Conditional random field based real time traffic state estimation method

Abstract

The invention discloses a conditional random field based real time traffic state estimation method. Through learning of history data, the corresponding relation between change condition of a speed sequence in a motion process of a single vehicle and road jam degree is constructed. A conditional random field model is adopted for learning the implication relation between the change condition of thespeed sequence in the motion process of the single vehicle and road jam degree and the learning result is a corresponding conditional random field model. After the model is constructed, real time accurate estimation on traffic condition of a specific road segment can be realized through speed sequence data of a single vehicle in the specific road segment acquired in real time. Therefore, strict requirements for data acquisition re reduced and a good estimation effect can be achieved in a comparatively simple collection condition (such as that only the speed sequence data of a single vehicle isneeded).

Description

technical field [0001] The invention relates to the field of intelligent traffic systems, in particular to a real-time traffic state estimation method based on a conditional random field. Background technique [0002] With the development of society, especially the continuous progress of science and technology, the development represented by computer intelligence and communication informatization has brought about earth-shaking changes in people's lives. With the development of society, especially the convenience of transportation, private cars have entered thousands of households, and driving has become an important part of most social families and work units. Where it has already appeared is very important. [0003] In real life, it has become a common way to arrange travel plans based on various traffic conditions information provided by maps. At present, Baidu, AutoNavi, and Tencent Maps all provide a wealth of road condition references. These road condition informati...

AI Technical Summary

Problems solved by technology

[0004] However, the traffic information provided by these systems must be realized based on rich data collection systems.

When there is only one or a few vehicle trajectories in such data on some road sections, less data information may lead to a large deviation in road condition estimation, and even a deviation from the destination; especially for wild environments or sparsely populated villages Or the mountainous area far away from the city will cause delays in travel planning for map users, and even life-threatening situations; therefore, how to achieve accurate estimation of traffic conditions under the condition of a small amount of data is very important for improving the display of road conditions on the map. Accuracy is of great significance to improve customer travel experience

Images