Sampling trajectory data-based vehicle periodic queuing length estimation method and device

Abstract

The invention discloses a sampling trajectory data-based vehicle periodic queuing length estimation method and device. The method includes the following steps that: sampling vehicle trajectory data are acquired, the arrival distribution characteristics of vehicles are obtained by means of trajectory data superposition; on the basis of the arrival distribution characteristics of the vehicles, withthe compliance of the arrival of the vehicles to time-varying Poisson distribution considered, an average arrival rate during a vehicle time period is estimated; the queuing length range of each period is determined on the basis of the parking position of the last queued sampling vehicle and the passage time of the first non-queued sampling vehicle; and a probability theory model is established according to the arrival distribution and queuing length range of the vehicles, the likelihood function of queuing length is inferred, and the maximum likelihood estimation method is adopted to obtain asolution. According to the sampling trajectory data-based vehicle periodic queuing length estimation method and device of the invention, the arrival distribution characteristic of the vehicles are obtained on the basis of historical trajectory data, and therefore, the method and device are suitable for conditions of extremely low permeability. Simulation and empirical verification show that the method has high estimation precision for periodic queuing length.

Description

Technical field [0001] The invention relates to the field of traffic control, and in particular to a method and device for estimating the length of vehicle periodic queues based on sampled trajectory data. Background technique [0002] As the throat of the city, the operation status of the intersection directly affects the operation status of its associated road sections. When traffic congestion occurs, the congestion will gradually spread, and the itinerary will have a chain reaction, which may even lead to the paralysis of the entire road network. It can be seen that optimizing the operating conditions of intersections plays an important role in improving urban traffic. However, before optimization, an indispensable step is the estimation of the traffic state at the intersection. Common indicators for measuring traffic status at intersections include flow, delays, queuing, and parking times. At present, the main data sources for traffic state estimation at signal-controlled i...

AI Technical Summary

Problems solved by technology

However, the existing probabilistic methods either require additional information such as vehicle arrivals, or require training a large number of model parameters, which are difficult to put into practice, and the accuracy is also difficult to guarantee under low penetration rates.

Images