Method for predicting travel time on urban ground level roads

Abstract

The invention provides a method for predicting travel time on urban ground level roads, which comprises the following steps of: matching the real-time traffic flow data collected by a road section traffic information collecting device and a signal control device with the arrival and discrete function of traffic stream in a traffic construction, and calculating the traffic flow in a certain time scope by a calculation model; uniformly inputting parameters such as the traffic flow, the saturation level, the road junction signal timing period, the split green ratio and the like in each travel direction of the road junction to calculate the travel time of the road section at the moment. The method can be used to predict the travel time in urban road network so as to select the optimum path with minimum travel time.

Description

technical field [0001] The invention relates to the field of traffic management, and in particular to a method for predicting travel time of urban ground roads. Background technique [0002] Today, as intelligent transportation informatization is getting more and more attention, how to improve the level of transportation services and make full use of limited road resources, thereby reducing the time and cost of transportation and saving energy, is a key issue for traffic management departments at all levels and intelligent transportation enterprises. and other issues. At present, large and medium-sized cities have built or are building large-scale traffic signal control systems, video surveillance systems, and electric police bayonet systems. For traffic participants, the information provided to them is far from enough, such as front-end traffic conditions, especially road congestion ahead, secondary information such as travel time from the departure point to the destinatio...

AI Technical Summary

Problems solved by technology

Although the travel time prediction method using license plate recognition and comparison collects the actual travel time of each vehicle, there will be the following problems: because not all lanes in the four directions of each intersection are equipped with license plate recognition equipment, and the license plate recognition The rate is not high, so the number of samples is very small, especially when there is no traffic or congestion, the number of available vehicle travel time samples is not enough, so that the accuracy requirements of travel time prediction cannot be met, and false positives are prone to occur; All lanes in one direction are equipped with license plate recognition equipment, which is too expensive and not realistic enough

Images