CN109979197B

CN109979197B - Method and system for constructing highway traffic time map based on fusion data

Info

Publication number: CN109979197B
Application number: CN201910270545.7A
Authority: CN
Inventors: 李敏; 张红龙; 刘良伟; 张睿; 赵章宗; 邓朝栋
Original assignee: Chongqing Tongtochi Information Technology Co Ltd
Current assignee: Chongqing Tongtochi Information Technology Co Ltd
Priority date: 2019-04-04
Filing date: 2019-04-04
Publication date: 2021-01-29
Anticipated expiration: 2039-04-04
Also published as: CN109979197A

Abstract

The invention provides a method and a system for constructing a highway traffic time map based on fusion data, which can predict the traffic condition of a highway through highway traffic data (video monitoring data and toll station traffic data), explore and improve the utilization rate of the highway traffic data to the maximum extent, fully consider the continuity of traffic flow and be beneficial to improving the prediction accuracy; in addition, the traffic development situation is a decision basis for traffic managers to implement management and control, and the traffic development situation can help traffic management departments to make decisions scientifically, actively and intelligently, and is favorable for realizing intelligent traffic.

Description

Method and system for constructing highway traffic time map based on fusion data

Technical Field

The invention relates to the technical field of intelligent traffic, in particular to a method and a system for constructing a highway traffic time map based on fusion data.

Background

By the end of 2017, the highway in China reaches 13.65 kilometers, and the highway is the first in the world and occupies half of the traffic mileage of the highway in the world. The highway infrastructure is established in the aspect of promoting the social and economic development of China, and plays a vital leading and basic supporting role and a service effect.

China's expressway intelligentization and informatization is vigorously constructed and is still continuously carried out, and the market scale is estimated to be close to 900 hundred million yuan by 2023 years. The two main subsystems of the highway intelligent system are as follows: intelligent charging system and security protection monitored control system. According to incomplete statistics, 8 thousands of video monitoring resources are deployed and built by each level of highway administration and management departments in the country, and 25505 toll stations (no harbor, Australia platform, and origin from Baidu maps) are arranged in the whole country.

With the continuous increase of networking operation mileage and the continuous expansion of operation management coverage of the expressway in China, people put forward higher expectations on traffic technology, hope that traffic can step into the era of intelligent traffic, make correct prediction on the future traffic flow of the expressway on the basis of mastering the traffic flow of the expressway, and the method is an important condition for traffic construction and stepping into the field of intelligent traffic in China. The prediction of the traffic flow is a research hotspot in the field of intelligent transportation, and the accurate prediction result can provide decision and support for traffic control of traffic management departments.

With the arrangement and the use of a large number of traffic video devices along the highway and the networking of a toll collection system, real-time traffic video data and toll station traffic data are uninterruptedly transmitted to a highway traffic management department, the traffic data is increased explosively, and in addition, the analysis and judgment of video signals in the past mainly depend on human decisions, so that the problems of highway management and service in China are increasingly highlighted. The perception capability of the expressway network is insufficient, the traffic prediction early warning means is insufficient, and the like. The traffic video monitoring data and toll station passing data are fully utilized, passive information collection to active monitoring of an expressway manager in the management process is achieved, the decision maker is changed from passive management to active and humanized management, and the expressway road traffic passing efficiency and the traveling experience of travelers are improved.

Disclosure of Invention

The invention provides a method and a system for constructing a highway traffic time map based on fusion data, which mainly solve the technical problems that: how to more accurately predict the traffic behavior of a highway.

In order to solve the above technical problems, the present invention provides a highway traffic time map system based on fusion data, comprising: the system comprises a toll station supervision subsystem, a road supervision subsystem and a traffic situation prediction subsystem;

the toll station supervision subsystem comprises toll stations arranged at each entrance and exit of a target highway, each toll station comprises an identification module, a recording module, a data analysis module and a first storage module, and the identification module is used for identifying license plates and vehicle types of vehicles passing through the toll stations; the recording module is used for recording the arrival time and the departure time of each vehicle; the data analysis module is used for realizing the tracking of the vehicles on the basis of the license plates of the vehicles, respectively calculating the running time of the vehicles, and calculating the first average speed of the vehicle according to the running time of the vehicle and the distance between the vehicle on-station charging station and the vehicle off-station charging station; for each vehicle of the same incoming toll station and the same outgoing toll station, the same incoming toll station and the same outgoing toll station are adjacent toll stations, and according to the first average speed of each vehicle and the vehicle type corresponding to each vehicle, all vehicles in the same vehicle type are counted, and the second average speed of each vehicle in the target road section is counted; the target road section is a road section between the same entrance toll station and the same exit toll station; the first storage module is used for storing the license plate, the vehicle type, the arrival time, the departure time, the first average speed and the second average speed of each vehicle;

the road supervision subsystem comprises one or more video monitoring points, an image processing module and a second storage module, wherein the one or more video monitoring points, the image processing module and the second storage module are distributed along the target highway, and each video monitoring point is used for acquiring a traffic video image in an effective monitoring area of the video monitoring point in real time; the image processing module is used for calculating the single vehicle speed of each vehicle in the effective monitoring area of each video monitoring point and the third average speed of each vehicle based on the traffic video images acquired by the video monitoring points; the second storage module is used for storing the traffic video image, the bicycle speed and the third average speed;

the traffic situation prediction subsystem comprises a setting module, a first acquisition module, a second acquisition module, a processing module and a construction module, wherein the setting module is used for receiving an external setting instruction so as to set an initial observation time point T0, a rolling prediction period T and a prediction time T;

the first obtaining module is used for obtaining vehicles entering from the toll stations at present from the toll station supervision subsystem according to the rolling prediction period T to serve as first vehicles to be detected, obtaining a second average speed matched with the vehicle type and the driving direction of the first vehicles to be detected, and taking the matched second average speed as a first target speed V1;

the second obtaining module is configured to obtain, from the road supervision subsystem, the vehicle speed, the third average speed, and the third average speed of the downstream video monitoring point adjacent to the video monitoring point of each vehicle in the effective monitoring area currently acquired by each video monitoring point according to the rolling prediction period T, and use each vehicle in the effective monitoring area currently acquired by the video monitoring point as a second vehicle to be measured; determining a second target speed V2 of the second vehicle to be tested based on the bicycle speed, the third average speed and the third average speed of the adjacent downstream video monitoring points;

the processing module is configured to load the current positions of the first to-be-tested vehicles and the second to-be-tested vehicles into a traffic network of the target highway, predict a first travel distance S1 of each first to-be-tested vehicle after a predicted time t based on the first target speed V1, and predict a second travel distance S2 of each second to-be-tested vehicle after the predicted time t based on the second target speed V2;

the building module is configured to load the first vehicle to be detected to a first target position in a traffic network of the target highway according to the first driving distance S1, and load the second vehicle to be detected to a second target position in the traffic network of the target highway according to the second driving distance S2, so as to generate a highway traffic time map.

Optionally, the toll station further includes a first direction prediction module, configured to obtain a latest historical passage record of the first vehicle to be detected passing through the toll station, and if the latest historical passage record is that the toll station is taken as an entry station, take an exit station of the latest historical passage record as the traveling direction of the current passage; and if the latest historical traffic record is that the toll station is taken as an outbound, taking the inbound of the latest historical traffic record as the driving direction of the current traffic.

Optionally, the toll station further includes a second direction prediction module, configured to obtain a historical passage record of the first vehicle to be detected passing through the toll station, count occurrence probabilities of passing in different directions according to the historical passage record, and select a traveling direction corresponding to the maximum occurrence probability as the traveling direction of the current passage.

Optionally, the second obtaining module is configured to determine whether a third average speed of the downstream video monitoring point exists, if yes, calculate a magnitude relationship between the third average speed and the third average speed of the downstream video monitoring point, and calculate the second target speed V2 according to the magnitude relationship and the bicycle speed; if not, the bicycle speed is taken as the second target speed V2.

Optionally, the second obtaining module is configured to, when it is determined that a third average speed of the downstream video monitoring point exists, calculate a first difference between the bicycle speed and the third average speed of the downstream video monitoring point, and when the first difference is greater than 0, calculate an average between the bicycle speed and the third average speed of the downstream video monitoring point, where the average is used as the second target speed V2; and when the first difference is less than or equal to 0, setting the bicycle speed as the second target speed V2.

Optionally, the second obtaining module is configured to, when it is determined that the third average speed of the downstream video monitoring point exists, calculate a first ratio between the third average speed of the downstream video monitoring point and the third average speed, use the first ratio as an influence weight of a future traveling speed of the second vehicle to be detected, and calculate a product between the vehicle speed and the influence weight, which is used as the second target speed V2.

Optionally, the second obtaining module is configured to, when it is determined that a third average speed of the downstream video monitoring point exists, calculate an absolute value of a second difference between the third average speed and the third average speed of the downstream video monitoring point, calculate a second ratio between the absolute value and the third average speed, and when the second ratio is smaller than or equal to a preset threshold, take the bicycle speed as the second target speed V2; and when the second ratio is greater than the preset threshold, calculating an average value between the bicycle speed and a third average speed of the downstream video monitoring point, and taking the average value as the second target speed V2.

Optionally, the second obtaining module is configured to, when it is determined that the third average speed of the downstream video monitoring point exists, directly calculate an average value of the vehicle speed of the second vehicle to be detected and the third average speed of the downstream video monitoring point, and use the average value as the second target speed V2.

Optionally, the traffic situation prediction subsystem further includes an alarm module, configured to determine a traffic density in each monitoring interval after the prediction time t, and perform an alarm when the traffic density reaches a preset traffic density threshold.

The invention also provides a method for constructing the highway traffic time map system based on the fusion data, which comprises the following steps:

setting an initial observation time point T0, a rolling prediction period T and a prediction time T;

according to the rolling prediction period T, acquiring vehicles entering the toll stations from the toll station supervision subsystem to serve as first vehicles to be detected, acquiring a second average speed matched with the types and driving directions of the first vehicles to be detected, and taking the matched second average speed as a first target speed V1;

according to the rolling prediction period T, acquiring the vehicle speed and the third average speed of each vehicle in the effective monitoring area of each video monitoring point currently acquired by each video monitoring point from a road supervision subsystem, and taking each vehicle in the effective monitoring area of each video monitoring point currently acquired by the video monitoring point as a second vehicle to be detected; determining a second target speed V2 of the second vehicle to be tested based on the bicycle speed, the third average speed and the third average speed of the adjacent downstream video monitoring points;

loading the current positions of each first vehicle to be tested and each second vehicle to be tested into a traffic network of the target highway, predicting a first running distance S1 of each first vehicle to be tested after a predicted time t based on the first target speed V1, and predicting a second running distance S2 of each second vehicle to be tested after the predicted time t based on the second target speed V2;

and loading the first vehicle to be tested to a first target position according to the first driving distance S1 corresponding to the first target position in the traffic network of the target highway, and loading the second vehicle to be tested to a second target position according to the second driving distance S2 corresponding to the second target position in the traffic network of the target highway, so as to generate a highway traffic time map.

The invention has the beneficial effects that:

according to the method and the system for constructing the highway traffic time map based on the fusion data, provided by the invention, the traffic condition of the highway is predicted through the highway traffic data (video monitoring data and toll station traffic data), the utilization rate of the highway traffic data is explored and improved to the maximum extent, the continuity of traffic flow is fully considered, and the prediction accuracy is favorably improved; in addition, the traffic development situation is a decision basis for traffic managers to implement management and control, and the traffic development situation can help traffic management departments to make decisions scientifically, actively and intelligently, and is favorable for realizing intelligent traffic.

Drawings

Fig. 1 is a schematic structural diagram of a highway traffic time map system according to a first embodiment of the present invention;

FIG. 2 is a schematic structural diagram of another highway traffic time map system according to a first embodiment of the present invention;

FIG. 3 is a schematic diagram of a highway traffic time map according to a first embodiment of the present invention;

fig. 4 is a flowchart illustrating a method for constructing a traffic time map of a highway according to a second embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following detailed description and accompanying drawings. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

The first embodiment is as follows:

in order to realize accurate prediction of the traffic operation condition of the highway and provide decision and support for traffic control and dispersion for the traffic management part, the embodiment provides a scheme of a highway traffic time map system, the traffic condition prediction at the future prediction time is realized by fusing the video data of the monitoring points of the current expressway network and the traffic data of toll stations, loading the forecast information to the highway network to form a highway traffic time map, wherein the highway traffic time map is the change trend of traffic conditions on the road network along with time, and based on the time map, background management personnel of a traffic management department, the traffic flow distribution condition of the future prediction time can be intuitively known, and an alarm and control plan can be conveniently and timely made, so that the traffic jam condition is relieved, the processing efficiency of abnormal events is improved, and intelligent traffic is realized.

Referring to fig. 1, fig. 1 is a schematic structural diagram of a highway traffic time map system provided in this embodiment, where the highway traffic time map system 100 includes a toll station supervision subsystem 110, a road supervision subsystem 120, and a traffic situation prediction subsystem 130; wherein

The toll booth supervision subsystem 110 includes toll booths established at respective entrances and exits of the target highway, which include an identification module 111, a recording module 112, a data analysis module 113, and a first storage module 114.

In this embodiment, the target highway may be selected according to a road section that needs to be predicted actually, and the road section that needs to be predicted actually may be, for example, a road section with a relatively congested traffic condition. The target highway can be a whole highway section, or a part of highway sections selected from the whole highway section, or all highway sections in the whole city of a certain city, or all highway sections in the whole country.

The recognition module 111 is used for recognizing license plates and vehicle types of all vehicles passing through the toll station; including but not limited to, minicars, minivans, medium sized customers, large sized customers, medium sized vans, large vans, and the like.

The recording module 112 is used for recording the arrival time and the departure time of each vehicle passing through the toll station.

The data analysis module 113 is configured to track the entering and exiting of each vehicle based on a license plate of each vehicle, calculate the running time of each vehicle, and calculate a first average speed of the vehicle according to the running time of the vehicle and a distance between the entering toll station and the exiting toll station of the vehicle. See table 1 below:

TABLE 1

The data analysis module 113 is further configured to, for each vehicle at the same incoming toll station and the same outgoing toll station, where the same incoming toll station and the same outgoing toll station are adjacent toll stations, count all vehicles in the same vehicle type and a second average speed in the target road section according to the first average speed of each vehicle and the vehicle type corresponding to each vehicle. The target road section is a road section between the same entrance toll station and the same exit toll station.

Optionally, the data analysis module 113 is further configured to count second average speeds of different vehicle types in the target road segment in different time periods. Such as daytime, evening, peak and off-peak or other typical hours. This may make the prediction more accurate.

Taking the above table 1 as an example, assuming TOLL stations TOLL1 and TOLL2 are adjacent TOLL stations, the data analysis module 113 can obtain a second average speed of the car on the target road (the road from TOLL1 to TOLL2) based on the historical traffic data as follows: 100km/h +83km/h is 91.5 km/h.

In practical application, for the same vehicle type with the same entrance toll station and the same exit toll station, the actually calculated second average speed can more accurately reflect the average speed of the vehicle type.

The first storage module 114 is used for storing the license plate, the model, the arrival time, the departure time, the first average speed and the second average speed of each vehicle passing through the toll station.

The toll station further includes a first direction prediction module 115, please refer to fig. 2, where the first direction prediction module 115 is configured to obtain a last past traffic record of the first vehicle to be detected passing through the toll station, and if the last past traffic record is that the toll station is taken as an entrance, an exit of the last past traffic record is taken as a traveling direction of the current traffic; and if the latest historical passage record is that the toll station is taken as an outbound, taking the inbound of the latest historical passage record as the traveling direction of the current passage.

For example, when a vehicle "hush a 1234" entering from the TOLL station TOLL1 is currently used as the first vehicle to be tested, the direction prediction module 115 may obtain a historical passage record of the last time the vehicle passes through the TOLL station TOLL1, where the TOLL station TOLL1 may be an exit or an entry, and when the vehicle enters from the TOLL station TOLL1 last time, the vehicle is likely to pass in the same direction as the previous pass, i.e., travel in the direction of the previous exit; if the TOLL station TOLL1 is left for the last time, the TOLL station TOLL1 is likely to be driven back to the original destination, that is, the vehicle is driven in the direction of the last passing to the original destination. Based on the mode, the driving direction of the first vehicle to be detected can be accurately predicted to a certain extent, but the method cannot be completely guaranteed.

In practical application, when the first vehicle to be tested does not have a historical passage record passing through the toll station, a randomly determined mode can be adopted to determine the driving direction of the time. Or determined in other ways known in the art.

The embodiment also provides a method for predicting the driving direction, which comprises the following steps: the toll station further comprises a second direction prediction module 116, which is used for acquiring a historical passing record of the first vehicle to be detected passing through the toll station, counting the occurrence probability of different directions according to the historical passing record, and selecting the driving direction corresponding to the maximum occurrence probability as the driving direction of the current passing.

Counting the historical passing records of the first vehicle to be tested, determining the driving rule of the first vehicle to be tested, determining the probability of the first vehicle to be tested in different directions during the current walking and distributing the probability; for example, if the number of times of entering from the toll station is 10, where 9 times is the direction of walking a, the probability of walking in the direction of walking a at this time is 90%, where 1 time is the direction of walking B at this time is 10%. The a direction may be selected as the current direction of travel. If the rule is not obvious, determining the driving direction according to the latest historical traffic record; and then randomly assigned again or without recording.

Optionally, a monitoring camera may be disposed at a lane of the ramp after entering the toll station, so as to accurately capture the driving direction of the first vehicle to be detected.

The road supervision subsystem 120 includes one or more video surveillance points 121 deployed along the target highway, an image processing module 122, and a second storage module 123.

Each video monitoring point 121 is used for acquiring a traffic video image in each effective monitoring area in real time.

The image processing module 122 is configured to calculate, based on the traffic video images acquired by the video monitoring points 121, a single vehicle speed of each vehicle in an effective monitoring area of each video monitoring point 121 and a third average speed of each vehicle passing through the effective monitoring area.

Optionally, the video image frame currently acquired by the video monitoring point 121 and the previous frame (actually, not limited to the previous frame) of the current video image frame are acquired according to the rolling prediction period T, and since there may be multiple vehicles in the previous and next frame images, the corresponding relationship between the vehicle in the previous frame image and the vehicle in the next frame image needs to be identified, that is, the vehicle needs to be tracked, so that the single vehicle speed of each vehicle to be detected can be accurately obtained.

Optionally, based on the previous frame image and the current frame image, the coincidence degree of the vehicle image is obtained, and vehicle tracking can be achieved. The higher the coincidence of the images of the two vehicles, the more likely it is that the same vehicle is. Then, based on the movement of the pixel point position of the vehicle to be detected in the front frame image and the rear frame image, determining the actual movement distance of the vehicle to be detected; according to the image acquisition frame rate of the video camera, the time difference (namely the moving time) between two frames of images can be determined, so that the bicycle speed of the vehicle to be detected can be calculated according to the actual moving distance and the actual moving time.

And obtaining a third average speed of each first vehicle to be detected under the monitoring point after calculating the vehicle speed of each first vehicle to be detected under the monitoring point. Assuming that three first vehicles to be tested exist under the monitoring point and the single vehicle speeds are a1, a2 and a3 respectively, the third average speed of each vehicle under the monitoring point is (a1+ a2+ a 3)/3.

It should be understood that the above-mentioned manner of calculating the bicycle speed of the first vehicle under test is not limited thereto, and other manners may be adopted as long as the bicycle speed of each first vehicle under test can be determined.

It should be noted that, when there is no first vehicle to be detected in the currently effective monitoring area of the video monitoring point 121, the marking process may be performed.

The second storage module 123 is configured to store the traffic video images collected by the video monitoring point 121, the bicycle speed of each first vehicle to be detected, and the third average speed.

The traffic situation prediction subsystem 130 includes a setting module 131, a first obtaining module 132, a second obtaining module 133, a processing module 134, and a construction module 135.

The setting module 131 is configured to receive an external setting instruction to set a starting observation time point T0, a rolling prediction period T, and a prediction time T.

For a target highway to be predicted, considering the influence of the arrangement distance of monitoring points on a road network, a monitoring blind area exists between adjacent monitoring points, and vehicles in the monitoring blind area cannot be monitored by the two adjacent monitoring points. Therefore, when the highway traffic time map is constructed, the vehicles in the monitoring blind areas are omitted, so that the constructed highway traffic time map has large errors. For this reason, the present embodiment reduces the influence of such an error by setting the initial observation time point t0 in advance. Specifically, the selected principle of the initial observation time point t0 is: and the influence of the error of the initial state on the traffic flow density prediction is controlled to the maximum extent at the moment when the vehicles on the target expressway are minimum. For example, 3 am is selected as the initial observation time point t 0.

The rolling prediction period T can be flexibly set according to the system processing performance, the monitoring data volume of the target highway or the requirements of traffic management departments, if the system processing performance is high, the monitoring data volume is small, and the rolling prediction period T can be set to be small, so that the prediction frequency is high, and the prediction accuracy is improved; on the contrary, if the system processing performance is low and the monitoring data volume is large, the rolling prediction period T needs to be set relatively large, so as to avoid the problem of paralysis caused by heavy system processing load.

The scroll prediction period T is set to, for example, 1 second, 10 seconds, 1 minute, 5 minutes, 10 minutes, or the like.

The first obtaining module 132 is configured to obtain, from the toll station monitoring subsystem 110, a vehicle currently entering from each toll station as a first vehicle to be tested according to the rolling prediction period T, obtain a second average speed matched with a vehicle type and a driving direction of the first vehicle to be tested, and use the matched second average speed as the first target speed V1.

When the second average speed is counted in different vehicle types in different time periods, the matched second average speed also needs to be considered in the time period. For example, Monday rush hour should refer to the second average vehicle speed of the same vehicle type in historical Monday rush hour.

It should be noted that the toll station supervision subsystem 110 may obtain the historical traffic data to obtain a second average speed of different vehicle types on the target road section. Such as a car, the second average speed of the target road segments in one direction (e.g., TOLL stops TOLL 1-TOLL 2) is 91.5km/h, and the second average speed of the target road segments in the other direction (e.g., TOLL stops TOLL 1-TOLL 3) is 90 km/h. When the model of the first vehicle to be tested arriving from the TOLL station TOLL1 is a car and the driving direction is TOLL1, the first obtaining module 132 can determine that the second average speed matched with the first vehicle to be tested is 91.5 km/h.

It should be understood that typically a toll station has two directions of travel, but this embodiment is not limiting, and when there are multiple directions of travel, the toll station supervisory subsystem 110 determines a second average speed in each direction for different vehicle types based on historical traffic records. Similarly, the first obtaining module 132 can determine the second average speed matching with the vehicle type in the driving direction only according to the vehicle type and the driving direction of the first vehicle to be tested (the current time period also needs to be known during the time period statistics).

A second obtaining module 133, configured to obtain, according to the rolling prediction period T, a vehicle speed and a third average speed of each vehicle in the effective monitoring area currently acquired by each video monitoring point 121 from the road supervision subsystem 120, and use each vehicle in the effective monitoring area currently acquired by the video monitoring point 121 as a second vehicle to be measured; and determining a second target speed V2 of the second vehicle to be tested based on the speed of the second vehicle, the third average speed and the third average speed of the adjacent downstream video monitoring points.

Optionally, the second obtaining module 133 is configured to determine whether the third average speed of the downstream video monitoring point exists, if so, calculate a magnitude relationship between the third average speed of the video monitoring point 121 and the third average speed of the downstream video monitoring point, and calculate a second target speed V2 of the second vehicle to be detected according to the magnitude relationship and the vehicle speed of the second vehicle to be detected; if the third average speed does not exist in the adjacent downstream video monitoring points, it is indicated that the vehicle to be tested does not exist in the effective monitoring area of the downstream video monitoring points, and therefore the corresponding road section passing condition of the downstream video monitoring points can be considered to be smooth, and therefore the vehicle speed of the second vehicle to be tested can be used as the second target speed V2 for prediction.

The third average speed of the video surveillance point 121 may reflect the vehicle passing status in the current effective surveillance area of the video surveillance point 121, and the third average speed of the adjacent downstream video surveillance point may also reflect the vehicle passing status in the effective surveillance area of the downstream video surveillance point. The continuity characteristic of the traffic flow is comprehensively considered, and the current traffic condition and the downstream traffic condition are combined to obtain the second target speed for prediction, so that the prediction result is more accurate.

Optionally, the second obtaining module 133 is configured to, when it is determined that the third average speed of the adjacent downstream video monitoring point exists, directly calculate an average value of the single vehicle speed of the second vehicle to be detected and the third average speed of the adjacent downstream video monitoring point, and use the average value as the second target speed V2.

Optionally, the second obtaining module 133 is configured to, when it is determined that the third average speed of the downstream video monitoring point exists, calculate a first difference between the single vehicle speed of the second vehicle to be detected and the third average speed of the downstream video monitoring point, when the first difference is greater than 0, that is, when the magnitude relationship is that the single vehicle speed of the second vehicle to be detected is greater than the third average speed of an adjacent downstream monitoring point, calculate an average value between the single vehicle speed of the second vehicle to be detected and the third average speed of the downstream video monitoring point, and use the average value as the second target speed V2 of the second vehicle to be detected; when the first difference is less than or equal to 0, that is, the magnitude relationship is that the bicycle speed of the second vehicle to be detected is less than or equal to the third average speed of the adjacent downstream monitoring point, it indicates that the downstream road section is relatively more smooth, and the passing of the second vehicle to be detected of the video monitoring point 121 is not greatly affected, so that the bicycle speed is taken as the second target speed V2.

Optionally, the second obtaining module 133 is configured to, when it is determined that the third average speed of the adjacent downstream video monitoring point exists, calculate a first ratio between the third average speed of the downstream video monitoring point and the third average speed of the video monitoring point, use the first ratio as an influence weight of a future traveling speed of the second vehicle to be detected, and calculate a product between a vehicle speed and the influence weight, which is used as the second target speed V2.

Optionally, the second obtaining module 133 is configured to, when it is determined that the third average speed of the adjacent downstream video monitoring point exists, calculate an absolute value of a second difference between the third average speed of the current video monitoring point and the third average speed of the downstream video monitoring point, calculate a second ratio between the absolute value and the third average speed of the current video monitoring point, and when the second ratio is less than or equal to a preset threshold, indicate that a traffic condition of a road segment where the current video monitoring point is located is equivalent to a traffic condition of a road segment where the adjacent downstream video monitoring point is located, at this time, take a vehicle speed of a second vehicle to be detected as the second target speed V2; when the second ratio is greater than the preset threshold, the difference between the current road section and the downstream road section is more, or the current road section is smooth to pass, and the downstream is congested, or the current road section is congested and the downstream is smooth; at this time, an average value between the vehicle speed of the second vehicle to be measured and the third average speed of the downstream video monitoring point is calculated, and the average value is taken as a second target speed V2.

The preset threshold value can be flexibly set according to actual conditions, for example, set to be 5%, 10%, 20%, and the like.

The processing module 134 is configured to load the current positions of each first vehicle to be tested and each second vehicle to be tested into a traffic network of the target highway, predict a first travel distance S1 of the first vehicle to be tested after the predicted time t based on the first target speed V1, and predict a second travel distance S2 of the second vehicle to be tested after the predicted time t based on the second target speed V2.

Wherein S1 ═ V1 × t; s2 ═ V2 × t. A plurality of first vehicles to be tested generally exist, and the corresponding first target speeds V1 are generally different due to different entrance toll stations, different driving directions, different vehicle types and the like; similarly, for each second vehicle to be tested, the corresponding second target speed V2 may also be different because the vehicle speed, the third average speed of the current video monitoring point, and the third average speed of the adjacent downstream video monitoring point are different.

The building module 135 is configured to load a corresponding first vehicle to be detected to a first target position in a traffic network of the corresponding target highway according to the first travel distance S1; and loading a corresponding second vehicle to be tested to a second target position in a traffic network of the corresponding target highway according to the second driving distance S2 so as to generate the highway traffic time map.

Optionally, with continued reference to fig. 2, the traffic situation prediction subsystem 130 may further include an alarm module 136, configured to determine the traffic density in each monitoring interval after the predicted time t, and alarm when the traffic density reaches a preset traffic density threshold. It should be understood that traffic density refers to the amount of traffic per length of road segment.

Optionally, with continued reference to fig. 2, the traffic situation prediction subsystem 130 may further include a presentation module 137 for presenting a highway transit time map.

Referring to fig. 3, fig. 3 is a schematic view of a traffic time map of an expressway provided in the present embodiment:

with respect to the first to-be-measured vehicle a1, it is predicted that the vehicle will travel from the position K1 to the position K1' (the first target position) after the predicted time t; for the second vehicle under test a2, after the predicted time t, it is predicted that the vehicle will travel from the position K2 to the position K2' (the second target position). Through the traffic time map, firstly, the accurate prediction of traffic conditions is realized, and in addition, the traffic time map is displayed in a form of the traffic time map, so that the traffic time map is visual and clear.

The highway traffic time map system provided by the embodiment realizes the construction of a highway traffic time map and the estimation of traffic density according to the video data of the monitoring points of the current highway network and the traffic data of toll stations, and based on the highway traffic time map, a traffic management department can conveniently know and predict the traffic condition of a target highway at any time within the time t; meanwhile, the traffic time map is displayed in a form of a traffic time map, so that the traffic time map is more visual and clear.

Example two:

on the basis of the first embodiment, the embodiment provides a method for constructing a traffic time map of an expressway, please refer to fig. 4, which mainly includes the following steps:

s401, setting a starting observation time point T0, a rolling prediction period T and a prediction time T.

S402, according to the rolling prediction period T, obtaining vehicles entering the toll station from the toll station supervision subsystem as first vehicles to be tested, obtaining a second average speed matched with the vehicle type and the driving direction of the first vehicles to be tested, and taking the matched second average speed as a first target speed V1.

Optionally, the latest historical passage record of the first vehicle to be detected passing through the toll station is obtained, and if the latest historical passage record is that the toll station is taken as an entrance station, the exit station of the latest historical passage record is taken as the traveling direction of the current passage; and if the latest historical passage record is that the toll station is taken as an outbound, taking the inbound of the latest historical passage record as the traveling direction of the current passage.

It should be noted that the toll station supervision subsystem may obtain historical traffic data to obtain a second average speed of different vehicle types on the target road section. Such as a car, the second average speed of the target road segments in one direction (e.g., TOLL stops TOLL 1-TOLL 2) is 91.5km/h, and the second average speed of the target road segments in the other direction (e.g., TOLL stops TOLL 1-TOLL 3) is 90 km/h. When the model of the first vehicle to be tested arriving from the TOLL station TOLL1 is a car and the driving direction is TOLL2, then it can be determined that the second average speed matched with the first vehicle to be tested is 90 km/h.

S403, according to the rolling prediction period T, acquiring the vehicle speed and the third average speed of each vehicle in the effective monitoring area of each video monitoring point currently acquired by the video monitoring points from the road supervision subsystem, and taking each vehicle in the effective monitoring area of each video monitoring point currently acquired by the video monitoring points as a second vehicle to be detected; and determining a second target speed V2 of the second vehicle to be tested based on the speed of the second vehicle, the third average speed and the third average speed of the adjacent downstream video monitoring points.

Optionally, determining the second target speed V2 of the second vehicle under test based on the vehicle speed, the third average speed, and the third average speed of the adjacent downstream video monitoring point includes:

judging whether the third average speed of the adjacent downstream video monitoring point exists, if so, calculating the magnitude relation between the third average speed of the video monitoring point and the third average speed of the adjacent downstream video monitoring point, and calculating a second target speed V2 of a second vehicle to be detected according to the magnitude relation and the single vehicle speed of the second vehicle to be detected; if the third average speed does not exist in the adjacent downstream video monitoring points, it is indicated that the vehicle to be tested does not exist in the effective monitoring area of the downstream video monitoring points, and therefore the corresponding road section passing condition of the downstream video monitoring points can be considered to be smooth, and therefore the vehicle speed of the second vehicle to be tested can be used as the second target speed V2 for prediction.

The third average speed of the video monitoring point can reflect the vehicle passing condition in the current effective monitoring area of the video monitoring point, and the third average speed of the adjacent downstream video monitoring point can also reflect the vehicle passing condition in the effective monitoring area of the adjacent downstream video monitoring point. The combination of the current traffic condition and the downstream traffic condition to obtain the second target speed V2 for prediction can make the prediction result more accurate.

Optionally, when it is determined that the third average speed of the adjacent downstream video monitoring point exists, an average value of the single vehicle speed of the second vehicle to be detected and the third average speed of the adjacent downstream video monitoring point is directly calculated, and the average value is used as the second target speed V2.

Optionally, when it is determined that the third average speed of the adjacent downstream video monitoring point exists, a first difference between the third average speed and the third average speed of the adjacent downstream video monitoring point is calculated, and when the first difference is greater than 0, that is, when the magnitude relationship is that the third average speed of the second vehicle to be measured is greater than the third average speed of the adjacent downstream video monitoring point, it indicates that the downstream road section is relatively congested, and a certain influence may be exerted on future passing of the second vehicle to be measured.

When the first difference is less than or equal to 0, namely the magnitude relation is that the third average speed of the second vehicle to be detected is less than or equal to the third average speed of the adjacent downstream monitoring point, the downstream road section is relatively smoother, and the passing of the second vehicle to be detected of the video monitoring point cannot be greatly influenced, so that the single vehicle speed is directly used as the second target speed V2.

Optionally, when it is determined that the third average speed of the adjacent downstream video monitoring point exists, a first ratio of the third average speed of the adjacent downstream video monitoring point to the third average speed of the video monitoring point is calculated, the first ratio is used as an influence weight of a future traveling speed of the second vehicle to be detected, and a product of the vehicle speed and the influence weight is calculated and used as a second target speed V2.

Optionally, when the third average speed of the adjacent downstream video monitoring point is determined to exist, calculating an absolute value of a second difference between the third average speed of the video monitoring point and the third average speed of the downstream video monitoring point, calculating a second ratio of the absolute value to the third average speed, and when the second ratio is less than or equal to a preset threshold, taking the bicycle speed of the second vehicle to be detected as a second target speed V2; and when the second ratio is greater than the preset threshold value, calculating an average value between the bicycle speed of the second vehicle to be detected and the third average speed of the downstream video monitoring point, and taking the average value as a second target speed V2.

It should be noted that the execution order of steps S402 and S403 is not limited, and both steps may be processed in parallel.

S404, loading the current positions of the first to-be-tested vehicles and the second to-be-tested vehicles into a traffic network of a target highway, predicting a first driving distance S1 of the first to-be-tested vehicles after the predicted time t based on the first target speed V1, and predicting a second driving distance S2 of the second to-be-tested vehicles after the predicted time t based on the second target speed V2.

Wherein S1 ═ V1 × t; s2 ═ V2 × t. A plurality of first vehicles to be tested generally exist, and due to different station charging stations, different driving directions or different vehicle types and the like, the first target speeds V1 corresponding to the first vehicles to be tested are generally different; similarly, for each second vehicle to be tested, since the vehicle speed, the third average speed of the current video monitoring point, and the third average speed of the adjacent downstream video monitoring point are different, the corresponding second target speed V2 is likely to be different.

S405, loading a first vehicle to be tested to a first target position in a traffic network of the target highway according to the first running distance S1, and loading a second vehicle to be tested to a second target position in the traffic network of the target highway according to the second running distance S2, so as to generate the highway traffic time map.

It will be apparent to those skilled in the art that the modules or steps of the invention described above may be implemented in a general purpose computing device, they may be centralized on a single computing device or distributed across a network of computing devices, and optionally they may be implemented in program code executable by a computing device, such that they may be stored on a computer storage medium (ROM/RAM, magnetic disks, optical disks) and executed by a computing device, and in some cases, the steps shown or described may be performed in an order different than that described herein, or they may be separately fabricated into individual integrated circuit modules, or multiple ones of them may be fabricated into a single integrated circuit module. Thus, the present invention is not limited to any specific combination of hardware and software.

The foregoing is a more detailed description of the present invention that is presented in conjunction with specific embodiments, and the practice of the invention is not to be considered limited to those descriptions. For those skilled in the art to which the invention pertains, several simple deductions or substitutions can be made without departing from the spirit of the invention, and all shall be considered as belonging to the protection scope of the invention.

Claims

1. A highway traffic time map system based on fusion data is characterized by comprising a toll station supervision subsystem, a road supervision subsystem and a traffic situation prediction subsystem;

2. The system according to claim 1, wherein the toll station further comprises a first direction prediction module for obtaining a last past passage record of the first vehicle to be detected passing through the toll station, and if the last past passage record is that the toll station is taken as an entrance, taking an exit of the last past passage record as the driving direction of the current passage; and if the latest historical traffic record is that the toll station is taken as an outbound, taking the inbound of the latest historical traffic record as the driving direction of the current traffic.

3. The system according to claim 1, wherein the toll station further comprises a second direction prediction module, configured to obtain a historical passage record of the first vehicle to be detected passing through the toll station, count occurrence probabilities of different directions according to the historical passage record, and select a driving direction corresponding to the maximum occurrence probability as the driving direction of the current passage.

4. The system according to claim 2 or 3, wherein the second obtaining module is configured to determine whether a third average speed of the downstream video surveillance point exists, if so, calculate a magnitude relationship between the third average speed and the third average speed of the downstream video surveillance point, and calculate the second target speed V2 according to the magnitude relationship and the vehicle speed; if not, the bicycle speed is taken as the second target speed V2.

5. The highway traffic time map system according to claim 2 or 3, wherein the second obtaining module is configured to calculate a first difference between the speed of the single vehicle and a third average speed of the downstream video monitoring point when judging that the third average speed of the downstream video monitoring point exists, and calculate an average between the speed of the single vehicle and the third average speed of the downstream video monitoring point when the first difference is greater than 0, as the second target speed V2; and when the first difference is less than or equal to 0, setting the bicycle speed as the second target speed V2.

6. The highway traffic time map system of claim 4, wherein the second obtaining module is configured to calculate a first ratio of a third average speed of the downstream video surveillance point to the third average speed when judging that the third average speed of the downstream video surveillance point exists, use the first ratio as an influence weight of a future traveling speed of the second vehicle under test, and calculate a product between the vehicle speed and the influence weight as the second target speed V2.

7. The highway traffic time map system of claim 4, wherein the second obtaining module is configured to, when determining that a third average speed of the downstream video surveillance point exists, calculate an absolute value of a second difference between the third average speed and the third average speed of the downstream video surveillance point, and calculate a second ratio of the absolute value to the third average speed, and when the second ratio is less than or equal to a preset threshold, take the vehicle speed as the second target speed V2; and when the second ratio is greater than the preset threshold, calculating an average value between the bicycle speed and a third average speed of the downstream video monitoring point, and taking the average value as the second target speed V2.

8. The system as claimed in claim 2 or 3, wherein the second obtaining module is configured to directly calculate an average value of the vehicle speed of the second vehicle under test and the third average speed of the downstream video monitoring point as the second target speed V2 when determining that the third average speed of the downstream video monitoring point exists.

9. The highway traffic time map system of claim 2 or 3, wherein the traffic situation prediction subsystem further comprises an alarm module for determining traffic density in each monitoring interval after the predicted time t and giving an alarm when the traffic density reaches a preset traffic density threshold.

10. A method for constructing a highway traffic time map based on fusion data is characterized by comprising the following steps: