CN114879194A - Control method and device for multi-radar vehicle data homologous fusion - Google Patents
Control method and device for multi-radar vehicle data homologous fusion Download PDFInfo
- Publication number
- CN114879194A CN114879194A CN202111416029.4A CN202111416029A CN114879194A CN 114879194 A CN114879194 A CN 114879194A CN 202111416029 A CN202111416029 A CN 202111416029A CN 114879194 A CN114879194 A CN 114879194A
- Authority
- CN
- China
- Prior art keywords
- vehicle data
- vehicle
- data
- valid
- radar
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
- G01S13/931—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of land vehicles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Electromagnetism (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Networks & Wireless Communication (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The application provides a control method and a device for multi-radar vehicle data homologous fusion, wherein the method comprises the following steps: preprocessing the received vehicle data to obtain an effective data set; obtaining second valid vehicle data in the valid data set that is related to the first valid vehicle data; determining third valid vehicle data that is homologous to the first valid vehicle data from the second valid vehicle data; and when the third valid vehicle data exists, determining the target vehicle after data fusion and the target vehicle data after data fusion according to the proximity principle. The method is beneficial to improving the efficiency of data fusion, gives consideration to higher accuracy and smaller calculated amount, and can efficiently process the vehicle data of the fusion area of multi-radar detection in the intelligent traffic. And performing homologous recognition and fusion under the accuracy of the current traffic radar equipment, and fully utilizing the reported data to perform association degree judgment, thereby enhancing the accuracy of the judgment result.
Description
Technical Field
The application relates to a multi-radar fusion technology, in particular to a control method and device for multi-radar vehicle data homologous fusion.
Background
The intelligent traffic system needs to report vehicle track data in real time, and a radar is arranged on a highway to monitor the vehicle data for reporting. In order to enable the range of the radar to cover the whole area of the highway, certain cross overlapping exists in the detection areas of the adjacent radars. The vehicle data reported by the radar has errors, so when the vehicle is in the overlapped monitoring range of two adjacent radars, two groups of incompletely identical data from the vehicle can exist in the data reported by the two groups of radars, and if the data is considered as two vehicles, the problem obviously exists.
The fundamental reason for the above problem is the accuracy of radar information, and ideally, if the radar can report accurate data at the same time, it can be determined whether two sets of data are from the same vehicle through simple comparison. However, due to the radar device, different devices cannot report accurate data at the same time. Therefore, the data of two adjacent groups of radars need to be subjected to homologous target division so as to determine which two groups of data come from the same vehicle, and then information fusion is performed according to a proper principle.
The target homologous division is carried out on the adjacent radar data, and because the measurement data of the same target are similar,
the current processing method mainly comprises a classification algorithm and a clustering algorithm. Common algorithms include a DBSCAN algorithm based on a density clustering principle, a K-means algorithm, a maximum likelihood homologous division method and the like. The principle of the DBSCAN algorithm is to classify closely connected data into one class so as to obtain a cluster class, and the method carries out cluster division by setting a neighborhood radius and a neighborhood sample threshold value. The K-means algorithm takes the distance between the multidimensional data between the objects as an evaluation index of the correlation, the closer the distance between the objects is, the greater the similarity between the two objects is, the most similar objects are clustered, and the objects are finally classified into K classes in an iterative mode. And (3) carrying out maximum likelihood homologous division, carrying out all possible structural division on the data set, and taking a likelihood function from the division set to obtain the optimal possible division.
In actual conditions, vehicles are difficult to be uniformly distributed on a road, a DBSCAN algorithm is difficult to determine an effective neighborhood radius and a neighborhood threshold, the clustering effect is poor, and the method cannot be directly adapted to the problem. For the K-means algorithm, firstly, the size of the K value cannot be determined, and the total number of corresponding vehicles in the data cannot be determined before the solution of the homologous division. And the data of different dimensions have different dimensions, and the difference of the dimensions can cause the influence of the data of different dimensions on the weight of the correlation to be different, thereby amplifying the influence of some dimension data and causing waste of other data. For the maximum likelihood homologous division method, the total number of the vehicle targets can not be determined, feasible division can not be made, the overlapping area is small in the practical problem, most of the vehicle targets are divided into single data, and the method has no practicability.
Disclosure of Invention
The technical purpose to be achieved by the embodiment of the application is to provide a control method and device for multi-radar vehicle data homologous fusion, so as to solve the problems of large calculated amount and low accuracy existing in the conventional multi-radar vehicle data homologous fusion.
In order to solve the above technical problem, an embodiment of the present application provides a control method for multi-radar vehicle data homologous fusion, including:
preprocessing received vehicle data uploaded by at least two radars to obtain an effective data set;
acquiring second effective vehicle data related to the first effective vehicle data in the effective data set, wherein the first effective vehicle data is any effective vehicle data in the effective data set, and a first radar corresponding to the first effective vehicle data is adjacent to a second radar corresponding to the second effective vehicle data;
determining third valid vehicle data that is homologous to the first valid vehicle data from the second valid vehicle data;
and when the third valid vehicle data exist, determining that the first vehicle corresponding to the first valid vehicle data or the second vehicle corresponding to the third valid vehicle data is the target vehicle after data fusion according to a proximity principle, wherein the valid vehicle data corresponding to the target vehicle is the target vehicle data after data fusion.
Specifically, the method for controlling the multi-radar vehicle data homologous fusion, which is described above, includes the steps of preprocessing the received vehicle data uploaded by at least two radars to obtain an effective data set:
comparing the vehicle data with a preset range of the preprocessing parameters, setting the vehicle data as valid vehicle data when the values of the preprocessing parameters in the vehicle data are determined to be within the corresponding preset range, and otherwise, removing the vehicle data;
wherein the pre-processing parameters include at least one of:
the vehicle speed;
vehicle length;
vehicle width;
the vehicle height;
difference between heading angle and heading.
Preferably, the control method for multi-radar vehicle data homologous fusion as described above, before acquiring the second valid vehicle data related to the first valid vehicle data, the method further includes:
judging whether the first valid vehicle data carries a fusion mark;
when the first effective vehicle data carries a fusion mark, determining third effective vehicle data which are homologous with the first effective vehicle data from a fusion vehicle list related to the first radar and the second radar, performing data fusion on the first effective vehicle data and the third effective vehicle data to obtain target vehicle data, and determining a first vehicle corresponding to the first effective vehicle data or a second vehicle corresponding to the third effective vehicle data as a target vehicle corresponding to the target vehicle data according to a proximity principle;
and when the fusion mark is not carried in the first valid vehicle data, executing the step of acquiring second valid vehicle data related to the first valid vehicle data.
Preferably, in the control method of multi-radar vehicle data homologous fusion as described above, the step of determining third valid vehicle data homologous to the first valid vehicle data from the second valid vehicle data includes:
acquiring an absolute value of a preset judgment parameter between the first effective vehicle data and each second effective vehicle data according to the first effective vehicle data and the second effective vehicle data, and constructing an absolute value set;
when the absolute value corresponding to second valid vehicle data is determined from the absolute value set and is smaller than the threshold value of the preset judgment parameter, setting the second valid vehicle data as candidate vehicle data;
constructing an absolute value matrix of the candidate vehicle data, and determining the correlation degree of each candidate vehicle data and the first effective vehicle data according to a grey correlation analysis method;
and determining the candidate vehicle data corresponding to the maximum value of the association degree as third effective vehicle data.
Specifically, in the control method for multi-radar vehicle data homologous fusion, the preset judgment parameter includes at least one of the following:
vehicle speed difference;
vehicle length difference;
vehicle width difference;
vehicle height difference;
a road-to-vehicle distance;
a transverse vehicle distance;
and (4) course angle difference.
Specifically, the method for controlling the multi-radar vehicle data homologous fusion, according to the gray correlation analysis method, determining the degree of correlation between each candidate vehicle data and the first valid vehicle data includes:
acquiring a maximum value and a minimum value corresponding to each preset judgment parameter;
determining a correlation coefficient corresponding to each candidate vehicle data and the first effective vehicle data about each preset judgment parameter according to a preset first algorithm, a maximum value and a minimum value;
and determining the association degree of each candidate vehicle data and the first valid vehicle data according to a preset second algorithm and the association coefficient.
Preferably, the control method for multi-radar vehicle data homologous fusion as described above, wherein the step of determining, according to a proximity rule, that the first vehicle corresponding to the first valid vehicle data or the second vehicle corresponding to the third valid vehicle data is the target vehicle after data fusion, includes:
acquiring a first distance between a first vehicle and a first radar and a second distance between a second vehicle and a second radar;
when the first distance is smaller than the second distance, selecting the first vehicle as a target vehicle;
when the first distance is larger than the second distance, selecting a second vehicle as a target vehicle;
and when the first distance is equal to the second distance, determining the first vehicle or the second vehicle as the target vehicle according to a preset rule.
Preferably, after obtaining the determined target vehicle and the target vehicle data, the method for controlling the multi-radar vehicle data homologous fusion further includes:
the target vehicle data and the identity information of the first vehicle and the second vehicle are added to a fused vehicle list associated with the first radar and the second radar.
Another embodiment of the present application also provides a control apparatus including:
the first processing module is used for preprocessing the received vehicle data uploaded by at least two radars to obtain an effective data set;
the second processing module is used for acquiring second effective vehicle data related to the first effective vehicle data in the effective data set, wherein the first effective vehicle data is any effective vehicle data in the effective data set, and a first radar corresponding to the first effective vehicle data is adjacent to a second radar corresponding to the second effective vehicle data;
the third processing module is used for determining third valid vehicle data which are homologous with the first valid vehicle data from the second valid vehicle data;
and the fourth processing module is used for determining that the first vehicle corresponding to the first valid vehicle data or the second vehicle corresponding to the third valid vehicle data is the target vehicle after data fusion according to the proximity principle when the third valid vehicle data exists, and the valid vehicle data corresponding to the target vehicle is the target vehicle data after data fusion.
Specifically, as the control device described above, the first processing module includes:
the first processing unit is used for comparing the vehicle data with a preset range of the preprocessing parameters, setting the vehicle data as valid vehicle data when the vehicle data are determined to be located in the corresponding preset range relative to the preprocessing parameters, and otherwise, rejecting the vehicle data;
wherein the pre-processing parameters include at least one of:
the vehicle speed;
vehicle length;
vehicle width;
vehicle height;
difference between heading angle and heading.
Preferably, the control device as described above, further comprising:
the fifth processing module is used for judging whether the first effective vehicle data carries the fusion mark or not;
a sixth processing module, configured to, when the first valid vehicle data carries a fusion flag, determine third valid vehicle data that is homologous to the first valid vehicle data from a fusion vehicle list related to the first radar and the second radar, perform data fusion on the first valid vehicle data and the third valid vehicle data to obtain target vehicle data, and determine, according to a proximity principle, that the first vehicle corresponding to the first valid vehicle data or the second vehicle corresponding to the third valid vehicle data is a target vehicle corresponding to the target vehicle data;
and the seventh processing module is used for executing the step of acquiring second valid vehicle data related to the first valid vehicle data when the fusion mark is not carried in the first valid vehicle data.
Preferably, as the control device described above, the third processing module includes:
the second processing unit is used for acquiring an absolute value of a preset judgment parameter between the first effective vehicle data and each second effective vehicle data according to the first effective vehicle data and the second effective vehicle data, and constructing an absolute value set;
the third processing unit is used for setting the second valid vehicle data as candidate vehicle data when the absolute value corresponding to the second valid vehicle data is determined from the absolute value set and is smaller than the threshold value of the preset judgment parameter;
the fourth processing unit is used for constructing an absolute value matrix of the candidate vehicle data and determining the correlation degree of each candidate vehicle data and the first effective vehicle data according to a grey correlation analysis method;
and the fifth processing unit is used for determining the candidate vehicle data corresponding to the maximum value of the association degree as the third effective vehicle data.
Specifically, in the control device, the preset determination parameter includes at least one of the following:
vehicle speed difference;
vehicle length difference;
vehicle width difference;
vehicle height difference;
a road-to-vehicle distance;
a transverse vehicle distance;
and (4) course angle difference.
Specifically, as the control device described above, the fourth processing unit includes:
the first processing subunit is used for acquiring a maximum value and a minimum value corresponding to each preset judgment parameter;
the second processing subunit is used for determining the association coefficient corresponding to each preset judgment parameter of each candidate vehicle data and the first effective vehicle data according to a preset first algorithm, a maximum value and a minimum value;
and the third processing subunit is used for determining the association degree of each candidate vehicle data and the first effective vehicle data according to a preset second algorithm and the association coefficient.
Preferably, as the control device, the fourth processing module includes:
a sixth processing unit for acquiring a first distance between the first vehicle and the first radar, and a second distance between the second vehicle and the second radar;
the seventh processing unit is used for selecting the first vehicle as the target vehicle when the first distance is smaller than the second distance;
the eighth processing unit is used for selecting the second vehicle as the target vehicle when the first distance is greater than the second distance;
and the ninth processing unit is used for determining the first vehicle or the second vehicle as the target vehicle according to a preset rule when the first distance is equal to the second distance.
Preferably, the control device as described above, further comprising:
an eighth processing module to add the target vehicle data and the identity information of the first vehicle and the second vehicle to a fused vehicle list associated with the first radar and the second radar.
Yet another embodiment of the present application also provides a computer readable storage medium, on which a computer program is stored, which, when being executed by a processor, realizes the steps of the control method of multi-radar vehicle data homosource fusion as described above.
Compared with the prior art, the control method and the control device for the multi-radar vehicle data homologous fusion provided by the embodiment of the application have the following beneficial effects at least:
the received vehicle data are preprocessed to remove unreasonable and ineffective vehicle data, clustering is carried out through the radar range, data fusion is carried out according to the principle of proximity, the efficiency of data fusion is improved, high accuracy and small calculated amount are considered, the vehicle data of a fusion area detected by multiple radars in intelligent traffic can be efficiently processed, homologous recognition and fusion are carried out under the accuracy of the current final traffic radar equipment, and vehicle data statistics and display are better achieved. And homologous identification can be performed by a grey correlation analysis method, and correlation degree judgment is performed by fully utilizing reported data, so that the accuracy of a judgment result is enhanced.
Drawings
FIG. 1 is a schematic flow chart of a control method for multi-radar vehicle data homography fusion according to the present application;
FIG. 2 is a second flowchart of the control method for multi-radar vehicle data homonymy fusion according to the present application;
FIG. 3 is a third schematic flow chart of a control method for multi-radar vehicle data homography fusion according to the present application;
FIG. 4 is a fourth flowchart illustrating a control method for multi-radar vehicle data homography fusion according to the present application;
fig. 5 is a schematic structural diagram of the control device of the present application.
Detailed Description
To make the technical problems, technical solutions and advantages to be solved by the present application clearer, the following detailed description is made with reference to the accompanying drawings and specific embodiments. In the following description, specific details such as specific configurations and components are provided only to help the embodiments of the present application be fully understood. Accordingly, it will be apparent to those skilled in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present application. In addition, descriptions of well-known functions and constructions are omitted for clarity and conciseness.
It should be appreciated that reference throughout this specification to "one embodiment" or "an embodiment" means that a particular feature, structure or characteristic described in connection with the embodiment is included in at least one embodiment of the present application. Thus, the appearances of the phrases "in one embodiment" or "in an embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
In various embodiments of the present application, it should be understood that the sequence numbers of the following processes do not mean the execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.
It should be understood that the term "and/or" herein is merely one type of association relationship that describes an associated object, meaning that three relationships may exist, e.g., a and/or B may mean: a exists alone, A and B exist simultaneously, and B exists alone. In addition, the character "/" herein generally indicates that the former and latter related objects are in an "or" relationship.
In the embodiments provided herein, it should be understood that "B corresponding to a" means that B is associated with a from which B can be determined. It should also be understood that determining B from a does not mean determining B from a alone, but may be determined from a and/or other information.
Referring to fig. 1, a preferred embodiment of the present application provides a control method for multi-radar vehicle data homosource fusion, including:
step S101, preprocessing received vehicle data uploaded by at least two radars to obtain an effective data set;
step S102, second valid vehicle data related to first valid vehicle data in the valid data set is obtained, wherein the first valid vehicle data is any valid vehicle data in the valid data set, and a first radar corresponding to the first valid vehicle data is adjacent to a second radar corresponding to the second valid vehicle data;
step S103, determining third valid vehicle data which is homologous with the first valid vehicle data from the second valid vehicle data;
and step S104, when third valid vehicle data exists, determining that the first vehicle corresponding to the first valid vehicle data or the second vehicle corresponding to the third valid vehicle data is a target vehicle after data fusion according to a proximity principle, wherein the valid vehicle data corresponding to the target vehicle is the target vehicle data after data fusion.
In a preferred embodiment of the application, a control method for multi-radar vehicle data homologous fusion is provided, wherein a control device capable of implementing the control method can receive vehicle data uploaded by at least two radars, two radars are adjacently arranged in the at least two radars, and an overlapping area can exist. After receiving the uploaded vehicle data, the control device can carry out effective preprocessing on the uploaded vehicle data, and removes unreasonable and ineffective vehicle data through the preprocessing, only effective vehicle data are reserved, and an effective data set is obtained, so that the validity of the adopted data is ensured in the subsequent homologous identification step, and the accuracy of the obtained homologous identification result is improved; before the homologous identification step is carried out, based on the limitation of factors such as radar monitoring range and vehicle speed, when a vehicle is located near the overlapping area of at least two adjacent radars, the vehicle can be monitored by the multiple radars, therefore, when any effective vehicle data in an effective data set, namely homologous data of first effective vehicle data, is obtained, effective vehicle data, namely second effective vehicle data, monitored by a second radar adjacent to a first radar corresponding to the first effective vehicle data can be obtained firstly, and then the homologous identification step is executed, third effective vehicle data homologous with the first effective vehicle data is determined from the multiple second effective vehicle data, the data range during data homologous identification is favorably reduced, and the efficiency is favorably improved. When it is determined that third valid vehicle data which is the same as the first valid vehicle data exists, data fusion may be performed according to a preset fusion rule, in this embodiment, it is preferable to determine a target vehicle after data fusion from a first vehicle corresponding to the first valid vehicle data and a second vehicle corresponding to the second valid vehicle data by using a proximity principle, and use valid vehicle data corresponding to the target vehicle as target vehicle data after data fusion, where it is to be noted that the first vehicle and the second vehicle are used to refer to different identifiers of the same vehicle in different radars.
In conclusion, in the embodiment of the application, unreasonable invalid vehicle data are removed through preprocessing the received vehicle data, clustering is performed through the radar range, data fusion is performed according to the principle of proximity, the efficiency of data fusion is improved, high accuracy and small calculated amount are considered, the vehicle data of a fusion area detected by multiple radars in intelligent traffic can be efficiently processed, homologous recognition and fusion are performed under the accuracy of the current final traffic radar equipment, and statistics and display of the vehicle data are better achieved.
Specifically, the method for controlling the multi-radar vehicle data homologous fusion, which is described above, includes the steps of preprocessing the received vehicle data uploaded by at least two radars to obtain an effective data set:
comparing the vehicle data with a preset range of the preprocessing parameters, setting the vehicle data as valid vehicle data when the values of the preprocessing parameters in the vehicle data are determined to be within the corresponding preset range, and otherwise, removing the vehicle data;
wherein the pre-processing parameters include at least one of:
the vehicle speed;
the length of the vehicle;
vehicle width;
vehicle height;
difference between heading angle and heading.
In a specific embodiment of the present application, in the process of preprocessing the vehicle data, a value related to a preset preprocessing parameter in the vehicle data is compared with a preset range of the corresponding preprocessing parameter, and if the value related to the preset preprocessing parameter in the vehicle data is located within the corresponding preset range, the value can be determined, the vehicle data is valid vehicle data, and if one value is located outside the corresponding preset range, it can be determined that the vehicle data is unreasonable invalid data, and at this time, the vehicle data can be removed, so as to ensure accuracy and reasonableness of the subsequently used vehicle data.
Specifically, the pre-processing parameters include, but are not limited to, at least one of the following: the vehicle speed; vehicle length; vehicle width; vehicle height; difference between heading angle and heading.
The vehicle speed, the vehicle length, the vehicle width and the vehicle height can be used for judging whether a vehicle corresponding to the vehicle data is a real vehicle or not so as to avoid the occurrence of non-vehicle objects such as: birds, floats, etc. are mistaken for vehicles; the difference value of the course angle and the road direction is used for judging whether the vehicle runs reasonably, so that the subsequent judgment of the vehicle which has a reverse trend with the road direction and is obviously unreasonable is carried out, the calculated amount is increased, and the like.
Optionally, the preprocessing parameters may be associated with each other, and may also be associated with the road segment, for example: and determining the type of the vehicle according to the vehicle specification parameters such as the length, the width and the height of the vehicle, and further determining the vehicle speed range according to the type of the vehicle. And unreasonable vehicle data can be further removed. Taking a small vehicle as an example, the corresponding range of the vehicle length is 3-8 meters, the corresponding range of the vehicle width is 1-3 meters, and the difference between the course angle and the road direction is within 90 degrees.
Referring to fig. 2, preferably, the control method for multi-radar vehicle data homosource fusion as described above, before acquiring the second valid vehicle data related to the first valid vehicle data, the method further includes:
step S201, judging whether the first valid vehicle data carries a fusion mark;
step S202, when the first effective vehicle data carries a fusion mark, determining third effective vehicle data which are homologous with the first effective vehicle data from a fusion vehicle list related to the first radar and the second radar, performing data fusion on the first effective vehicle data and the third effective vehicle data to obtain target vehicle data, and determining a first vehicle corresponding to the first effective vehicle data or a second vehicle corresponding to the third effective vehicle data as a target vehicle corresponding to the target vehicle data according to a proximity principle;
in step S203, when the first valid vehicle data does not carry the fusion flag, a step of acquiring second valid vehicle data related to the first valid vehicle data is performed.
In a preferred embodiment of the present application, before obtaining second valid vehicle data related to the first valid vehicle data, it is further determined whether the first valid vehicle data has been fused, specifically, the fused vehicle data carries a fusion flag, and when the first valid vehicle data carries the fusion flag, it may be indicated that the first valid vehicle data has been fused, at this time, third valid vehicle data that is homologous to the first valid vehicle data may be directly determined from a fusion vehicle list related to the first radar and the second radar, and then data fusion may be directly performed; by setting a fusion mark and a fusion vehicle list, the steps of homologous identification can be further simplified, the calculated amount is reduced, and the efficiency is further improved; and if the first effective vehicle data does not carry the fusion mark, executing the steps of obtaining the second effective vehicle data and further determining third effective vehicle data which are homologous with the first effective vehicle data from the second effective vehicle data.
Referring to fig. 3, preferably, the step of determining the third valid vehicle data that is homologous with the first valid vehicle data from the second valid vehicle data according to the control method for multi-radar vehicle data homologous fusion described above includes:
step S301, according to the first effective vehicle data and the second effective vehicle data, obtaining an absolute value of a preset judgment parameter between the first effective vehicle data and each second effective vehicle data, and constructing an absolute value set;
step S302, when determining that the absolute values corresponding to second valid vehicle data in the absolute value set are all smaller than the threshold value of a preset judgment parameter, setting the second valid vehicle data as candidate vehicle data;
step S303, constructing an absolute value matrix of candidate vehicle data, and determining the correlation degree of each candidate vehicle data and the first effective vehicle data according to a grey correlation analysis method;
in step S304, the candidate vehicle data corresponding to the maximum value of the degree of association is determined as the third valid vehicle data.
In a specific embodiment of the present application, when determining third valid vehicle data that is identical to the first valid vehicle data from the second valid vehicle data, the third valid vehicle data is obtained by first obtaining, according to a preset determination parameter that is preset, an absolute value of the preset determination parameter between the first valid vehicle data and each of the second valid vehicle data, and constructing an absolute value set; and further screening the second effective vehicle data according to the threshold value of the preset judgment parameter to obtain candidate vehicle data, wherein the calculation amount can be further reduced through the further screening step. By constructing an absolute value matrix of the candidate vehicle data, the association degree of each candidate vehicle data with the first valid vehicle data can be determined according to a grey association analysis method; according to the relevance degree of each candidate vehicle data and the first effective vehicle data, the candidate vehicle data corresponding to the maximum value of the relevance degree can be determined to be the third effective vehicle data. The grey correlation analysis method is adopted for correlation analysis, so that the reported data can be more fully utilized for correlation degree judgment, and the accuracy of the judgment result is enhanced. In one embodiment, taking the example of the data of 3 candidate vehicles with six preset judgment parameters, the absolute value matrix constructed is as follows:
wherein x, y and z respectively represent candidate vehicle data, and 1-6 respectively represent a preset judgment parameter.
It should be noted that, in the present embodiment, an absolute value is adopted, which is beneficial to avoiding the situation that the analysis result changes due to the fact that the same pair of data changes in the front-back order, for example, the first valid vehicle data and the second valid vehicle data are exchanged.
Specifically, in the control method for multi-radar vehicle data homologous fusion, the preset judgment parameter includes at least one of the following:
vehicle speed difference;
vehicle length difference;
vehicle width difference;
vehicle height difference;
a road-to-vehicle distance;
a transverse vehicle distance;
and (4) course angle difference.
In an embodiment, the preset determination parameter is exemplified, and the preset determination parameter includes, but is not limited to, at least one of the following: vehicle speed difference; vehicle length difference; vehicle width difference; vehicle height difference; a road-to-vehicle distance; a transverse vehicle distance; and (4) course angle difference. It should be noted that the route-to-route distance is a distance between longitude and latitude coordinates corresponding to the two vehicle data on the route; the transverse vehicle distance is the distance between the longitude and latitude coordinates corresponding to the two vehicle data in the direction of the vertical road, and the condition of the corresponding vehicle lane is reflected. The heading angle difference is the angle difference between the heading angles corresponding to the two pieces of vehicle data.
In one embodiment, the threshold value of the absolute value of the road-to-vehicle distance is preferably 2000 cm; the threshold value of the absolute value of the heading angle difference is preferably 90 degrees; the threshold value of the absolute value of the vehicle length difference is preferably 800 cm; the threshold value of the absolute value of the vehicle width difference is preferably 260 cm; the threshold value of the absolute value of the vehicle height difference is preferably 150 cm; the threshold value for the absolute value of the lateral vehicle distance is preferably 100 cm.
Referring to fig. 4, in particular, the control method for multi-radar vehicle data homologous fusion as described above, the step of determining the association degree of each candidate vehicle data with the first valid vehicle data according to the gray association analysis method includes:
step S401, obtaining a maximum value and a minimum value corresponding to each preset judgment parameter;
step S402, determining a correlation coefficient corresponding to each candidate vehicle data and the first effective vehicle data about each preset judgment parameter according to a preset first algorithm, a maximum value and a minimum value;
step S403, determining a degree of association between each candidate vehicle data and the first valid vehicle data according to a preset second algorithm and the association coefficient.
In a specific embodiment, when determining the degree of association between each candidate vehicle data and the first valid vehicle data according to a gray association analysis method, a maximum value and a minimum value corresponding to each preset judgment parameter are obtained based on an absolute value matrix, where the maximum values are:
Δi(max)＝max{xi，yi，zi}i＝1，2，…，6
the minimum values are:
Δi(min)＝min{xi，yi，zi}i＝1，2，…，6
and then, the obtained maximum value and the obtained minimum value are substituted into a first algorithm, so that a correlation coefficient corresponding to each candidate vehicle data and each first effective vehicle data with respect to each preset judgment parameter can be obtained, wherein the first algorithm is preferably:
wherein, i is 1, 2, …, 6; j is x, y, z; ρ is a resolution coefficient, preferably ρ is 0.5.
And then according to a preset second algorithm and the correlation coefficient, the correlation degree of each candidate vehicle data and the first effective vehicle data can be determined. Wherein the second algorithm is preferably:
preferably, the control method for multi-radar vehicle data homologous fusion as described above, wherein the step of determining, according to a proximity rule, that the first vehicle corresponding to the first valid vehicle data or the second vehicle corresponding to the third valid vehicle data is the target vehicle after data fusion, includes:
acquiring a first distance between a first vehicle and a first radar and a second distance between a second vehicle and a second radar;
when the first distance is smaller than the second distance, selecting the first vehicle as a target vehicle;
when the first distance is larger than the second distance, selecting a second vehicle as a target vehicle;
and when the first distance is equal to the second distance, determining the first vehicle or the second vehicle as the target vehicle according to a preset rule.
In one embodiment, when determining the target vehicle according to the proximity principle, a first distance between the first vehicle and the first radar and a second distance between the second vehicle and the second radar are obtained first, and then the first distance and the second distance are compared, so as to select the vehicle with a small distance as the target vehicle. The preset rule includes, but is not limited to, determining that the first vehicle or the second vehicle is the target vehicle, or determining that the vehicle in front of the vehicle is the target vehicle according to the vehicle driving tendency.
Preferably, after obtaining the determined target vehicle and the target vehicle data, the method for controlling the multi-radar vehicle data homologous fusion further includes:
the target vehicle data and the identity information of the first vehicle and the second vehicle are added to a fused vehicle list associated with the first radar and the second radar.
In a preferred embodiment of the present application, since the identities of the same vehicle in the same radar are the same, the target vehicle data and the identity information of the first vehicle and the second vehicle are also added to the fused vehicle list associated with the first radar and the second radar after the target vehicle and the target vehicle data are determined. So that the homologous data can be directly determined according to the identity information in the fusion vehicle list in the next period. Thereby contributing to a reduction in the amount of calculation.
Referring to fig. 5, another embodiment of the present application also provides a control apparatus including:
the first processing module 501 is configured to pre-process received vehicle data uploaded by at least two radars to obtain an effective data set;
a second processing module 502, configured to obtain second valid vehicle data related to first valid vehicle data in the valid data set, where the first valid vehicle data is any valid vehicle data in the valid data set, and a first radar corresponding to the first valid vehicle data is adjacent to a second radar corresponding to the second valid vehicle data;
a third processing module 503, configured to determine third valid vehicle data that is homologous to the first valid vehicle data from the second valid vehicle data;
the fourth processing module 504 is configured to determine, according to a rule of proximity, that the first vehicle corresponding to the first valid vehicle data or the second vehicle corresponding to the third valid vehicle data is the target vehicle after data fusion, and determine that the valid vehicle data corresponding to the target vehicle is the target vehicle data after data fusion, when the third valid vehicle data exists.
Specifically, as the control device described above, the first processing module includes:
the first processing unit is used for comparing the vehicle data with a preset range of the preprocessing parameters, setting the vehicle data as valid vehicle data when the vehicle data are determined to be located in the corresponding preset range relative to the preprocessing parameters, and otherwise, rejecting the vehicle data;
wherein the pre-processing parameters include at least one of:
the vehicle speed;
vehicle length;
vehicle width;
vehicle height;
difference between heading angle and heading.
Preferably, the control device as described above, further comprising:
the fifth processing module is used for judging whether the first effective vehicle data carries the fusion mark or not;
a sixth processing module, configured to, when the first valid vehicle data carries a fusion flag, determine third valid vehicle data that is homologous to the first valid vehicle data from a fusion vehicle list related to the first radar and the second radar, perform data fusion on the first valid vehicle data and the third valid vehicle data to obtain target vehicle data, and determine, according to a proximity principle, that the first vehicle corresponding to the first valid vehicle data or the second vehicle corresponding to the third valid vehicle data is a target vehicle corresponding to the target vehicle data;
and the seventh processing module is used for executing the step of acquiring second valid vehicle data related to the first valid vehicle data when the fusion mark is not carried in the first valid vehicle data.
Preferably, as the control device described above, the third processing module includes:
the second processing unit is used for acquiring an absolute value of a preset judgment parameter between the first effective vehicle data and each second effective vehicle data according to the first effective vehicle data and the second effective vehicle data, and constructing an absolute value set;
the third processing unit is used for setting the second valid vehicle data as candidate vehicle data when the absolute value corresponding to the second valid vehicle data is determined from the absolute value set and is smaller than the threshold value of the preset judgment parameter;
the fourth processing unit is used for constructing an absolute value matrix of the candidate vehicle data and determining the correlation degree of each candidate vehicle data and the first effective vehicle data according to a grey correlation analysis method;
and the fifth processing unit is used for determining the candidate vehicle data corresponding to the maximum value of the association degree as the third effective vehicle data.
Specifically, in the control device, the preset determination parameter includes at least one of the following:
vehicle speed difference;
vehicle length difference;
vehicle width difference;
vehicle height difference;
a road-to-vehicle distance;
a transverse vehicle distance;
and (4) course angle difference.
Specifically, as the control device described above, the fourth processing unit includes:
the first processing subunit is used for acquiring a maximum value and a minimum value corresponding to each preset judgment parameter;
the second processing subunit is used for determining the association coefficient corresponding to each preset judgment parameter of each candidate vehicle data and the first effective vehicle data according to a preset first algorithm, a maximum value and a minimum value;
and the third processing subunit is used for determining the association degree of each candidate vehicle data and the first effective vehicle data according to a preset second algorithm and the association coefficient.
Preferably, as the control device, the fourth processing module includes:
a sixth processing unit for acquiring a first distance between the first vehicle and the first radar, and a second distance between the second vehicle and the second radar;
the seventh processing unit is used for selecting the first vehicle as the target vehicle when the first distance is smaller than the second distance;
the eighth processing unit is used for selecting the second vehicle as the target vehicle when the first distance is greater than the second distance;
and the ninth processing unit is used for determining the first vehicle or the second vehicle as the target vehicle according to a preset rule when the first distance is equal to the second distance.
Preferably, the control device as described above, further comprising:
an eighth processing module to add the target vehicle data and the identity information of the first vehicle and the second vehicle to a fused vehicle list associated with the first radar and the second radar.
The embodiment of the control device of the invention is the control device corresponding to the embodiment of the method, and all implementation means in the embodiment of the method are applicable to the embodiment of the control device, so that the same technical effects can be achieved.
Yet another embodiment of the present application also provides a computer readable storage medium, on which a computer program is stored, which, when being executed by a processor, realizes the steps of the control method of multi-radar vehicle data homosource fusion as described above.
Further, the present application may repeat reference numerals and/or letters in the various examples. This repetition is for the purpose of simplicity and clarity and does not in itself dictate a relationship between the various embodiments and/or configurations discussed.
It is further noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion.
The foregoing is a preferred embodiment of the present application, and it should be noted that, for those skilled in the art, several modifications and refinements can be made without departing from the principle described in the present application, and these modifications and refinements should be regarded as the protection scope of the present application.
Claims (10)
1. A control method for multi-radar vehicle data homologous fusion is characterized by comprising the following steps:
preprocessing received vehicle data uploaded by at least two radars to obtain an effective data set;
acquiring second valid vehicle data related to first valid vehicle data in the valid data set, wherein the first valid vehicle data is any valid vehicle data in the valid data set, and a first radar corresponding to the first valid vehicle data is adjacent to a second radar corresponding to the second valid vehicle data;
determining third valid vehicle data that is homologous to the first valid vehicle data from the second valid vehicle data;
when the third valid vehicle data exists, determining that the first vehicle corresponding to the first valid vehicle data or the second vehicle corresponding to the third valid vehicle data is a target vehicle after data fusion according to a proximity principle, wherein the valid vehicle data corresponding to the target vehicle is target vehicle data after data fusion.
2. The method for controlling the homologous fusion of the multi-radar vehicle data according to claim 1, wherein the step of preprocessing the received vehicle data uploaded by at least two radars to obtain an effective data set comprises:
comparing the vehicle data with a preset range of a preprocessing parameter, and when the value of the preprocessing parameter in the vehicle data is determined to be within the corresponding preset range, setting the vehicle data as the valid vehicle data, otherwise, rejecting the vehicle data;
wherein the pre-processing parameters include at least one of:
the vehicle speed;
vehicle length;
vehicle width;
vehicle height;
difference between heading angle and heading.
3. The method of claim 1, wherein prior to obtaining second valid vehicle data related to the first valid vehicle data, the method further comprises:
judging whether the first valid vehicle data carries a fusion mark or not;
when the first effective vehicle data carries a fusion mark, determining third effective vehicle data which are homologous with the first effective vehicle data from a fusion vehicle list related to the first radar and the second radar, performing data fusion on the first effective vehicle data and the third effective vehicle data to obtain target vehicle data, and determining a first vehicle corresponding to the first effective vehicle data or a second vehicle corresponding to the third effective vehicle data as a target vehicle corresponding to the target vehicle data according to a proximity principle;
and when the first valid vehicle data does not carry the fusion mark, executing the step of acquiring second valid vehicle data related to the first valid vehicle data.
4. The method of claim 1, wherein the step of determining third valid vehicle data from the second valid vehicle data that is homologous to the first valid vehicle data comprises:
acquiring an absolute value of a preset judgment parameter between the first valid vehicle data and each second valid vehicle data according to the first valid vehicle data and the second valid vehicle data, and constructing an absolute value set;
when the absolute value corresponding to one second valid vehicle data is determined from the absolute value set and is smaller than the threshold value of the preset judgment parameter, setting the second valid vehicle data as candidate vehicle data;
constructing an absolute value matrix of the candidate vehicle data, and determining the correlation degree of each candidate vehicle data and the first valid vehicle data according to a grey correlation analysis method;
and determining the candidate vehicle data corresponding to the maximum value of the association degree as the third valid vehicle data.
5. The control method for multi-radar vehicle data homology fusion according to claim 4, wherein the preset judgment parameter comprises at least one of:
vehicle speed difference;
vehicle length difference;
vehicle width difference;
vehicle height difference;
a road-to-vehicle distance;
a transverse vehicle distance;
and (4) course angle difference.
6. The method for controlling multi-radar vehicle data homonymous fusion according to claim 4 or 5, wherein the step of determining the degree of association of each candidate vehicle data with the first valid vehicle data according to a gray association analysis method includes:
acquiring a maximum value and a minimum value corresponding to each preset judgment parameter;
determining a correlation coefficient corresponding to each candidate vehicle data and the first effective vehicle data with respect to each preset judgment parameter according to a preset first algorithm and the maximum value and the minimum value;
and determining the association degree of each candidate vehicle data and the first valid vehicle data according to a preset second algorithm and the association coefficient.
7. The method for controlling multi-radar vehicle data homonymous fusion according to claim 1, wherein the step of determining, on a near basis, that the first vehicle corresponding to the first valid vehicle data or the second vehicle corresponding to the third valid vehicle data is a data-fused target vehicle includes:
acquiring a first distance between the first vehicle and the first radar and a second distance between the second vehicle and the second radar;
when the first distance is smaller than the second distance, selecting the first vehicle as the target vehicle;
when the first distance is larger than the second distance, selecting the second vehicle as the target vehicle;
when the first distance is equal to the second distance, determining that the first vehicle or the second vehicle is the target vehicle according to a preset rule.
8. The method for controlling multi-radar vehicle data homonymous fusion according to claim 1 or 3, wherein after determining the target vehicle and the target vehicle data, the method further comprises:
adding the target vehicle data and the identity information of the first vehicle and the second vehicle to a list of fused vehicles associated with the first radar and the second radar.
9. A control device, comprising:
the first processing module is used for preprocessing the received vehicle data uploaded by at least two radars to obtain an effective data set;
the second processing module is used for acquiring second valid vehicle data related to first valid vehicle data in the valid data set, wherein the first valid vehicle data is any valid vehicle data in the valid data set, and a first radar corresponding to the first valid vehicle data is adjacent to a second radar corresponding to the second valid vehicle data;
a third processing module for determining third valid vehicle data from the second valid vehicle data that is homologous to the first valid vehicle data;
and the fourth processing module is used for determining that the first vehicle corresponding to the first valid vehicle data or the second vehicle corresponding to the third valid vehicle data is a target vehicle after data fusion according to a proximity principle when the third valid vehicle data exists, and the valid vehicle data corresponding to the target vehicle is target vehicle data after data fusion.
10. A computer-readable storage medium, characterized in that a computer program is stored thereon, which computer program, when being executed by a processor, carries out the steps of a method of controlling a multi-radar vehicle data homonymy fusion according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111416029.4A CN114879194A (en) | 2021-11-25 | 2021-11-25 | Control method and device for multi-radar vehicle data homologous fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111416029.4A CN114879194A (en) | 2021-11-25 | 2021-11-25 | Control method and device for multi-radar vehicle data homologous fusion |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114879194A true CN114879194A (en) | 2022-08-09 |
Family
ID=82667566
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111416029.4A Pending CN114879194A (en) | 2021-11-25 | 2021-11-25 | Control method and device for multi-radar vehicle data homologous fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114879194A (en) |
-
2021
- 2021-11-25 CN CN202111416029.4A patent/CN114879194A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Tian et al. | An automatic car accident detection method based on cooperative vehicle infrastructure systems | |
| CN106980113B (en) | Object detection device and object detection method | |
| CN113155173B (en) | Perception performance evaluation method and device, electronic device and storage medium | |
| CN111796286B (en) | Brake grade evaluation method and device, vehicle and storage medium | |
| CN109344886B (en) | Occlusion number plate distinguishing method based on convolutional neural network | |
| CN115294767B (en) | Real-time detection and traffic safety early warning method and device for expressway lane line | |
| Quiros et al. | Machine vision of traffic state estimation using fuzzy logic | |
| CN114694060B (en) | Road casting detection method, electronic equipment and storage medium | |
| Tak et al. | Development of AI‐Based Vehicle Detection and Tracking System for C‐ITS Application | |
| CN108344997B (en) | Road guardrail rapid detection method based on trace point characteristics | |
| CN113378751A (en) | Traffic target identification method based on DBSCAN algorithm | |
| CN114460598A (en) | Target identification method, device, equipment and storage medium | |
| CN113658427A (en) | Road condition monitoring method, system and equipment based on vision and radar | |
| Sabour et al. | Deepflow: Abnormal traffic flow detection using siamese networks | |
| CN113297939B (en) | Obstacle detection method, obstacle detection system, terminal device and storage medium | |
| CN113740874A (en) | Road edge detection method, electronic equipment and vehicle | |
| CN113432615B (en) | Detection method and system based on multi-sensor fusion drivable area and vehicle | |
| CN117130010A (en) | Obstacle sensing method and system for unmanned vehicle and unmanned vehicle | |
| CN112163521A (en) | Vehicle driving behavior identification method, device and equipment | |
| CN114879194A (en) | Control method and device for multi-radar vehicle data homologous fusion | |
| CN115520216A (en) | Driving state judging method and device, computer equipment and storage medium | |
| CN115526837A (en) | Abnormal driving detection method and device, electronic equipment and medium | |
| Lidman et al. | Clustering, shape extraction and velocity estimation applied to radar detections | |
| CN114895274A (en) | Guardrail identification method | |
| CN114882470A (en) | Vehicle-mounted anti-collision early warning method and device, computer equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 400040 No. 35, Jinghe Road, Huxi street, high tech Zone, Shapingba District, Chongqing Applicant after: CITIC Technology Zhilian Technology Co.,Ltd. Address before: 400040 No. 35, Jinghe Road, Huxi street, high tech Zone, Shapingba District, Chongqing Applicant before: Datang Gaohong Zhilian Technology (Chongqing) Co.,Ltd. |