Real-time traffic speed prediction method, device and electronic equipment

Abstract

The invention provides a real-time traffic speed prediction method, a device, electronic equipment and a computer readable storage medium. The method comprises the following steps: obtaining current weather conditions and time characteristics; obtaining an attribute feature z of each road link; obtaining a current speed average value r' on each road link and a historical speed average value r in apreset time range corresponding to the current weather conditions and time characteristics; obtaining a current average traffic flow g' of each predetermined area of an urban road network and a historical average traffic flow g in a predetermined time range corresponding to the current weather conditions and time characteristics; and predicting the real-time traffic speed of all the roads Link according to the attribute feature z, the current speed average value r', the historical speed average value r, the current average traffic flow g' and the historical average traffic flow g. The real-time traffic speed prediction method can accurately predict the driving traffic speed of different roads at different moments in real time.

Description

technical field [0001] The invention relates to the field of vehicles, in particular to a real-time traffic speed prediction method, device, electronic equipment and computer-readable storage medium. Background technique [0002] The driving speed of motor vehicles on each road link accurately reflects the current traffic conditions on the road, and can be used to estimate the driving time of a given starting and ending point and evaluate the traffic load capacity of the corresponding road. Existing schemes or empirical data are used to fill in (corresponding roads 2 / 3 of the maximum allowable speed of the road as the estimated speed), or perform conventional matrix decomposition based on the historical average speed of each road link. [0003] The empirical filling method cannot reflect the specific characteristics and real-time characteristics of different roads. The conventional matrix decomposition method may cause the decomposition to obtain negative speed values ​​that...

AI Technical Summary

Problems solved by technology

[0003] The empirical filling method cannot reflect the specific characteristics and real-time characteristics of different roads. The conventional matrix decomposition method may decompose and obtain negative speed values ​​that require additional adjustments, resulting in slow iterative convergence and poor interpretability.

Images