolute error between the estimated value of the queuing length model and the simulated value is 4.12 meters at the smallest, 7.83 meters at most, and the average relative error is at least 8.51% and 15.34% at most, indicating that even under the low permeability rate, the intersection queuing length estimation model based on vehicle trajectory can still estimate the maximum queuing length and initial queue length of supersaturated intersections more effectively.