WO2012024957A1 - Method for fusing real-time traffic stream data and device thereof - Google Patents

Abstract

A method for fusing real-time traffic stream data and a device thereof are provided. The method comprises: determining the trust level of the road in various road condition states; when the difference between the trust level of the road condition state with the highest trust level and the trust level of the road condition state with the secondary high trust level is not less than a preset threshold, adopting a road condition state S with the highest trust level as the current road condition state of the road, and working out the current running speed of the road according to a corresponding running speed V of the road under the road condition state S with the highest trust level; and otherwise, recomputing the current running speed of the road according to the corresponding running speed V of the road under each type of real-time traffic stream data, and determining the current road condition state of the road according to the current running speed.

Description

 The present invention claims the priority of the Chinese patent application entitled "Real-time traffic flow data fusion method and device" submitted by the Chinese Patent Office on August 23, 2010, with the application number 201010260603.7, The entire contents are incorporated herein by reference.

Technical field

 The present invention relates to the field of intelligent transportation technologies, and in particular, to a real-time traffic flow data fusion method and apparatus. Background technique

Intelligent transportation systems use computers, communications, artificial intelligence, sensors and other technologies to provide users with real-time traffic information on the road. During the driving process, the user can know the current traffic conditions of each road section through GPS/GIS, broadcasting, information release board, etc. The traffic management department can know the traffic conditions of each road section through the vehicle sensors and video cameras on the road. At any time, the traffic signals of various traffic intersections are adjusted and information is released to the outside world. At present, the prior art mainly obtains traffic conditions of roads by separately processing various traffic flow data such as FCD (Floating Car Data), traffic flow induction coil data or event information data, and then selecting the traffic conditions of the road. The road condition information obtained by processing the data is filled as the current road condition information of the road. For example, the road condition information obtained by processing the FCD is selected as the current road condition information of the road. However, the inventors have found that the prior art separately treats various traffic flow data separately, and then selects one of the traffic flow data to obtain the road condition information as the current road condition of the road, if the selected traffic flow data occurs. Anomalies can result in reduced accuracy of road conditions derived from traffic flow data of this type.

Summary of the invention

 Embodiments of the present invention provide a real-time traffic flow data fusion method and apparatus, which improves the accuracy of road condition information.

 In order to achieve the above object, embodiments of the present invention use the following technical solutions:

 A real-time traffic flow data fusion method includes:

 Calculating, according to each of the real-time traffic flow data of the at least two real-time traffic flow data, the road state s and the running speed of the road under each of the real-time traffic flow data

V;

 Determining, in turn, a trust degree corresponding to each of the real-time traffic flow data;

 Determining the trust of the road in various road conditions;

 When the difference between the trust state of the road state with the highest degree of trust and the trust state of the road state with the second highest trust state is not less than a preset threshold, the road state S with the highest trust state is used as the current road state state of the road. And calculating a current running speed of the road according to the running speed V corresponding to the road in the road state S with the highest trust state;

Trust in the state of the road with the highest degree of trust and trust in the state of the road with the highest degree of trust The difference between the degrees is less than a preset threshold, and the current running speed of the road is recalculated according to the corresponding running speed V of the road under each of the real-time AC data, and the current running speed is determined according to the current running speed. The current traffic status of the road.

 A real-time traffic flow data fusion device includes:

 a first processing unit, configured to sequentially calculate, according to each real-time traffic flow data of the at least two real-time traffic flow data, a road state S and a running speed V corresponding to the road under each of the real-time traffic flow data ;

 a second processing unit, configured to sequentially determine a trust level corresponding to each of the real-time traffic flow data;

 a determining unit, configured to determine a trust degree of the road in each road state;

 a state fusion unit, configured to use, when the degree of trust of the road state with the highest degree of trust, and the degree of trust of the road state with the second highest degree of trust are not less than a preset threshold, using the state of the road state with the highest trust as the The current road state of the road, and the current running speed of the road is calculated according to the running speed V corresponding to the road in the state of the road state S with the highest trust state; the speed fusion unit is used for the road state with the highest trust degree The difference between the trust degree of the state and the trust state of the road state with the second highest degree of trust is less than a preset threshold, and the road is recalculated according to the corresponding running speed V of the road under each of the real-time exchange data. The current running speed, and determining the current road state of the road based on the current running speed.

The real-time traffic flow data fusion method provided by the embodiment of the present invention is calculated by sequentially calculating Corresponding to the traffic state S and the running speed V of each of the real-time traffic flow data, and sequentially determining the trust degree corresponding to each of the real-time traffic flow data, and determining that the road is in each road state Trust. Then, if the difference between the trust degree of the most trusted road state state and the trust degree of the second trustworthy road state state is not less than a preset threshold, then the road state s with the highest trust degree is used as the current state of the road a road condition state, and calculating a current running speed of the road according to the running speed V corresponding to the road in the road state S having the highest trust state; otherwise, according to the road, each of the real-time AC data is The lower corresponding running speed V recalculates the current running speed of the road, and determines the current road state of the road according to the current running speed.

 It can be seen from the implementation process of the embodiment of the present invention that the current road state and the running speed of the road are calculated by combining at least two real-time traffic flow data, and the at least two traffic flow data respectively correspond to different trust degrees. Compared with the prior art, the road condition information obtained by selecting one of the traffic flow data is used as the current road condition information of the road, and the embodiment of the present invention can effectively utilize the accuracy of various real-time traffic flow data, thereby improving the road condition information of the road. The accuracy.

BRIEF DESCRIPTION OF THE DRAWINGS The accompanying drawings, which are to be regarded as Without the creative work, you can also obtain other attachments based on these drawings. Figure.

 1 is a flow chart of a real-time traffic flow data fusion method according to Embodiment 1 of the present invention;

 2 is a schematic diagram of an operation speed and a road condition state of acquiring a road by using a coil according to Embodiment 1 of the present invention;

 3 is a flow chart of another real-time traffic flow data fusion method according to Embodiment 1 of the present invention;

 4 is a structural diagram of a real-time traffic flow data fusion device according to Embodiment 2 of the present invention; FIG. 5 is a structural diagram of another real-time traffic flow data fusion device according to Embodiment 2 of the present invention.

detailed description

 The technical solutions in the embodiments of the present invention are clearly and completely described in conjunction with the drawings in the embodiments of the present invention. It is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.

 Example 1

The embodiment shown in FIG. 1 provides a real-time traffic flow data fusion method, which specifically includes the following steps: 1 01. Calculate, according to each real-time traffic flow data of the at least two real-time traffic flow data, a road state S and a running speed V corresponding to the road under each of the real-time traffic flow data.

 In actual application, when the real-time traffic flow data is acquired, at least two real-time traffic flow data for fusion may be aggregated in a certain period of time. Since the selection of the time period is related to the frequency of publication of real-time traffic flow data. If the time period is selected to be shorter, the amount of real-time traffic flow data acquired will be limited, and the probability of data having extreme values will increase. If the time period is selected too long, it will affect the traffic flow information release, for example, Update traffic flow information in a timely manner. Therefore, a preferred solution is that the value of the above time period can be selected as 5 minutes.

 It should be noted that the real-time traffic flow data mentioned in the embodiments of the present invention includes: floating car data, traffic flow induction coil data or event information data.

Different types of data have different data characteristics, such as: FCD (F l oa ti ng Ca r Da ta , floating car data) is characterized by large amount of information, wide coverage, and accuracy is greatly affected by the floating car processing model. The higher the accuracy of the road information, the higher the accuracy of the FCD, and the higher the accuracy of the information. For example, the data characteristic of the traffic flow induction coil data is that the information is mainly for a highway, and the coverage is small, but the accuracy is high. For example, the event information data may be specifically classified into two types: traffic flow information and event information, the traffic flow information may be used for traffic flow information fusion, and the event information may be used for verification. Most of this type of data needs to be manually entered, with higher The accuracy, however, generally has a certain time delay. In particular, when an unexpected event occurs, the impact of the event information data on the traffic flow is not well judged.

 Therefore, according to the different types of real-time traffic flow data described above, the method for calculating the corresponding road state S and the running speed V of the road under each of the real-time traffic flow data is different, and the following describes how to calculate the road in the FCD, traffic, respectively. Corresponding road state S and running speed V under flow induction coil data and event information data.

 First, use FCD to calculate the road condition S and the running speed V corresponding to the road.

 Step 1. First classify the FCD according to the data source and the vehicle I D (each floating car has its own unique number).

 Step 2: Then, extract the GPS latitude, longitude, speed, direction, and time information in the FCD, and filter the abnormal data in the FCD. The abnormal data to be filtered includes:

 Single point data, which means that there is only one GPS point data for the same vehicle I D .

 Abnormal data, which refers to the time difference between two adjacent GPS points of the same vehicle I D multiplied by the maximum speed value is less than the linear distance between the two GPS points.

 Step 3. Through map matching, the path is estimated to get the path corresponding to all GPS points of each vehicle. Specifically, it can be realized by the following methods: Selecting the possible matching roads based on the latitude and longitude coordinates of the GPS points, generally there are multiple, filtering the roads that are too large with the angle of the road through the direction of the GPS points, and the time passing through the GPS points The order and the road connection are determined to match the road.

Step 4, after the above three steps, the road set through which each floating car passes can be determined, so that The running speed V of the road is obtained by the actual running length, the running time, and the instantaneous speed of the GPS point, and the road state S is determined according to the running speed V.

 Second, use the traffic flow induction coil data to calculate the road state corresponding to the road S and running speed

V.

 Generally, the traffic flow induction coil is generally divided into a single coil and a double coil, and the double coil is taken as an example to illustrate the speed extraction.

 As shown in FIG. 2, when the vehicle passes through two adjacent loop coils, the vehicle detector can obtain the time T1 passing through the first coil and the time T2 passing through the second coil, respectively, assuming the two adjacent coils. The actual distance is D, and the speed value V of the vehicle can be calculated from V = D / ( T1 - T2 ).

 Obtaining, in unit time, the speed values of all the vehicles passing through the two adjacent coils, and calculating an average value of the speed values of all the passing vehicles, and using the average value as the running speed V of the road in the unit time, The road state S is obtained based on the operating speed.

3. Calculate the road condition state S and the running speed V corresponding to the road by using the event information data. Event information data is usually collected by hand, and can be divided into internal collections and field collections. The internal collection mainly collects traffic information from FM stations or collects data through video observations. The field collection mainly depends on the collection. Personnel manually visualize the specific road traffic flow conditions. Both of these methods can directly obtain a more accurate road state S, but the running speed V is generally based on the road condition. It should be noted that the state of the road condition mentioned in the embodiment of the present invention includes: smooth, slow, or congested. According to the operating speed V, the road state S is specifically: when the running speed V is less than 20 km/h, determining that the road state S is congested; when the running speed V is greater than or equal to 20 km/h and less than 40 km/h, determining that the road state S is slow; When the running speed V is greater than or equal to 40 km/h, it is determined that the road condition state S is unblocked.

 102. Determine, in turn, a trust degree corresponding to each of the real-time traffic flow data.

 Specifically, this step can be implemented by the following sub-steps (not shown):

 102 A. Calculate, in sequence, a state accuracy rate of each of the real-time traffic flow data in a preset time range;

 102B. Determine a trust degree corresponding to each real-time traffic flow data according to a conversion formula and a state accuracy rate of each of the real-time traffic flow data within a preset time range. The conversion formula is: trust degree = [state accuracy rate *1000+0.5], where [ ] is a rounding symbol.

 103. Determine a degree of trust of the road under various road conditions.

上述表一中的 "FCD1" 表示从公司 1获取的浮动车数据， "FCD2" 表 示从公司 2 获取的浮动车数据， "线圈" 表示交通流感应线圈数据， "事 件" 表示事件信息数据。 Assume that in the application scenario 1, at least two kinds of real-time traffic flow data in step 101 are specifically FCD acquired from company 1, FCD acquired from company 2, traffic flow induction coil data, and event information data. Assume that the running speed and road condition corresponding to the road G calculated by using the above four real-time traffic flow data are as follows:

"FCD1" in Table 1 above represents floating car data obtained from company 1, "FCD2" represents floating car data obtained from company 2, "coil" represents traffic flow induction coil data, and "event" represents event information data.

Assume that step 102 determines that the trustworthiness of each of the four real-time traffic flow data is as shown in Table 2:

 Then, the trust degree of the road in the road condition state in this step is: the sum of trust degrees corresponding to all real-time traffic flow data used when calculating the road state. The following describes in detail how to calculate the trust of the road under various road conditions.

 For example, according to Table 1, it can be known that the real-time traffic flow data used in the unblocked state is "FCD1", and according to Table 2, the trust degree corresponding to the "FCD1" is 6, so that the The trust of road G in the unblocked state is 6;

For another example, according to Table 1, it can be known that the real-time traffic flow data used in the slow state is "FCD2", "coil" and "event". According to Table 2, the trust corresponding to "FCD2" can be known. The degree is 5, the "coil" corresponds to a degree of trust of 10, and the "event" corresponds to a degree of trust of 9, then the trust of the road in a slow state is: all that is used when calculating the slow state The sum of the trusts corresponding to the real-time traffic flow data, that is, 5 + 10+9=24. According to Table 1 obtained in step 101 and Table 2 obtained in step 102, this step can determine the trust degree of the road G in each road state as shown in Table 3 below:

 104. When the difference between the trust degree of the most trusted road state state and the trust state of the second trustworthy road state state is not less than a preset threshold, the road state S having the highest trust degree is used as the current road condition of the road. a state, and calculating a current running speed of the road according to the running speed V corresponding to the road in the road state S with the highest degree of trust.

 Specifically, taking the application scenario 1 in step 103 as an example, the preset threshold may be calculated according to the formula F = ( M1+M2+.. +Mn ) / n , where n represents real-time traffic flow The number of data, Mn represents the trust degree of the nth real-time traffic flow data. According to the above Table 2, the preset threshold can be calculated as F = (6+5 + 10+9) /4=7.5.

 It can be seen from the above table 3 that the state of the road with the highest trust is slow, the corresponding trust degree is 24, the state of the road with the second highest trust is unblocked, and the corresponding trust degree is 6; then the highest trust can be obtained. The difference between the trust of the traffic state and the trust of the traffic state with the second highest trust is: 24-6 = 18.

At this time, since 18>7.5, that is, the difference between the trust degree of the road state with the highest degree of trust and the trust state of the road state with the second highest trust state is not less than a preset threshold, it may be determined that the current road state S of the road G is slow. The current running speed V of the road G is calculated by the following process: 值 using the weighted average value of the road running speed V corresponding to the road state with the highest trust state as the current state of the road The running speed, wherein the weight value is a trust degree corresponding to the real-time traffic flow data used when calculating the running speed V.

 Specifically, according to Table 1, it can be known that the corresponding running speeds of the road G in the slow state are 21 km/h, 29 km/h and 27 km/h, respectively, and the calculation results are "21 km/h". The real-time traffic flow data is "FCD2", and the real-time traffic flow data used when calculating "29 km/h" is "coil", and the real-time traffic flow data used when "27 km/h" is calculated is " Event "; according to Table 2, the "FCD2" corresponds to a degree of trust of 5, the "coil" corresponds to a degree of trust of 10, and the "event" corresponds to a degree of trust of 9. Then, the current running speed of the road G is V = ( 21 X 5 + 29 X 10+27 χ 9 ) / (5 + 10+9) = 26.6 km/h.

 105. The difference between the trust degree of the highest trust state of the road state and the trust state of the road state state with the second highest trust state is less than a preset threshold, according to the running speed of the road under each of the real-time AC data. Recalculating the current running speed of the road and determining the current road state of the road according to the current running speed.

 In order to explain the implementation process of the step 105 more clearly, it is assumed that in the application scenario 2, the running speed and the road state corresponding to the road W calculated by using the four real-time traffic flow data obtained by the above step 101 are as shown in Table 4 below:

 Data Name FCD1 FCD2 Coil Event

The state of the road w is smooth and slow, and the speed of the passage W (km/h) 45 30 21 50 Table 4

 Among them, "FCD 1, represents the floating car data obtained from company 1, "FCD2" represents the floating car data obtained from company 2, "coil" represents traffic flow induction coil data, and "event" represents event information data.

 It is assumed that, under the application scenario 2, the trust degree corresponding to each of the four kinds of real-time traffic flow data is the same as that in the above table 2, and the preset threshold used in the application scenario 2 can be calculated according to the above table 2. Specifically, F = ( 6+5 + 1 0+9 ) / 4 = 7. 5.

Then, the trust degree of the road W that can be determined through step 103 in each road state is as shown in Table 5 below:

 Table 5

 It can be known from Table 5 above that the state of the road with the highest degree of trust is unblocked, and the corresponding trust degree is 1 5; the state of the road with the second highest degree of trust is slow, and the corresponding trust degree is also 15 . The difference is less than the preset threshold of 7.5, and the difference may be less than the preset threshold of 7.5. A corresponding running speed V under real-time AC data is recalculated to obtain a current running speed of the road, and a current road condition state of the road is determined according to the current running speed.

Specifically, a weighted average of the running speed V of the road under each of the real-time AC data may be used as the current running speed of the road, where the weight value is In order to calculate the trust degree corresponding to the real-time traffic flow data used when the running speed V is obtained.

 The current running speed of the road raft is calculated by taking the application scenario 2 as an example. According to Table 4 and Table 2, the running speed V of the road w under the four kinds of real-time AC data is: 45 km. /h, 30 km/h, 21 km/h, 50 km/h. Calculate the real-time traffic flow data used for the "45 km/h" as "FCD1", and the corresponding trust degree is 6; calculate the real-time traffic flow when the "30 km/h" is used. The data is "FCD2", and its corresponding trust degree is 5; the real-time traffic flow data used when the "21 km/h" is calculated is "coil", and the corresponding trust degree is 10; The real-time traffic flow data used in the description of "50 km/h" is "event", and its corresponding trust degree is 9. Then, the current running speed of the road W is V = ( 45 X 6 + 30 X 5 + 21 X 10 + 50 χ 9 ) / (6 + 5 + 10 + 9) = 36 km / h, since the running speed is "36km" The corresponding road condition state of /h" is "slowman", so that it can be determined that the current road state S of the road W is "slow".

 The embodiment of the present invention obtains different trust degrees according to the state accuracy of different real-time traffic flow data, and obtains the current degree of trust by analyzing the trust distribution of the road in each road state and the weighted average of the trust. Running speed and traffic status. Compared with the prior art, the road condition information obtained by one of the traffic flow data is used as the current road condition information of the road, and the embodiment of the present invention can effectively utilize the accuracy of various real-time traffic flow data, thereby improving the road condition information of the road. accuracy.

Further, as shown in FIG. 3, the foregoing method may further include the following step 106: 106. Verify, by using the event information data, the current road state and the current running speed of the calculated road.

 Re-verification using event information data is mainly to verify the restricted class information. For example, when a restricted traffic event occurs on a road, the road should not have traffic flow information. For example, when an unexpected event that causes congestion is present, the threshold value of the running speed corresponding to the road state can be lowered by referring to the speed value on the road, so that the state tends to be congested.

 Example 2:

 As shown in FIG. 4, an embodiment of the present invention provides a real-time traffic flow data fusion apparatus, including: a first processing unit 11, a second processing unit 12, a determining unit 13, a state fusion unit 14, and a speed fusion unit 15.

 The first processing unit 11 is configured to sequentially calculate, according to each real-time traffic flow data of the at least two real-time traffic flow data, a road state S and a corresponding road condition corresponding to each real-time traffic flow data. Speed V;

 The second processing unit 12 is configured to sequentially determine the reliability corresponding to each of the real-time traffic flow data;

The determining unit 13 is configured to determine the trust degree of the road in each road condition state; the state fusion unit 14 is configured to: when the trust degree of the road state with the highest trust degree and the trust state of the road state state with the second highest trust degree are not smaller than Presetting a threshold value, using the road state S having the highest degree of trust as the current road state of the road, and according to the road, the trust degree is the most The running speed V corresponding to the high road condition S calculates the current running speed of the road; the speed fusing unit 15 is used for the trust degree of the road state with the highest trust degree and the trust state of the road state with the second highest trust degree. The difference is less than the preset threshold, and the current running speed of the road is recalculated according to the corresponding running speed V of the road under each of the real-time AC data, and the road is determined according to the current running speed. Current traffic status.

 Further, the second processing unit may perform function subdivision (not shown), and specifically includes: a calculation module and a conversion module.

 The calculation module is configured to sequentially calculate a state accuracy rate of each of the real-time traffic flow data within a preset time range; and the conversion module is configured to sequentially perform the data according to the conversion formula and each of the real-time traffic flow data. The state accuracy rate in the time range is determined to determine the trust degree corresponding to each of the real-time traffic flow data; the conversion formula is: trust degree = [state accuracy rate * 1 000 + 0.5.

 It should be noted that, in a specific application, the state fusion unit specifically uses the value obtained by weighting and averaging the corresponding running speed V of the road in the road state with the highest trust degree as the current running speed of the road. Wherein, the weight value is a trust degree corresponding to the real-time traffic flow data used when calculating the running speed V.

The speed fusing unit specifically uses the weighted average of the running speed V of the road under each of the real-time AC data as the current running speed of the road, wherein the weight value is calculated and calculated. The letter corresponding to the real-time traffic flow data used when the running speed V is used Ten degrees.

 Further, as shown in FIG. 5, the above apparatus may further include: an inspection unit 16.

 The checking unit 16 is configured to verify the current road state and the current running speed of the calculated road using the event information data.

 The real-time traffic flow data fusion device provided by the embodiment of the present invention combines at least two real-time traffic flow data to calculate the current road state and the running speed of the road, and the at least two traffic flow data respectively correspond to different trust degrees. Compared with the prior art, the road condition information obtained by selecting one of the traffic flow data is used as the current road condition information of the road, and the embodiment of the present invention can effectively utilize the accuracy of various real-time traffic flow data, thereby improving the road condition information of the road. The accuracy.

 The embodiments of the present invention are mainly applied to the process of integrating real-time traffic flow data, and can improve the accuracy of the road condition information of the road.

 The above is only the specific embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily think of changes or substitutions within the technical scope of the present invention. It should be covered by the scope of the present invention. Therefore, the scope of the invention should be determined by the scope of the appended claims.

Claims

Claim  A real-time traffic flow data fusion method, comprising:  Calculating, according to each of the real-time traffic flow data of the at least two kinds of real-time traffic flow data, the road state S and the running speed V corresponding to the road under each of the real-time traffic flow data; determining each of the Trustworthiness corresponding to real-time traffic flow data;  Determining the trust of the road in various road conditions;  When the difference between the trust degree of the road state with the highest degree of trust and the trust state of the road state with the second highest degree of trust is not less than a preset threshold, the road state S having the highest degree of trust is adopted as the current road state of the road, and Calculating a current running speed of the road according to the running speed V corresponding to the road in the road state S with the highest degree of trust;  When the difference between the trust degree of the road state with the highest degree of trust and the trust state of the road state with the second highest trust state is less than a preset threshold, the road is recalculated according to the corresponding running speed V of each of the real-time AC data. Deriving a current running speed of the road, and determining a current road state of the road according to the current running speed.  The real-time traffic flow data fusion method according to claim 1, wherein the real-time traffic flow data comprises: floating car data, traffic flow induction coil data or event information data;  The road conditions include: unblocked, slow, or congested. 3. The real-time traffic flow data fusion method according to claim 1, wherein: Determining, in turn, determining the trust degree corresponding to each of the real-time traffic flow data includes:  Calculating, in turn, the state accuracy of each of the real-time traffic flow data within a preset time range;  Determining the trust degree corresponding to each of the real-time traffic flow data according to the conversion formula and the state accuracy rate of each of the real-time traffic flow data within a preset time range;  The conversion formula is: Trust = [state accuracy rate * 1 000 + 0. 5 ].  The real-time traffic flow data fusion method according to claim 1, wherein the trust degree of the road in the road state is: corresponding to all real-time traffic flow data used when calculating the road state The sum of trust.  The real-time traffic flow data fusion method according to claim 1, further comprising:  The current road state and the current running speed of the calculated road are verified using event information data.  The real-time traffic flow data fusion method according to claim 1, wherein the preset threshold is calculated according to a formula F = ( M1 + M2 + . . + Mn ) / n, wherein n represents real-time The number of traffic flow data, Mn represents the trust degree of the nth real-time traffic flow data; Calculating, according to the running speed V of the road in the road state with the highest degree of trust, the current running speed of the road includes: using the road corresponding to the running in the road state with the highest trust degree The value obtained by weighted averaging of the velocity V as the road The front running speed, wherein the weight value is a trust degree corresponding to the real-time traffic flow data used when calculating the running speed V;  Recalculating the current running speed of the road according to the running speed V corresponding to the road under each of the real-time alternating current data includes: using the road corresponding to each of the real-time alternating current data The operating speed V is a weighted average obtained as the current running speed of the road, wherein the weight value is a trust degree corresponding to the real-time traffic flow data used when calculating the running speed V.  7. A real-time traffic flow data fusion device, comprising:  a first processing unit, configured to sequentially calculate, according to each real-time traffic flow data of the at least two real-time traffic flow data, a road state S and a running speed V corresponding to the road under each of the real-time traffic flow data ;  a second processing unit, configured to sequentially determine a trust level corresponding to each of the real-time traffic flow data;  a determining unit, configured to determine a trust degree of the road in each road state; a state fusion unit, configured to use the road state S with the highest trust as the road when the difference between the trust state of the road state with the highest degree of trust and the trust state of the road state with the second highest trust state is not less than a preset threshold. Current road condition state, and calculating a current running speed of the road according to the running speed V corresponding to the road in the road state S with the highest trust state; a speed fusion unit, configured to be the state state with the highest trust state Trust and trust The difference in the degree of trust of the road state is less than a preset threshold, and the current running speed of the road is recalculated according to the corresponding running speed V of the road under each of the real-time AC data, and according to the current The operating speed determines the current road state of the road.  The real-time traffic flow data fusion device according to claim 7, wherein the second processing unit comprises:  a calculation module, configured to sequentially calculate a state accuracy rate of each of the real-time traffic flow data within a preset time range;  a conversion module, configured to determine a trust degree corresponding to each real-time traffic flow data according to a conversion formula and a state accuracy rate of each of the real-time traffic flow data within a preset time range;  The conversion formula is: Trust = [state accuracy rate * 1 000 + 0. 5].  The real-time traffic flow data fusion device according to claim 7, further comprising: a verification unit, configured to use the event information data to check the current road state and the current running speed of the calculated road. The real-time traffic flow data fusion device according to claim 7, wherein the state fusion unit specifically adopts a weighted average of the roads corresponding to the running speed V in the road state with the highest trust degree. The value of the road is the current running speed of the road, wherein the weight value is a trust degree corresponding to the real-time traffic flow data used when calculating the running speed V; The speed fusing unit specifically adopts a weighted average of the running speed V of the road under each of the real-time AC data as a current running speed of the road, wherein the weight value is calculated and calculated. The trust level corresponding to the real-time traffic flow data used when running the speed V.