CN109543704B - Multi-target clustering method and device for automobile and vehicle-mounted radar - Google Patents
Multi-target clustering method and device for automobile and vehicle-mounted radar Download PDFInfo
- Publication number
- CN109543704B CN109543704B CN201710887429.0A CN201710887429A CN109543704B CN 109543704 B CN109543704 B CN 109543704B CN 201710887429 A CN201710887429 A CN 201710887429A CN 109543704 B CN109543704 B CN 109543704B
- Authority
- CN
- China
- Prior art keywords
- array
- polling
- elements
- state
- target
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention discloses a multi-target clustering method and a multi-target clustering device for automobiles and vehicle-mounted radars, wherein the method comprises the following steps: s1, acquiring an original target array to be polled, and marking each element in the array as an unaccessed state; s2, polling each element in the original target array, and acquiring a corresponding new array according to the polled element in the unaccessed state; s3, polling each element in the new array, and acquiring the corresponding cluster array according to the polled element in the unaccessed state; s4, polling each element in the cluster array, and adding the element in the state of not being accessed into the corresponding new array; s5, returning to S3 after completing cluster array polling, and returning to S2 after completing new array polling; and S6, after the original target array polling is completed, forming a clustering target array by the elements with the maximum amplitude in each new array. The method can improve the target clustering accuracy of the vehicle-mounted radar and has a good clustering effect on large targets.
Description
Technical Field
The invention relates to the technical field of vehicle-mounted radars, in particular to a multi-target clustering method of a vehicle-mounted radar, a multi-target clustering device of the vehicle-mounted radar and an automobile.
Background
In the related technology, when a vehicle-mounted radar target is clustered, a radar detection system is used for periodic detection to obtain a multi-dimensional parameter of a point trace in a detection area, and then a DBSCAN (Density-Based Spatial Clustering of Applications with Noise) Clustering algorithm is used for Clustering the point trace in each detection period to obtain a plurality of clusters, wherein each cluster represents one vehicle-mounted radar target. Input parameters of the DBSCAN clustering algorithm comprise a point trace multidimensional parameter and a set of preset neighborhood parameters (e, MinPts), wherein the e is the radius of a circle with the object point trace as the center, and the MinPts is a threshold value of the number of the adjacent point traces in the circle with the object point trace as the center.
And further, filtering a plurality of clusters obtained after clustering, calculating the average distance of the clusters after filtering, and giving out early warning if the change trend of the average distance of the clusters in a plurality of continuous detection periods meets the early warning condition. Wherein the average distance of a cluster is the average of the average trace-to-point distances of all clusters in a detection period, and the average trace-to-point distance of a cluster is the average of the distances of all trace-to-point distances in the cluster.
However, the above technology is an automatic detection technology for long-term monitoring and providing early warning in high-risk areas under various weather conditions, and is only for the purpose of early warning, and only for comparing the average distance variation trend of clusters in a detection period with the early warning conditions, and does not integrate the detected target points into one target.
Disclosure of Invention
The present invention is directed to solving, at least to some extent, one of the technical problems in the art described above. Therefore, the invention aims to provide a multi-target clustering method for a vehicle-mounted radar, which can improve the target clustering accuracy of the vehicle-mounted radar and has a good clustering effect on large targets.
A second object of the invention is to propose a computer-readable storage medium.
The third purpose of the invention is to provide a multi-target clustering device for the vehicle-mounted radar.
A fourth object of the invention is to provide a motor vehicle.
In order to achieve the above object, an embodiment of a first aspect of the present invention provides a multi-target clustering method for a vehicle radar, including the following steps: s1, acquiring an original target array to be polled, and marking each element in the original target array as an unaccessed state, wherein each element in the original target array corresponds to a target of the vehicle-mounted radar; s2, polling each element in the original target array, and classifying the elements in the original target array according to the parameter information of the element in the unaccessed state being polled and the parameter information of the remaining elements in the original target array except the element being polled to obtain a new array corresponding to the element in the unaccessed state, wherein the parameter information of the element at least comprises the amplitude of the corresponding target; s3, polling each element in the new array, and classifying the elements in the original target array according to the parameter information of the element in the non-access state being polled and the parameter information of the remaining elements in the original target array except the element being polled to obtain a cluster number array corresponding to the element in the non-access state; s4, polling each element in the cluster array, and adding the element in the cluster array in the state of not being accessed to a new array corresponding to the cluster array; s5, when the polling to the cluster array is completed, returning to execute the step S3, and when the polling to the new array is completed, returning to execute the step S2; s6, when the original target array is polled, acquiring the element with the maximum amplitude value in each new array, and adding the element with the maximum amplitude value in each new array to the clustering target array to complete target clustering.
According to the multi-target clustering method of the vehicle-mounted radar, an original target array to be polled is obtained, each element in the original target array is marked to be in an unaccessed state, each element in the original target array is polled, a corresponding new array is obtained according to the polled element in the unaccessed state, each element in the new array is polled, a corresponding cluster array is obtained according to the polled element in the unaccessed state, each element in the cluster array is polled, the element in the unaccessed state in the cluster array is added into the new array corresponding to the cluster array, the elements in the new array are polled continuously after the polling of the cluster array is finished, the elements in the original target array are polled continuously after the polling of the new array is finished, and the original target array is polled continuously, and obtaining the element with the largest amplitude in each new array, and forming the element with the largest amplitude in each new array into a clustering target array, so that the clustering accuracy of the vehicle-mounted radar target can be improved, and a good clustering effect is achieved on a large target.
Further, the present invention proposes a computer-readable storage medium on which a computer program is stored, which when executed by a processor implements the above-mentioned multi-target clustering method for vehicle-mounted radars.
The computer-readable storage medium of the embodiment of the invention can improve the clustering accuracy of the vehicle-mounted radar target and has good clustering effect on a large target by executing the program corresponding to the multi-target clustering method of the vehicle-mounted radar stored on the computer-readable storage medium.
In order to achieve the above object, an embodiment of a third aspect of the present invention provides a multi-target clustering device for a vehicle radar, including: the acquisition module is used for acquiring an original target array to be polled; the marking module is used for marking each element in the original target array into an unvisited state, wherein each element in the original target array corresponds to one target of the vehicle-mounted radar; the first polling module is used for polling each element in the original target array, and classifying the elements in the original target array according to the parameter information of the element which is in the non-access state and the parameter information of the rest elements except the element which is in the polling state in the original target array to obtain a new array corresponding to the element which is in the non-access state, wherein the parameter information of the element at least comprises the amplitude value of the corresponding target; the second polling module is used for polling each element in the new array and classifying the elements in the original target array according to the parameter information of the element which is in the non-access state and is being polled and the parameter information of the rest elements in the original target array except the element which is being polled so as to obtain a cluster array corresponding to the element which is in the non-access state; the third polling module is used for polling each element in the cluster array and adding the element in the non-access state in the cluster array to a new array corresponding to the cluster array; and the control module is used for controlling the second polling module to continuously poll the elements in the new array after the third polling module finishes polling the cluster array, controlling the first polling module to continuously poll the elements in the original target array after the second polling module finishes polling the new array, acquiring the element with the largest amplitude in each new array after the first polling module finishes polling the original target array, and adding the element with the largest amplitude in each new array into the clustering target array to finish target clustering.
According to the multi-target clustering device of the vehicle-mounted radar, the original target array to be polled is obtained through the obtaining module, each element in the original target array is marked to be in an unaccessed state through the marking module, each element in the original target array is polled through the first polling module, a corresponding new array is obtained according to the polled element in the unaccessed state, each element in the new array is polled through the second polling module, a corresponding cluster array is obtained according to the polled element in the unaccessed state, each element in the cluster array is polled through the third polling module, the element in the unaccessed state in the cluster array is added to the new array corresponding to the cluster array, and the second polling module is controlled to continuously poll the elements in the new array after the third polling module finishes polling on the cluster array through the control module, and after the second polling module finishes polling the new array, controlling the first polling module to continuously poll the elements in the original target array, acquiring the elements with the maximum amplitude in each new array after the first polling module finishes polling the original target array, and forming the elements with the maximum amplitude in each new array into a clustering target array.
Further, the invention provides an automobile, which comprises the multi-target clustering device of the vehicle-mounted radar of the embodiment.
According to the automobile provided by the embodiment of the invention, the multi-target clustering device of the vehicle-mounted radar can improve the clustering accuracy of the vehicle-mounted radar target, and has a good clustering effect on a large target.
Drawings
FIG. 1 is a flow chart of a multi-target clustering method of vehicle-mounted radar according to an embodiment of the invention;
FIG. 2 is a flow chart of a multi-target clustering method for a vehicle radar according to an embodiment of the present invention;
fig. 3 is a block diagram of a multi-target clustering apparatus of a vehicle radar according to an embodiment of the present invention;
fig. 4 is a block diagram of an automobile according to an embodiment of the present invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are illustrative and intended to be illustrative of the invention and are not to be construed as limiting the invention.
The following describes a multi-target clustering method and device for an automobile and a vehicle-mounted radar according to an embodiment of the invention with reference to the accompanying drawings.
FIG. 1 is a flow chart of a multi-target clustering method of vehicle-mounted radar according to an embodiment of the invention. As shown in fig. 1, the clustering method includes the following steps:
s1, acquiring the original target array to be polled, and marking each element in the original target array as an unaccessed state.
Each element in the original target array corresponds to one target of the vehicle-mounted radar.
It should be noted that the purpose of clustering is to re-cluster the same target to target points spread to different positions of the spectrum, and therefore, in this embodiment, all the vehicle-mounted radar targets are initialized first, and all the vehicle-mounted radar targets are marked as an unaccessed state.
And S2, polling each element in the original target array, and classifying the elements in the original target array according to the parameter information of the element in the unaccessed state being polled and the parameter information of the rest elements in the original target array except the element being polled to obtain a new array corresponding to the element in the unaccessed state.
The parameter information of the element at least comprises the amplitude of the corresponding target, namely the amplitude of a signal between the vehicle-mounted radar target and the vehicle-mounted radar.
Specifically, calculating Euler distances between the element in the non-access state being polled and other elements except the element being polled in the original target array, and marking the state of the element in the non-access state being polled as the accessed state; acquiring all elements with Euler distances smaller than a preset distance, and recording the number M of the elements; if M is larger than the preset value, M elements with the Euler distances smaller than the preset distance are classified into one type. And if M is less than or equal to the preset value, returning to execute the step S2 to continue polling the elements in the original target array, or returning to execute the step S3 to continue polling the elements in the new array.
And the parameter information of the element also comprises the radial distance and the radial speed of the corresponding target and the vehicle-mounted radar.
Optionally, the preset distance and the preset value may be calibrated as required.
In this embodiment, the euler distance S between two targets can be calculated by the following equation (1):
S＝(Sa-Sb)2+(Va-Vb)2 (1)
wherein S isa、SbRespectively, the radial distance, V, between the targets a, b and the radar on boarda、VbRespectively the radial velocity between the targets a, b and the vehicle radar.
It should be noted that, when clustering is performed, the more dimensions considered, the higher the accuracy of target clustering. Therefore, based on the consideration of the calculated amount, compared with the clustering method which only takes one-dimensional distance as the judgment standard in the related art, the clustering method of the invention takes two-dimensional Euler distance as the judgment standard, and the clustering accuracy is high.
For example, when the original target array X is [ a1, a2, a3, a4], when the polling is started, the polling is performed to a1, the original target array X is in an unaccessed state, a1 is marked as an accessed state, euler distances S (a1, a2), S (a1, a3) and S (a1, a4) between a1 and a2, a1 and a3, and a1 and a4 are respectively calculated, if S (a1, a2) and S (a1, a3) are smaller than a preset distance, elements are a1, a2 and a3, and the number 3 is greater than a preset value 2, the elements in X are classified to obtain a new array [ a1, a2, a3] containing a 1.
And S3, polling each element in the new array, and classifying the elements in the original target array according to the parameter information of the element in the unaccessed state being polled and the parameter information of the rest elements in the original target array except the element being polled to obtain a cluster number array corresponding to the element in the unaccessed state.
For example, when polling the elements in the new array [ a1, a2, a3], a1 is in an accessed state, the polling is continued, when polling reaches a2, a2 is marked as accessed, S (a2, a1), S (a2, a3), S (a2, a4) are calculated, if S (a2, a1), S (a2, a4) are smaller than a preset distance, the elements are a1, a2, a4, and the number is 3 greater than a preset value 2, the elements in X are classified to obtain a cluster number group [ a1, a2, a4] containing a 2.
It can be understood that if S (a2, a1) and S (a2, a3) are smaller than the preset distance, the elements are a1, a2 and a3, and the number of the elements is 3 and is greater than the preset value 2, the elements in X are classified to obtain a cluster number group [ a1, a2 and a3] containing a 2.
It should be noted that, in order to reduce the computational complexity, the euler distance is calculated only once when any two elements in the original target array are calculated, i.e. when polling a1, S (a1, a2) is already calculated, and therefore, when polling a2, the calculation result S (a1, a2) of polling a1 can be used.
S4, polling each element in the cluster array, and adding the element in the cluster array in the state of not being accessed to the new array corresponding to the cluster array.
It should be noted that, while the element in the cluster array in the non-accessed state is added to the new array corresponding to the cluster array, the element in the non-accessed state is marked as accessed.
For example, if the cluster array is [ a1, a2, a4], a1, a2 are marked as accessed, and a4 is in an unaccessed state, a4 is added to the new array corresponding to the cluster array to obtain an updated new array [ a1, a2, a3, a4], and a4 is marked as accessed.
Similarly, if the cluster array is [ a1, a2, a3], and a1, a2 are marked as accessed, and a3 is marked as not accessed, a3 is added to the new array corresponding to the cluster array to obtain an updated new array [ a1, a2, a3], and meanwhile, a3 is marked as accessed.
S5, when the polling to the cluster array is completed, the step S3 is executed, and when the polling to the new array is completed, the step S2 is executed.
For example, if the cluster array is [ a1, a2, a4], after the polling is completed, the corresponding new array [ a1, a2, a3], a1, a2 is accessed, and a3 is in an unaccessed state. When polling a3, flag a3 has been accessed and S (a3, a1), S (a3, a2), S (a3, a4) are calculated. If only S (a3, a1) is less than the preset distance, then the original target array continues to be polled.
S6, when the original target array is polled, acquiring the element with the maximum amplitude value in each new array, and adding the element with the maximum amplitude value in each new array to the clustering target array to complete target clustering.
For example, after the original target array is polled, the obtained new arrays are [ a1, a2, a3, a3] and [ a1, a3, a4], the corresponding elements with the largest amplitudes are a1 and a4 respectively, and then a1 and a4 form a clustering target array [ a1, a4], and target clustering is completed.
It should be noted that, under the same condition, the RCS (Radar-Cross Section) value of the real target is 5-10 times larger than that of the noise or false target, and the larger the RCS value is, the higher the probability of obtaining the real target is, that is, the RCS value and the amplitude of the target are in direct proportion. Therefore, in order to improve the probability of the real target and ensure the accuracy of the clustering target, the invention selects the element with the maximum amplitude in each new array as the final target. In the related technology, the purpose of early warning is achieved only by clustering targets to obtain whether the average distance variation trend of the clusters and the early warning condition are met, and a good clustering effect is not achieved.
Also, the larger the generic target, the larger its RCS value. For targets with too high RCS values, such as large trucks with large volumes, many algorithms cluster these targets into multiple targets, but the method according to the present invention can still cluster these targets into one target.
In a specific embodiment of the present invention, after initializing all vehicle-mounted radar targets forming an original target array, i.e., marking all vehicle-mounted radar targets in an unaccessed state, a process of clustering by using the multi-target clustering method of the vehicle-mounted radar of the embodiment of the present invention is described with reference to the flowchart shown in fig. 2:
the method comprises the following steps: polling each element in the original target array, if the element being polled is in an unvisited state, marking the element as a visited state, calculating the Euler distances between the element and all other elements of the original target array, and if the element being polled is in a visited state, continuing to poll the element in the original target array.
Step two: and acquiring all elements of which the Euler distances are smaller than the preset distance Eps, and recording the number of the elements.
Step three: if the number of the recorded elements is larger than a preset value Min _ Pts, all elements with the Euler distance smaller than Eps are combined into a new array, and an ID is marked for the new array; otherwise, continuing to poll the original target array.
Step four: polling each element in the new array, and if the element being polled is in the accessed state, continuing to poll the new array; otherwise, the element is marked as a visited state, and the Euler distance between the element and other elements except the element in the original target array is calculated.
Step five: and acquiring all elements with Euler distances smaller than Eps, and recording the number of the elements.
Step six: if the number of the recorded elements is larger than Min _ Pts, all elements with the Euler distance smaller than Eps are formed into a cluster array; else continue polling for new array elements.
Step seven: polling each element in the cluster array, and if the element being polled is in the accessed state, continuing to poll the cluster array; otherwise, the element is added to the corresponding new array.
Step eight: if the cluster number group in the step seven completes the polling, returning to execute the step four; and if the new array in the step four completes the polling, returning to execute the step one.
Step nine: and if the original target array in the step one is polled, adding the element with the maximum amplitude value in each new array into the clustering target array to finish clustering the vehicle-mounted radar targets.
Therefore, the Euler distance is used as a judgment standard, and polling of multiple classification arrays is performed, so that the elements with the largest amplitude in each new array form a clustering target array, the accuracy of clustering the vehicle-mounted radar targets is improved, and a good clustering effect is achieved on large targets.
For example, an example is given in which the vehicle-mounted radar target includes six targets a, b, c, d, e, and f, and the original target array X ═ a, b, c, d, e, and f]And each element is in an unaccessed state. Polling each element a-f in X, marking a as visited when polling a in a non-visited state, and respectively calculating Euler distances S between a and b, c, d, e, fab、Sac、Sad、Sae、Saf. If S isab、SaeIf the distance is less than the preset distance Eps, the corresponding elements are a, b and e, the number of the elements is 3, and the distance is greater than the preset value Min _ Pts (for example, Min _ Pts is 2), the new array Y1 formed by the elements a, b and e is [ a, b, e ═]And the new array is marked with ID 1.
Then, polling each element in the new array Y1, and when a is polled, a is in an accessed state, and then continuing to poll the new array; when b is polled and is in an unaccessed state, b is marked as accessed, the ID of a new array where b is located is marked as 1, and the Euler distance S between b and other elements a, c, d, e and f of the original target array except b is calculatedba、Sbc、Sbd、Sbe、Sbf. If S isba、SbdIf the number of the corresponding elements is less than Eps, the number of the corresponding elements is a, b and d, the number of the elements is 3, and if the number of the elements is greater than the preset value 2, the elements a, b and d form a cluster array Z1 ═ a, b and d]。
Further, polling each element in the cluster array Z1, when polling the element a, b, a, b are in the accessed state, then continuing to poll the cluster array; when d is in an unaccessed state when polling d, d is added to the new array with ID equal to 1, and the new array Y1 with updated ID equal to 1 is [ a, b, e, d ], and d is marked as accessed. At this time, polling of cluster number group Z is completed.
Further, continue polling the new array Y1 ═ a, b, e]Each element of (1). When polling a, b, a, b are in the accessed state, and continue polling Y1. When polling to e, e is in an unaccessed state, marking e as accessed, and marking the ID of a new array where e is located as 1. And calculating to obtain Sea、SecIf the number of the corresponding elements is less than Eps, the corresponding elements are a, e and c, the number of the elements is 3, and if the number of the elements is greater than the preset value 2, the a, e and c form a cluster array Z2 ═ a, c and e]. Polling Z2, c is in an unaccessed state, c is added to the new array with ID 1, and the updated new array with ID 1, Y1 [ a, b, c, e, d ] } is added to the new array with ID 1]While c is marked as accessed. At this point, the new array [ a, b, e ] is complete]Is polled.
And continuing to poll the elements in the original target array X, wherein a, b, c, d and e are all in an accessed state, when f is polled to f, f is in an unvisited state and marked as accessed, and since a, b, c, d and e are classified into one type, f can be regarded as a non-clustering target. At this time, a, b, c, d, e, and f are all in the visited state, which indicates that polling on X is completed, and a new array Y1 with ID equal to 1 (e.g., a) with the largest amplitude is used to form a cluster target array T equal to [ a ], so as to complete clustering on the vehicle-mounted radar target. It should be noted that there is not necessarily only one element with the largest amplitude in the new array, but there may be two, three, etc.
In conclusion, according to the multi-target clustering method for the vehicle-mounted radar, provided by the embodiment of the invention, the Euler distance is taken as a judgment standard, the clustering accuracy of the vehicle-mounted radar target can be improved, and a good clustering effect is achieved on a large target.
Further, the present application proposes a computer-readable storage medium on which a computer program is stored, which when executed by a processor implements the multi-target clustering method of the vehicle-mounted radar of the above-described embodiments.
The computer-readable storage medium of the embodiment of the invention can improve the clustering accuracy of the vehicle-mounted radar target and has good clustering effect on a large target by executing the program corresponding to the multi-target clustering method of the vehicle-mounted radar stored on the computer-readable storage medium.
Fig. 3 is a multi-target clustering apparatus of a vehicle radar according to an embodiment of the present invention. As shown in fig. 3, the clustering apparatus 100 includes: an acquisition module 10, a tagging module 20, a first polling module 30, a second polling module 40, a third polling module 50, and a control module 60.
The obtaining module 10 is configured to obtain an original target array to be polled. The marking module 20 is configured to mark each element in the original target array as an unaccessed state, where each element in the original target array corresponds to a target of the vehicle-mounted radar. The first polling module 30 is configured to poll each element in the original target array, and classify the elements in the original target array according to the parameter information of the element in the unaccessed state being polled and the parameter information of the remaining elements in the original target array except the element being polled to obtain a new array corresponding to the element in the unaccessed state, where the parameter information of the element at least includes the amplitude of the corresponding target. The second polling module 40 is configured to poll each element in the new array, and classify the elements in the original target array according to the parameter information of the element in the unaccessed state being polled and the parameter information of the remaining elements in the original target array except the element being polled to obtain a cluster number group corresponding to the element in the unaccessed state. The third polling module 50 is configured to poll each element in the cluster array and add the element in the cluster array that is in the unaccessed state to a new array corresponding to the cluster array. The control module 60 is configured to control the second polling module to continue polling the elements in the new array after the third polling module completes polling the cluster array, control the first polling module to continue polling the elements in the original target array after the second polling module completes polling the new array, obtain the element with the largest amplitude in each new array after the first polling module completes polling the original target array, and add the element with the largest amplitude in each new array to the clustering target array to complete target clustering.
In some embodiments of the present invention, when classifying the elements in the original target array according to the parameter information of the element in the unaccessed state being polled and the parameter information of the remaining elements in the original target array except the element being polled, the first polling module 30 is specifically configured to calculate the euler distance between the element in the unaccessed state being polled in the original target array and the remaining elements in the original target array except the element being polled, and mark the state of the element in the unaccessed state being polled as the accessed state; acquiring all elements with Euler distances smaller than a preset distance, and recording the number M of the elements; and if the M is larger than the preset value, forming a new array by the M elements with the Euler distances smaller than the preset distance.
The second polling module 40 is specifically configured to calculate an euler distance between the element in the unaccessed state being polled in the new array and the remaining elements in the original target array except for the element being polled when classifying the elements in the original target array according to the parameter information of the element in the unaccessed state being polled and the parameter information of the remaining elements in the original target array except for the element being polled, and mark the state of the element in the unaccessed state being polled as an accessed state; acquiring all elements with Euler distances smaller than a preset distance, and recording the number M of the elements; and if the M is larger than the preset value, forming a cluster array by the M elements with the Euler distances smaller than the preset distance.
In some embodiments of the invention, the parameter information of the element further comprises a radial distance and a radial velocity of the corresponding target from the onboard radar.
Specifically, the euler distance S between two targets can be calculated by the following formula (1):
S＝(Sa-Sb)2+(Va-Vb)2 (1)
wherein S isa、SbRespectively, the radial distance, V, between the targets a, b and the radar on boarda、VbRespectively the radial velocity between the targets a, b and the vehicle radar.
In some embodiments of the present invention, the control module 60 is further configured to control the first polling module 30 to continue polling the elements in the original target array or control the second polling module 40 to continue polling the elements in the new array when M is less than or equal to the preset value.
It should be noted that, for a specific implementation of the multi-target clustering device for a vehicle-mounted radar in the embodiment of the present invention, reference may be made to a specific implementation of the multi-target clustering method for a vehicle-mounted radar in the above-mentioned embodiment of the present invention, and details are not repeated here in order to reduce redundancy.
According to the multi-target clustering device of the vehicle-mounted radar, the Euler distance is used as a judgment standard, the clustering accuracy of the vehicle-mounted radar target can be improved, and a good clustering effect is achieved on a large target.
Fig. 4 is a block diagram of an automobile according to an embodiment of the present invention. As shown in fig. 4, the automobile 1000 includes a multi-target clustering device 100 of an on-vehicle radar.
According to the automobile provided by the embodiment of the invention, the multi-target clustering device of the vehicle-mounted radar can improve the clustering accuracy of the vehicle-mounted radar target, and has a good clustering effect on a large target.
It should be noted that the logic and/or steps represented in the flowcharts or otherwise described herein, such as an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.
It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, any one or a combination of the following techniques, which are well known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
In the description of the present invention, it is to be understood that the terms "central," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," "circumferential," and the like are used in the orientations and positional relationships indicated in the drawings for convenience in describing the invention and to simplify the description, and are not intended to indicate or imply that the referenced devices or elements must have a particular orientation, be constructed and operated in a particular orientation, and are therefore not to be considered limiting of the invention.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present invention, "a plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.
In the present invention, unless otherwise expressly stated or limited, the terms "mounted," "connected," "secured," and the like are to be construed broadly and can, for example, be fixedly connected, detachably connected, or integrally formed; can be mechanically or electrically connected; they may be directly connected or indirectly connected through intervening media, or they may be connected internally or in any other suitable relationship, unless expressly stated otherwise. The specific meanings of the above terms in the present invention can be understood by those skilled in the art according to specific situations.
In the present invention, unless otherwise expressly stated or limited, the first feature "on" or "under" the second feature may be directly contacting the first and second features or indirectly contacting the first and second features through an intermediate. Also, a first feature "on," "over," and "above" a second feature may be directly or diagonally above the second feature, or may simply indicate that the first feature is at a higher level than the second feature. A first feature being "under," "below," and "beneath" a second feature may be directly under or obliquely under the first feature, or may simply mean that the first feature is at a lesser elevation than the second feature.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made to the above embodiments by those of ordinary skill in the art within the scope of the present invention.
Claims (6)
1. A multi-target clustering method for a vehicle-mounted radar is characterized by comprising the following steps:
s1, acquiring an original target array to be polled, and marking each element in the original target array as an unaccessed state, wherein each element in the original target array corresponds to a target of the vehicle-mounted radar;
s2, polling each element in the original target array, and classifying the elements in the original target array according to the parameter information of the polling element in the unaccessed state and the parameter information of the remaining elements in the original target array except the polling element to obtain a new array corresponding to the polling element, including: calculating Euler distances between the element in the non-access state which is being polled and other elements except the element in the polling state in the original target array, and marking the state of the element in the non-access state which is being polled as an access state; acquiring all elements with Euler distances smaller than a preset distance, and recording the number M of the elements; if M is larger than a preset value, classifying M elements with Euler distances smaller than the preset distance into one class, and marking the state of the polled elements in the original target array as an accessed state; the parameter information of the elements further comprises the radial distance and the radial speed between the corresponding target and the vehicle-mounted radar, and the Euler distance S between the two targets is calculated by the following formula:
S＝(Sa-Sb)2+(Va-Vb)2，
wherein S isa、SbRespectively, the radial distance, V, between the targets a, b and the radar on boarda、VbThe radial speeds between the targets a and b and the vehicle-mounted radar are respectively;
s3, polling each element in the new array, classifying the elements in the original target array according to the parameter information of the element in the unaccessed state being polled and the parameter information of the rest elements in the original target array except the element being polled to obtain a cluster number array corresponding to the element in the unaccessed state, and marking the state of the element polled in the new array as the accessed state;
s4, polling each element in the cluster array, adding the element in the cluster array in the state of not being accessed to a new array corresponding to the cluster array, and marking the state of the polled element in the cluster array as the accessed state;
s5, when the polling to the cluster array is completed, returning to execute the step S3, and when the polling to the new array is completed, returning to execute the step S2;
s6, when the original target array is polled, acquiring the element with the maximum amplitude in each new array, and adding the element with the maximum amplitude in each new array to the clustering target array to finish target clustering, wherein the parameter information of the element at least comprises the amplitude of the corresponding target.
2. The multi-target clustering method for vehicle-mounted radars according to claim 1, wherein if M is less than or equal to the preset value, returning to perform step S2 to continue polling the elements in the original target array, or returning to perform step S3 to continue polling the elements in the new array.
3. A computer-readable storage medium on which a computer program is stored, characterized in that the program, when processed and executed, implements a multi-objective clustering method of vehicle-mounted radars according to any one of claims 1-2.
4. A multi-target clustering device for vehicle-mounted radar is characterized by comprising:
the acquisition module is used for acquiring an original target array to be polled;
the marking module is used for marking each element in the original target array into an unvisited state, wherein each element in the original target array corresponds to one target of the vehicle-mounted radar;
a first polling module, configured to poll each element in the original target array, and classify the elements in the original target array according to parameter information of an element being polled and parameter information of remaining elements in the original target array except the element being polled to obtain a new array corresponding to the element being in the inaccessible state, where the first polling module includes: calculating Euler distances between the element in the non-access state which is being polled and other elements except the element in the polling state in the original target array, and marking the state of the element in the non-access state which is being polled as an access state; acquiring all elements with Euler distances smaller than a preset distance, and recording the number M of the elements; if M is larger than a preset value, classifying M elements with Euler distances smaller than the preset distance into one class, and marking the state of the polled elements in the original target array as an accessed state; the parameter information of the elements further comprises the radial distance and the radial speed between the corresponding target and the vehicle-mounted radar, and the Euler distance S between the two targets is calculated by the following formula:
S＝(Sa-Sb)2+(Va-Vb)2，
wherein S isa、SbRespectively, the radial distance, V, between the targets a, b and the radar on boarda、VbAre respectively the targeta. b radial velocity with the vehicle radar;
the second polling module is used for polling each element in the new array, classifying the elements in the original target array according to the parameter information of the element which is in the non-access state and the parameter information of the rest elements except the element which is in the polling state in the original target array to obtain a cluster number array corresponding to the element which is in the non-access state, and marking the state of the element which is polled in the new array as the access state;
the third polling module is used for polling each element in the cluster array, adding the element in the non-access state in the cluster array to a new array corresponding to the cluster array, and marking the state of the polled element in the cluster array as an accessed state;
and the control module is used for controlling the second polling module to continuously poll the elements in the new array after the third polling module finishes polling the cluster array, controlling the first polling module to continuously poll the elements in the original target array after the second polling module finishes polling the new array, acquiring the element with the largest amplitude in each new array after the first polling module finishes polling the original target array, and adding the element with the largest amplitude in each new array into the clustering target array to finish target clustering, wherein the parameter information of the element at least comprises the amplitude of the corresponding target.
5. The multi-target clustering apparatus of vehicle-mounted radars according to claim 4, wherein the control module is further configured to:
and when M is less than or equal to the preset value, controlling the first polling module to continuously poll the elements in the original target array, or controlling the second polling module to continuously poll the elements in the new array.
6. An automobile characterized by comprising the multi-target clustering device of the vehicle-mounted radar according to any one of claims 4 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710887429.0A CN109543704B (en) | 2017-09-22 | 2017-09-22 | Multi-target clustering method and device for automobile and vehicle-mounted radar |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710887429.0A CN109543704B (en) | 2017-09-22 | 2017-09-22 | Multi-target clustering method and device for automobile and vehicle-mounted radar |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109543704A CN109543704A (en) | 2019-03-29 |
| CN109543704B true CN109543704B (en) | 2021-02-23 |
Family
ID=65830736
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710887429.0A Active CN109543704B (en) | 2017-09-22 | 2017-09-22 | Multi-target clustering method and device for automobile and vehicle-mounted radar |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109543704B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110988856B (en) * | 2019-12-19 | 2021-08-03 | 电子科技大学 | A Density Clustering-Based Target Detection Point Trace Agglomeration Algorithm |
| CN111402629B (en) * | 2020-03-06 | 2021-11-12 | 杭州海康威视系统技术有限公司 | Vehicle turning control method and device and electronic equipment |
| CN111369828B (en) * | 2020-03-09 | 2022-02-01 | 杭州海康威视系统技术有限公司 | Safety early warning system and method for vehicle turning blind area |
| WO2021189328A1 (en) * | 2020-03-25 | 2021-09-30 | 华为技术有限公司 | Radar target clustering method and apparatus |
| CN112689775B (en) * | 2020-04-29 | 2022-06-14 | 华为技术有限公司 | Radar point cloud clustering method and device |
| CN112505682B (en) * | 2020-11-16 | 2022-12-23 | 上海无线电设备研究所 | Missile-borne radar multi-target track initial association method, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101430708A (en) * | 2008-11-21 | 2009-05-13 | 哈尔滨工业大学深圳研究生院 | Blog hierarchy classification tree construction method based on label clustering |
| CN105242271A (en) * | 2014-07-03 | 2016-01-13 | 通用汽车环球科技运作有限责任公司 | Object classification for vehicle radar systems |
| CN106096639A (en) * | 2016-06-06 | 2016-11-09 | 华侨大学 | A kind of clustering method of ionospheric radar data based on density temperature |
| CN106355194A (en) * | 2016-08-22 | 2017-01-25 | 广东华中科技大学工业技术研究院 | A surface target processing method for unmanned boats based on laser imaging radar |
| CN106557777A (en) * | 2016-10-17 | 2017-04-05 | 中国互联网络信息中心 | It is a kind of to be based on the improved Kmeans clustering methods of SimHash |
-
2017
- 2017-09-22 CN CN201710887429.0A patent/CN109543704B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101430708A (en) * | 2008-11-21 | 2009-05-13 | 哈尔滨工业大学深圳研究生院 | Blog hierarchy classification tree construction method based on label clustering |
| CN105242271A (en) * | 2014-07-03 | 2016-01-13 | 通用汽车环球科技运作有限责任公司 | Object classification for vehicle radar systems |
| CN106096639A (en) * | 2016-06-06 | 2016-11-09 | 华侨大学 | A kind of clustering method of ionospheric radar data based on density temperature |
| CN106355194A (en) * | 2016-08-22 | 2017-01-25 | 广东华中科技大学工业技术研究院 | A surface target processing method for unmanned boats based on laser imaging radar |
| CN106557777A (en) * | 2016-10-17 | 2017-04-05 | 中国互联网络信息中心 | It is a kind of to be based on the improved Kmeans clustering methods of SimHash |
Non-Patent Citations (2)
| Title |
|---|
| 基于改进DBSCAN算法的激光雷达目标物检测方法;黄钢;《交通信息与安全》;20150331;第33卷(第3期);论文第1-3节 * |
| 聚类方法：DBSCAN算法研究;XianenZhou;《https://blog.csdn.net/zhouxianen1987/article/details/68945844》;20170401;页面全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109543704A (en) | 2019-03-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109543704B (en) | Multi-target clustering method and device for automobile and vehicle-mounted radar | |
| CN109425852B (en) | Calibration method and device for automobile and vehicle-mounted radar | |
| CN113109802B (en) | Target motion state judging method, device, radar equipment and storage medium | |
| CN115616554A (en) | Configure the vehicle radar system to determine the free space | |
| US20160103208A1 (en) | System for filtering lidar data in vehicle and method thereof | |
| CN112597839B (en) | Road boundary detection method based on vehicle-mounted millimeter wave radar | |
| CN110659548B (en) | Vehicle and target detection method and device thereof | |
| CN110857983B (en) | Object velocity vector estimation using multiple radars with different observation angles | |
| CN113536850A (en) | Target object size testing method and device based on 77G millimeter wave radar | |
| CN113460084A (en) | Method and device for determining vehicle driving intention, vehicle and storage medium | |
| CN110758244B (en) | Method and system for automatically preventing rear-end collision of electric automobile | |
| US11428782B2 (en) | Neural network-based object surface estimation in radar system | |
| CN108020839A (en) | The vehicle detection of external object | |
| CN115542346A (en) | Ground detection method, device, vehicle and storage medium | |
| CN109263649B (en) | Vehicle, object recognition method and object recognition system thereof in automatic driving mode | |
| CN111562591B (en) | Automobile position measuring method and device based on laser radar | |
| CN119001621A (en) | Decoy suppression method and device, electronic equipment, storage medium and program product | |
| US11030764B2 (en) | Method and system for trailer size estimating and monitoring | |
| CN108116406A (en) | A kind of general automated parking system and method for popping one's head in | |
| JP6555132B2 (en) | Moving object detection device | |
| CN113589288A (en) | Target screening method, device and equipment based on millimeter wave radar and storage medium | |
| CN118294910A (en) | Detection threshold determining method, device, radar and storage medium | |
| CN114442075A (en) | A radar parameter configuration method, device and vehicle | |
| CN117970254A (en) | Method and device for filtering radar reflection points | |
| CN113428164A (en) | Driving habit learning method and device and computer readable 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |