CN117111053B

CN117111053B - Millimeter wave radar tracking result processing method, device and processing equipment

Info

Publication number: CN117111053B
Application number: CN202311368567.XA
Authority: CN
Inventors: 朱星; 左海波
Original assignee: Wuhan Future Phantom Technology Co Ltd
Current assignee: Wuhan Future Phantom Technology Co Ltd
Priority date: 2023-10-23
Filing date: 2023-10-23
Publication date: 2024-02-27
Anticipated expiration: 2043-10-23
Also published as: CN117111053A

Abstract

The application provides a processing method, a device and processing equipment for millimeter wave radar tracking results, which are used for optimizing the millimeter wave radar tracking results through an optimization framework constructed by first clustering processing, cluster merging processing, second clustering processing and matching processing so as to avoid the condition that a plurality of tracking identifications id appear in a real object, thereby providing more accurate data support for the millimeter wave radar tracking results, guaranteeing related user experience and reducing vehicle driving risks.

Description

Millimeter wave radar tracking result processing method, device and processing equipment

Technical Field

The application relates to the field of millimeter wave radar tracking, in particular to a method and a device for processing millimeter wave radar tracking results and processing equipment.

Background

In autopilot, it is easy to understand that obstacle perception is a very important link, thereby providing accurate and effective data support for autopilot. In specific application, the millimeter wave radar is a large obstacle sensing scheme, and a 2/3D image-based visual sensing scheme is also used, so that the two schemes can be used for fusing sensing results to realize a better obstacle sensing effect.

For a radar tracking module (also referred to as a millimeter wave radar product by a radar sensing module in the market) using a millimeter wave radar, the barrier target level information is usually output, and [ id, longitude_dist, laser_dist, longitude_vel, laser_vel ] respectively represents a tracking identifier id, a longitudinal distance, a transverse distance, a longitudinal speed, and a transverse speed.

However, the inventor of the application finds that in the radar tracking result output by the radar tracking module, ideally, one object should only correspond to one tracking identifier id and can be continuously tracked, but when the obstacles are relatively close or relatively large in practical situations, one object outputs a plurality of tracking identifiers ids, namely detection noise is generated, and the plurality of tracking identifiers ids continuously output redundant obstacle information in the follow-up processing such as fusion matching and the like, so that the vehicle in an automatic driving working mode is usually caused to stop autonomously, and obviously, the problem that the tracking identifiers id are not accurately recognized enough can be solved, the user experience can be reduced, and the driving risk of the vehicle can be improved.

Disclosure of Invention

In a first aspect, the present application provides a method for processing a millimeter wave radar tracking result, where the method includes:

acquiring a first millimeter wave radar tracking result output by a radar tracking module, wherein the radar tracking module is configured on a vehicle, and the first millimeter wave radar tracking result is obtained by the radar tracking module executing a sensing task on an obstacle around the vehicle in the working process of the vehicle;

performing first clustering processing on the first millimeter wave radar tracking result to obtain a second millimeter wave radar tracking result;

based on different clustering clusters related to the second millimeter wave radar tracking result, carrying out cluster combination processing on the second millimeter wave radar tracking result to obtain different third millimeter wave radar tracking results corresponding to the different clustering clusters;

sequentially carrying out second aggregation processing based on time sequence on different third millimeter wave radar tracking results with single frame properties and a currently existing track set to obtain different fourth millimeter wave radar tracking results;

taking the degree of distancing as a reference, carrying out matching processing between different fourth millimeter wave radar tracking results and a track set existing at present to obtain a matching result;

and updating the track set existing currently based on the matching result to obtain a fifth millimeter wave radar tracking result as output.

In a second aspect, the present application provides a processing apparatus for millimeter wave radar tracking results, where the apparatus includes:

the system comprises an acquisition unit, a radar tracking module and a control unit, wherein the acquisition unit is used for acquiring a first millimeter wave radar tracking result output by the radar tracking module, the radar tracking module is configured on a vehicle, and the first millimeter wave radar tracking result is obtained by the radar tracking module executing a sensing task on an obstacle around the vehicle in the working process of the vehicle;

the first clustering unit is used for carrying out first clustering processing on the first millimeter wave radar tracking result to obtain a second millimeter wave radar tracking result;

the cluster merging unit is used for carrying out cluster merging processing on the second millimeter wave radar tracking result based on different clusters related to the second millimeter wave radar tracking result to obtain different third millimeter wave radar tracking results corresponding to different clusters;

the second aggregation unit is used for sequentially carrying out second aggregation processing based on time sequences on different third millimeter wave radar tracking results with single frame properties and the track set existing at present to obtain different fourth millimeter wave radar tracking results;

the matching unit is used for carrying out matching processing between different fourth millimeter wave radar tracking results and a track set existing at present by taking the degree of distancing as a reference to obtain a matching result;

And the updating unit is used for updating the current track set based on the matching result to obtain a fifth millimeter wave radar tracking result as output.

In a third aspect, the present application provides a processing device, comprising a processor and a memory, the memory having stored therein a computer program, the processor executing the method provided by the first aspect of the present application or any one of the possible implementations of the first aspect of the present application when calling the computer program in the memory.

In a fourth aspect, the present application provides a computer readable storage medium storing a plurality of instructions adapted to be loaded by a processor to perform the method provided in the first aspect of the present application or any one of the possible implementations of the first aspect of the present application.

From the above, the present application has the following advantages:

according to the method, after a first millimeter wave radar tracking result output by a radar tracking module is obtained, first clustering processing is conducted on the first millimeter wave radar tracking result to obtain a second millimeter wave radar tracking result, then cluster merging processing is conducted on the second millimeter wave radar tracking result based on different clusters related to the second millimeter wave radar tracking result to obtain different third millimeter wave radar tracking results corresponding to different clusters, sequential second clustering processing is conducted on the different third millimeter wave radar tracking results with the current track set in a single frame mode to obtain different fourth millimeter wave radar tracking results, then the distance degree is used as a reference, matching processing is conducted on the different fourth millimeter wave radar tracking results with the current track set to obtain a matching result, at the moment, the current track set is updated based on the matching result to obtain a fifth millimeter wave radar tracking result, in the process, the optimizing architecture is constructed through the first clustering processing, the cluster merging processing, the second clustering processing and the matching processing, the obtained optimizing architecture is used for optimizing the radar tracking result, so that a plurality of real objects are prevented from being identified, the fact that the driving risk of a vehicle can be more accurately guaranteed, and the driving risk of a user can be more accurately tracked by the millimeter wave is guaranteed.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the description of the embodiments will be briefly introduced below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

Fig. 1 is a schematic view of a scenario of a method for processing millimeter wave radar tracking results in the present application;

FIG. 2 is a schematic flow chart of a method for processing millimeter wave radar tracking results;

FIG. 3 is a schematic view of a scenario of four cluster merge schemes of the present application;

fig. 4 is a schematic structural diagram of a processing device for millimeter wave radar tracking results in the present application;

fig. 5 is a schematic view of a structure of the processing apparatus of the present application.

Detailed Description

The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all, of the embodiments of the present application. All other embodiments, which can be made by those skilled in the art based on the embodiments herein without making any inventive effort, are intended to be within the scope of the present application.

The terms first, second and the like in the description and in the claims of the present application and in the above-described figures, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments described herein may be implemented in other sequences than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or modules is not necessarily limited to those steps or modules that are expressly listed or inherent to such process, method, article, or apparatus. The naming or numbering of the steps in the present application does not mean that the steps in the method flow must be executed according to the time/logic sequence indicated by the naming or numbering, and the execution sequence of the steps in the flow that are named or numbered may be changed according to the technical purpose to be achieved, so long as the same or similar technical effects can be achieved.

The division of the modules in the present application is a logical division, and may be implemented in another manner in practical application, for example, a plurality of modules may be combined or integrated in another system, or some features may be omitted or not implemented, and in addition, coupling or direct coupling or communication connection between the modules that are shown or discussed may be through some interfaces, and indirect coupling or communication connection between the modules may be in an electrical or other similar form, which is not limited in this application. The modules or sub-modules described as separate components may or may not be physically separate, or may be distributed in a plurality of circuit modules, and some or all of the modules may be selected according to actual needs to achieve the purposes of the present application.

Before describing the processing method of the millimeter wave radar tracking result provided by the application, the background content related to the application is first described.

The millimeter wave radar tracking result processing method, the millimeter wave radar tracking result processing device and the computer readable storage medium can be applied to processing equipment and used for optimizing the millimeter wave radar tracking result through an optimization framework constructed by first clustering processing, cluster merging processing, second clustering processing and matching processing so as to avoid the condition that a plurality of tracking identifications id appear in a real object, thereby providing more accurate data support of the millimeter wave radar tracking result, guaranteeing relevant user experience and reducing vehicle driving risks.

According to the millimeter wave radar tracking result processing method, an execution main body can be a millimeter wave radar tracking result processing device or different types of processing Equipment such as a server, a physical host or User Equipment (UE) and the like integrated with the millimeter wave radar tracking result processing device. The processing device of the millimeter wave radar tracking result can be realized in a hardware or software mode, the UE can be specifically a terminal device such as a smart phone, a tablet computer, a notebook computer, a desktop computer or a personal digital assistant (Personal Digital Assistant, PDA), and the processing device can be arranged in a device cluster mode.

Specifically, the processing device executing the processing method of the millimeter wave radar tracking result of the application, or that is, the processing device carrying application service implementing the processing method of the millimeter wave radar tracking result of the application is mainly used for providing data support of the millimeter wave radar tracking result for automatic driving of the vehicle, and further can assist in identifying obstacles, so the processing device can be a relevant vehicle terminal on the vehicle, and can be linked with a radar tracking module on the vehicle, millimeter wave radar tracking processing on the vehicle for obstacles around the vehicle, particularly in front of the vehicle can be performed locally, but the processing device can also provide data support of the millimeter wave radar tracking result for automatic driving of the local vehicle through remote service, so the processing device can also be a server, a physical host and other devices, and in consideration of flexible requirements in application, whether in a local mode or a remote mode, the processing device can be configured by a user UE at hand, and can be used as an additional supplementary mode, and can play the effect of a flexible deployment scheme.

As an example, as shown in fig. 1, a schematic view of a scenario of a processing method of a millimeter wave radar tracking result in the present application may be inserted into a position between an output end of a radar tracking module and a sensor fusion process (a "multi-tracking id processing" module in a corresponding diagram) in a multi-sensor fusion architecture related to millimeter wave radar, image sensing and other sensing schemes, so as to solve the problem of multi-tracking id of the same object, achieve the effect of removing obstacle noise, thereby providing more accurate and effective data support for the sensor fusion process, and further providing more accurate and effective data support for decisions performed by a PNC module (Prediction), planning (Planning) and Control (Control).

Meanwhile, the processing of the millimeter wave radar tracking result is considered from the automatic driving perspective to provide data support of the millimeter wave radar tracking result for related equipment control systems, and on the other hand, the method can also play a role in early warning for related personnel such as drivers, namely, in a common vehicle driving scene, the method for processing the millimeter wave radar tracking result provided by the method can assist in perception recognition of obstacles around a vehicle, particularly in front of the vehicle, and then plays a role in timely and effective early warning by combining with early warning modes such as voice and the like, so that higher safety is assisted, and more stable and safe vehicle driving effect is obtained.

As a further example, the scheme of the application can be applied to a low-speed scene with the vehicle speed less than 10km/h, such as a driving field of second driving school subjects, wherein millimeter wave radar is a short-range detection radar, and the detection radius is less than 40m.

Obviously, the processing equipment for executing the processing method of the millimeter wave radar tracking result of the application, the specific equipment form and the deployment form of the processing equipment can be flexibly adjusted according to actual requirements, and therefore the processing equipment is not particularly limited.

Next, a method for processing the millimeter wave radar tracking result provided by the present application will be described.

First, referring to fig. 2, fig. 2 shows a schematic flow chart of a processing method of a millimeter wave radar tracking result according to the present application, and the processing method of a millimeter wave radar tracking result according to the present application may specifically include steps S201 to S206 as follows:

step S201, a first millimeter wave radar tracking result output by a radar tracking module is obtained, wherein the radar tracking module is configured on a vehicle, and the first millimeter wave radar tracking result is obtained by the radar tracking module executing a sensing task on an obstacle around the vehicle in the working process of the vehicle;

it can be understood that the method and the device for optimizing the millimeter wave radar tracking result are to be realized aiming at the millimeter wave radar tracking result acquired by the radar tracking module configured on the vehicle, so that data support is provided for obstacle recognition of the vehicle.

In general, the radar tracking module is an original structure on the vehicle, or may also be a structure configured on the vehicle for applying the scheme of the present application, which adopts a millimeter wave radar working mechanism and is a radar for detecting through a millimeter wave band, the module may also be referred to as a radar sensing module, etc., and the specific working principle and hardware structure thereof are not the key point of the scheme, and obviously reference may be made to the prior art, so the present application does not make specific description.

Furthermore, as to the radar tracking module, it is understood that in practical cases, it may also be part of the processing device of the present application, which may be adjusted according to the flexible device form and deployment form of the processing device of the present application.

For the radar tracking result acquired and output by the radar tracking module, the first millimeter wave radar tracking result is recorded for convenience of explanation.

The first millimeter wave radar tracking result obtained by the method is of a single-frame nature, and is colloquially data collected at a time point (corresponding to a time stamp).

Specifically, the content of the first millimeter wave radar tracking result under the current frame may specifically include tracking identifier id, longitudinal distance_dist, transverse distance_dist, longitudinal speed longitude_vel, transverse speed transverse_vel, width, height, length, and circumscribed polygon.

As an example, the first millimeter wave radar tracking result may be: { timestamp, obj1, obj2, & gt, obj1, obj2, & gt, wherein timestamp represents the timestamp of the current frame, obj1, obj2, & gt, obj N represents the corresponding object of N objects (objects) in the current frame, each object identification obj has information detected by the radar tracking module, such as { id, center, vector, width, height, length, poly, theta, & gt, center comprises the above mentioned longitudinal distance_discrete and transverse distance_discrete, and vector comprises the above mentioned longitudinal speed longitude_vel and transverse speed laser_vel.

Meanwhile, the radar tracking module processes the following contents according to tracking information designed by an original manufacturer in concrete implementation:

1) Mapping the original 0-255 direct id value to the 0-INT_MAX value;

2) Filtering some flickering noise points;

3) Smoothing centers, vectors, etc.;

4) Assigning the variables of width, height, length and the like to a specified size, such as 1m;

although the radar tracking module has a certain filtering capability for tracking noise, the radar tracking module mainly sets a threshold value according to the time required by the radar tracking module to generate a track, but the problem that one object has a plurality of tracking ids cannot be essentially solved (one object theoretically has only one tracking id).

Step S202, performing first clustering processing on the first millimeter wave radar tracking result to obtain a second millimeter wave radar tracking result;

after the original first millimeter wave radar tracking results such as { timestamp, obj1, obj2, & gt, obj n } are obtained, the optimization process of the present application for millimeter wave radar tracking results can be started.

Specifically, the first clustering process (which may also be referred to as primary clustering process, corresponding to the secondary clustering process that follows) introduced in the present application may be started first.

The clustering purpose is to search similar tracking objects in the first millimeter wave radar tracking result, prepare for the combination of the tracking objects and provide data support.

It is understood that the clustering process needs to be implemented through a pre-configured and related clustering algorithm, and the clustering process can be configured according to the needs of staff.

As an exemplary implementation manner, performing a first clustering process on the first millimeter wave radar tracking result to obtain a second millimeter wave radar tracking result may specifically include:

and performing first clustering processing on the first millimeter wave radar tracking result through a Density-based clustering algorithm (Density-Based Spatial Clustering of Applications with Noise, DBSCAN) with noise to obtain a second millimeter wave radar tracking result.

It can be understood that the DBSCAN divides the points which are relatively close to each other in the space into a class, marks the outer points which are positioned in the low-density area, is very suitable for being used under the condition of multiple noise of the millimeter wave radar tracking result related to the application, and has better clustering effect.

Of course, in particular applications, other types of clustering algorithms may be employed in addition to DBSCAN.

In addition, in order to promote better clustering effect, the measurement distance between the clusters can be properly adjusted, or an optimized quantization scheme of the measurement distance between the clusters is provided.

Specifically, as yet another exemplary implementation manner, performing the first clustering processing on the first millimeter wave radar tracking result to obtain the second millimeter wave radar tracking result may include:

performing first clustering processing on the first millimeter wave radar tracking result to obtain a second millimeter wave radar tracking result, and in the clustering process, calculating the measurement distance between clusters according to the following formula:

，

wherein,representing the current object +.>And cluster center->Metric distance between>Is the speed.

It can be seen that the present application perceives the speed of the radar tracking module The method also comprises the step of taking the measurement distance between the clusters into consideration, and the step of specifically realizing the measurement distance by the formula, wherein after the speed consideration is added, the measurement distance is more adapted to the dynamic recognition scene of the vehicle obstacle, so that the dynamic obstacle can be clustered more effectively, and the method can be further used for promoting the optimization of the millimeter wave radar tracking result to be more effective.

For the setting of the clustering result, the clustering clusters can be taken as distinction, the content of the first millimeter wave radar tracking result is rearranged to obtain millimeter wave radar tracking results of different clustering clusters, and a second millimeter wave radar tracking result is formed, or the second millimeter wave radar tracking result can be obtained by modifying and updating corresponding attributes/labels on the basis of the first millimeter wave radar tracking result, specifically, the second millimeter wave radar tracking result can be obtained by adding a cluster identification cluster_id on the basis of the first millimeter wave radar tracking result to indicate the clustering cluster to which the object belongs, and it can be understood that the former format and content of the first millimeter wave radar tracking result are not influenced as much as possible and the advantage of lower processing workload is achieved.

For example, in the overall first millimeter wave radar tracking result { timestamp, obj1, obj2, & gt, obj n } a cluster identification cluster_id is added for each millimeter wave radar tracking result { id, center, density, width, height, length, polygon, theta, & gt, to obtain { id, center, density, width, height, length, polygon, density, theta, & gt.

Step S203, based on different clusters related to the second millimeter wave radar tracking result, carrying out cluster merging processing on the second millimeter wave radar tracking result to obtain different third millimeter wave radar tracking results corresponding to different clusters;

after the preliminary clustering is completed and different clusters are obtained in step S202, corresponding cluster merging processing can be performed based on the clusters, specifically, in the second millimeter wave radar tracking result, based on each cluster, a plurality of corresponding millimeter wave radar tracking results are merged, so that a new millimeter wave radar tracking result is formed, and the new millimeter wave radar tracking result is recorded as a third millimeter wave radar tracking result in the application.

It can be understood that in the process of executing the cluster merging process, on the basis of merging the tracking objects, the attributes of the tracking objects need to be merged, so as to obtain a new tracking object which characterizes a plurality of tracking objects before the cluster merging process.

Specifically, as yet another exemplary implementation manner, in the second millimeter wave radar tracking result herein, the cluster merging processing performed on the millimeter wave radar tracking result of each cluster may specifically include the following processing contents:

Combining the objects { obj1, obj2, & gt, obj K } of the millimeter wave radar tracking results belonging to the same cluster into a new object obj, taking the average of the longitudinal speed longitude_vel and the transverse speed transverse_vel of the millimeter wave radar tracking results belonging to the same cluster with reference to a scene diagram of four cluster combining schemes of the application shown in fig. 3, wherein the longitudinal distance longitude_dist, the transverse distance transverse_dist, the width height and the length are determined by a scheme selected from the following four external polygon updating schemes (the following 4 schemes are from top to bottom, the 4 schemes of the corresponding fig. 3 are from left to right 4):

clustering of the search objects

Taking the object obj closest to the center of the rear axle of the vehicle as the clustered object obj;

center object clustering

Taking all objects { obj1, obj2,.. The obj k } to be the center in x-direction and y-direction, width height and length are unchanged;

clustering method of binding box

Taking the circumscribed polygon of all objects { obj1, obj2,.. The circumscribed polygon of obj k } as a unit, taking the smallest circumscribed rectangle, taking the direction as the vertical direction, and taking the angle theta = 90 degrees (under the cartesian coordinate system, the positive x-axis direction is 0 degree, the counterclockwise rotation of the x-axis is positive angle, 90 degrees is the positive y-axis direction, the clockwise x-axis is negative angle, the value range of the angle theta is [ -180,180], and the width length is the width and length of the smallest circumscribed rectangle;

Clustering method of orientation box

Taking the circumscribed polygon of all objects { obj1, obj2, & gt, the circumscribed polygon of obj K } as a unit, taking the smallest rotation circumscribed rectangle, wherein the direction is the direction of the rotation rectangle, and the width and the length are the width and the length of the smallest rotation circumscribed rectangle.

It can be understood that, for the above four cluster merging schemes/external polygon updating schemes, in specific operations, the cluster merging schemes/external polygon updating schemes may be selected randomly, or alternatively used in advance, that is, one may be preset, or may be selected in real time according to an adaptation policy, and configured according to specific requirements.

Step S204, sequentially carrying out second aggregation processing based on time sequence on different third millimeter wave radar tracking results with single frame properties and a track set existing at present to obtain different fourth millimeter wave radar tracking results;

as already mentioned above, the first millimeter wave radar tracking result obtained in step S210 is a single frame, is a millimeter wave radar tracking result corresponding to one time stamp, and after obtaining a corresponding third millimeter wave radar tracking result through cluster merging processing, may be associated with another frame/time stamp, and perform a second clustering processing (secondary clustering processing) based on time sequence.

The first clustering process mainly considers spatial attribute, combines radar tracking points which are relatively close in space into the same object, the second clustering process mainly considers time attribute, and certain families which are relatively far in space after the first clustering process is completed, if the families are close in time dimension (mainly reflected on id of front and back frames), the families are further clustered and combined, and the second clustering process is complementary to the first clustering process, so that the number and the size of the barriers can be obtained more accurately.

Specifically, as yet another exemplary implementation, the timing-based second aggregation process referred to herein may include the following:

id similarity calculation

Comparing cluster identification cluster_id set a= { id_c1, id_c2, &..} of the current third millimeter wave radar tracking result with track set t= { T1, T2, &..tm } existing currently, and if the intersection ratio of the cluster identification cluster_id set a= { id_c1, id_c2, &..} and the track set t= { T1, T2, &..tm } exists currently, and if the intersection ratio of the cluster identification cluster_id set a= { id_c1, id_c2 }, incorporating the current third millimeter wave radar tracking result into a set to be clustered c= { A };

2. cluster merge operation

After comparison operation is carried out on all third millimeter wave radar tracking results, if the elements in the last set C to be clustered are equal to two, the Euclidean distance of the center point is adopted, the distances between the two third millimeter wave radar tracking results in the set are compared, and if the distances exceed the corresponding threshold values, clustering is not carried out;

If the elements in the final to-be-clustered set C are less than two, exiting the current processing flow of the second clustering processing based on the time sequence;

if the elements in the last set C to be clustered are more than two, combining the tracking results of each third millimeter wave radar in the last set C to be clustered to form a new millimeter wave radar tracking result.

It will be appreciated that a specific set of implementations is provided herein for the timing-based second aggregation process referred to herein.

Wherein the second aggregation process mainly considers the timing information, and the timing information is mainly embodied on the id change of the previous and subsequent frames. As a simple example of a scenario: one bicycle is traveling from a distance at a relatively high speed, the first frame of radar information is { time=15:05:120, id=1, center= (1 m, 10 m),. The second frame of radar information is { time=15:05:220, id=1, center= (1 m, 8 m) }. The two frames of information indicate that the bicycle is driven for 2m in 100ms, if only spatial clustering (first clustering processing) is performed, the two frames are regarded as two different objects, so the time sequence-based second clustering processing is developed herein, and in consideration of that each frame has id information given by a radar tracking module manufacturer, the time sequence-based second clustering processing can be performed by using the information, so that a more accurate clustering effect is obtained.

Step S205, carrying out matching processing between different fourth millimeter wave radar tracking results and a currently existing track set by taking the degree of distancing as a reference to obtain a matching result;

after the second time sequence-based aggregation processing is completed, the matching processing between elements formed by the fourth millimeter wave radar tracking result c= { C1, C2, & gt, cN } and the currently existing track set t= { T1, T2, & gt, tM } can be performed according to the distance degree between the fourth millimeter wave radar tracking result c= { C1, C2, & gt, and tM } so as to provide a final data basis for the subsequent updating operation of the track set.

The distance degree can also be called cost, and the smaller the distance degree is, the closer the cluster and the track are, and the minimum value is 0.

As yet another exemplary implementation manner, taking the distance degree as a reference, performing a matching process between different fourth millimeter wave radar tracking results and a currently existing track set to obtain a matching result may specifically include:

1. calculating different fourth millimeter wave radar tracking results c= { C1, C2,..:

Similar to the id similarity calculation above, comparing the cluster identifier of the current fourth millimeter wave radar tracking result with the current track set, and if the intersection ratio of the cluster identifier and the current track set is higher than the corresponding threshold, then Cost (i, j) =0;

if the distance similarity and the speed similarity are lower than the corresponding threshold values, calculating the distance similarity and the speed similarity of the two, and carrying out weighted average on the distance similarity and the speed similarity by the following formula to obtain a final Cost (i, j):wherein->Is weight(s)>Representing distance similarity, ++>Representing speed similarity;

the distance similarity may be a distance from a Center of the tracking object obj after cluster merging (including a longitudinal distance longitude_dist and a transverse distance transverse_dist) to an circumscribing polygon of a track object track obj (the latest updated obj in track) in the track set currently existing, where the distance similarity is 0 if the Center of the tracking object obj after cluster merging is inside the circumscribing polygon of the track object track obj in the track set currently existing, and is a distance from the Center to a nearest edge of the polygon if the Center of the tracking object obj after cluster merging is outside the circumscribing polygon of the track object track obj in the track set currently existing; the speed similarity takes the magnitude of the difference between the speed of the trace object obj after cluster merging and the speed of the trace object track obj in the currently existing trace set.

For the more popular point set here, reference may be made to the following:

the original radar tracking module is set to detect a set { obj1, obj2, & gt, obj N }, which indicates that the frame has N objects;

generating a plurality of clustered clusters after clustering, each cluster also being an object set { obj1, obj2,.. Sub.obj K }, indicating that the cluster has K objects;

each cluster is merged into a single object obj, and various attributes of obj are calculated: center, polygon, etc.,

these individual objects obj form a trajectory in time sequence, which is also denoted as the set of obj T { obj1, obj2,.. Sub.m., the set is denoted M frames, the object of the closest frame of the set being obj.

In a certain frame, there are usually multiple clusters and multiple track sets, and the calculation of distance similarity and speed similarity refers to the calculation of obj after a single cluster is combined and the calculation of obj m of the last frame in a single track.

2. Based on the Cost matrix Cost, matching different fourth millimeter wave radar tracking results c= { C1, C2,..cn } and a currently existing track set t= { T1, T2,..tm } to obtain matched clusters and tracks { (C1, T1), (C2, T2),.. } and unmatched track sets { T1, T2,.} and unmatched cluster sets { C1, C2,..}, wherein the matched clusters and tracks { (C1, T1), (C2, T2),...

It will be appreciated that other types of matching algorithms may be employed in particular operations other than the hungarian algorithm.

Meanwhile, as can be seen from the present disclosure, the three types of matching results herein are three updating operations directly corresponding to the subsequent track set, and the updating operation of the track set can directly perform the corresponding updating operation according to the matching results.

Step S206, updating the track set existing currently based on the matching result to obtain a fifth millimeter wave radar tracking result as output.

After the matching processing in step S205 is performed to obtain the matching result between the different fourth millimeter wave radar tracking results and the different tracks, the existing track in the currently existing track set may be updated or deleted based on the matching result, or the track may be newly added in the currently existing track set, so as to achieve the objective of optimizing the millimeter wave tracking result.

Among them, there are:

1. the track updating operation updates the track object track obj corresponding to the matched track set which exists currently by utilizing the track object obj after cluster merging processing, and filtering smoothing can be carried out on centers, width and height of the obj in addition to the attribute of the copied obj in the updating content, wherein the filtering smoothing can be realized by adopting a filtering scheme such as Kalman filtering or a low-pass filter, and the like, and can be adjusted according to actual needs;

2. The track creation operation creates a new track object track obj by using the track object obj after cluster merging processing on unmatched tracks, and endows a self-increasing track_id;

3. track deletion operation is performed by using an unmatched track, and the track is deleted if the current track is not updated for a long time according to the tracking time.

After the fifth millimeter wave radar tracking result is obtained, the fifth millimeter wave radar tracking result may be directly output, or related data processing may be continuously performed, for example, the above-mentioned multi-sensor fusion processing including image sensing may be further performed to obtain a final obstacle recognition result, or corresponding autonomous avoidance processing, reminding processing, etc. may be performed, and the subsequent data processing of the fifth millimeter wave tracking result may obviously refer to the prior art, and therefore, it is considered that the scheme is not important in the present application, and therefore description will not be further expanded.

For the scheme content, in general, after the first millimeter wave radar tracking result output by the radar tracking module is obtained, the first clustering processing is performed on the first millimeter wave radar tracking result to obtain a second millimeter wave radar tracking result, then, based on different clustering clusters related to the second millimeter wave radar tracking result, the clustering merging processing is performed on the second millimeter wave radar tracking result to obtain different third millimeter wave radar tracking results corresponding to different clustering clusters, then, sequential second clustering processing is performed on the different third millimeter wave radar tracking results with the current track set to obtain different fourth millimeter wave radar tracking results, then, the distance degree is used as a reference, the matching processing is performed on the different fourth millimeter wave radar tracking results with the current track set to obtain a matching result, at the moment, the current track set is updated based on the matching result to obtain a fifth millimeter wave radar tracking result, and the fifth millimeter wave radar tracking result is used as output.

The millimeter wave radar tracking result processing method is introduced, and in order to facilitate better implementation of the millimeter wave radar tracking result processing method, the millimeter wave radar tracking result processing device is further provided from the angle of a functional module.

Referring to fig. 4, fig. 4 is a schematic structural diagram of a processing device for millimeter wave radar tracking results in the present application, where the processing device 400 for millimeter wave radar tracking results may specifically include the following structure:

an obtaining unit 401, configured to obtain a first millimeter wave radar tracking result output by a radar tracking module, where the radar tracking module is configured on a vehicle, and the first millimeter wave radar tracking result is obtained by the radar tracking module executing a task of sensing an obstacle around the vehicle in a working process of the vehicle;

a first clustering unit 402, configured to perform a first clustering process on the first millimeter wave radar tracking result to obtain a second millimeter wave radar tracking result;

the cluster merging unit 403 is configured to perform cluster merging processing on the second millimeter wave radar tracking result based on different clusters related to the second millimeter wave radar tracking result, so as to obtain different third millimeter wave radar tracking results corresponding to different clusters;

A second aggregation unit 404, configured to sequentially perform a second aggregation process based on time sequence on different third millimeter wave radar tracking results with single frame properties and a track set existing at present, so as to obtain different fourth millimeter wave radar tracking results;

a matching unit 405, configured to perform matching processing between different fourth millimeter wave radar tracking results and a currently existing track set with the degree of distance as a reference, so as to obtain a matching result;

and the updating unit 406 is configured to update the currently existing track set based on the matching result, and obtain a fifth millimeter wave radar tracking result as output.

In an exemplary implementation, the first clustering unit 402 is specifically configured to:

and performing first clustering processing on the first millimeter wave radar tracking result through a clustering algorithm DBSCAN with noise and based on density to obtain a second millimeter wave radar tracking result.

In yet another exemplary implementation, the first clustering unit 402 is specifically configured to:

，

In still another exemplary implementation manner, the first millimeter wave radar tracking result specifically includes a tracking identifier id, a longitudinal distance_dist, a transverse distance_dist, a longitudinal speed longitude_vel, a transverse speed longitude_vel, a width, a height, a length, and an external polygon, and the second millimeter wave radar tracking result is based on the first millimeter wave radar tracking result, and a cluster identifier cluster_id is added to indicate a cluster to which the object belongs.

In still another exemplary implementation, in the second millimeter wave radar tracking result, cluster merging processing performed on the millimeter wave radar tracking result of each cluster includes the following processing contents:

combining the objects { obj1, obj2, & gt, obj K } of the millimeter wave radar tracking results belonging to the same cluster into a new object obj, averaging the longitudinal speed longitude_vel and the transverse speed transverse_vel of the millimeter wave radar tracking results belonging to the same cluster, wherein the longitudinal distance longitude_dist, the transverse distance longitude_dist, the width height and the length are determined by the selected scheme in the following four external polygon updating schemes:

taking the circumscribed polygon of all objects { obj1, obj2,.. The circumscribed polygon of obj k } as a unit, taking the smallest circumscribed rectangle, wherein the direction is the vertical direction, the angle=90 degrees, and the width and the length are the width and the length of the smallest circumscribed rectangle;

In yet another exemplary implementation, the timing-based second aggregation process includes the following:

cluster identification cluster for the current third millimeter wave radar tracking result, cluster_id set a= { id_c1, id_c2,.} and currently existing track set t= { T1, T2, comparing tM …, if the intersection ratio of the two is higher than the corresponding threshold, incorporating the current third millimeter wave radar tracking result into the set to be clustered c= { a };

In yet another exemplary implementation, the matching unit 405 is specifically configured to:

calculating different fourth millimeter wave radar tracking results c= { C1, C2,..:

comparing the cluster identification of the current fourth millimeter wave radar tracking result with the current track set, and if the intersection ratio of the cluster identification and the current track set is higher than the corresponding threshold value, performing Cost (i, j) =0;

if the distance similarity and the speed similarity are lower than the corresponding threshold values, calculating the distance similarity and the speed similarity of the two, and carrying out weighted average on the distance similarity and the speed similarity by the following formula to obtain a final Cost (i, j): Wherein->Is weight(s)>Representing distance similarity, ++>Representing speed similarity;

based on the Cost matrix Cost, matching different fourth millimeter wave radar tracking results c= { C1, C2,..cn } and a currently existing track set t= { T1, T2,..tm } to obtain matched clusters and tracks { (C1, T1), (C2, T2),.. } and unmatched track sets { T1, T2,.} and unmatched cluster sets { C1, C2,..}, wherein the matched clusters and tracks { (C1, T1), (C2, T2),...

The present application further provides a processing device from the perspective of a hardware structure, referring to fig. 5, fig. 5 shows a schematic structural diagram of the processing device of the present application, specifically, the processing device of the present application may include a processor 501, a memory 502, and an input/output device 503, where the processor 501 is configured to implement steps of a processing method of a millimeter wave radar tracking result in the corresponding embodiment of fig. 2 when executing a computer program stored in the memory 502; alternatively, the processor 501 is configured to implement functions of each unit in the corresponding embodiment of fig. 4 when executing the computer program stored in the memory 502, and the memory 502 is configured to store the computer program required for the processor 501 to execute the processing method of the millimeter wave radar tracking result in the corresponding embodiment of fig. 2.

By way of example, a computer program may be partitioned into one or more modules/units that are stored in the memory 502 and executed by the processor 501 to complete the present application. One or more of the modules/units may be a series of computer program instruction segments capable of performing particular functions to describe the execution of the computer program in a computer device.

The processing devices may include, but are not limited to, a processor 501, memory 502, and input-output devices 503. It will be appreciated by those skilled in the art that the illustrations are merely examples of processing devices, and are not limiting of processing devices, and may include more or fewer components than shown, or may combine some components, or different components, e.g., processing devices may also include network access devices, buses, etc., through which processor 501, memory 502, input output device 503, etc., are connected.

The processor 501 may be a central processing unit (Central Processing Unit, CPU), but may also be other general purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), field programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like, which is a control center for a processing device, with various interfaces and lines connecting the various parts of the overall device.

The memory 502 may be used to store computer programs and/or modules, and the processor 501 may implement various functions of the computer device by executing or executing the computer programs and/or modules stored in the memory 502, and invoking data stored in the memory 502. The memory 502 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, application programs required for at least one function, and the like; the storage data area may store data created according to the use of the processing device, or the like. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as a hard disk, memory, plug-in hard disk, smart Media Card (SMC), secure Digital (SD) Card, flash Card (Flash Card), at least one disk storage device, flash memory device, or other volatile solid-state storage device.

The processor 501, when configured to execute a computer program stored in the memory 502, may specifically implement the following functions:

It will be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working process of the processing device, the processing apparatus and the corresponding units of the millimeter wave radar tracking result described above may refer to the description of the processing method of the millimeter wave radar tracking result in the corresponding embodiment of fig. 2, which is not repeated herein.

Those of ordinary skill in the art will appreciate that all or a portion of the steps of the various methods of the above embodiments may be performed by instructions, or by instructions controlling associated hardware, which may be stored in a computer-readable storage medium and loaded and executed by a processor.

For this reason, the present application provides a computer readable storage medium, in which a plurality of instructions capable of being loaded by a processor are stored, so as to execute the steps of the processing method of the millimeter wave radar tracking result in the corresponding embodiment of fig. 2, and the specific operation may refer to the description of the processing method of the millimeter wave radar tracking result in the corresponding embodiment of fig. 2, which is not repeated herein.

Wherein the computer-readable storage medium may comprise: read Only Memory (ROM), random access Memory (Random Access Memory, RAM), magnetic or optical disk, and the like.

Because the instructions stored in the computer readable storage medium may execute the steps of the method for processing the millimeter wave radar tracking result in the corresponding embodiment of fig. 2, the beneficial effects that can be achieved by the method for processing the millimeter wave radar tracking result in the corresponding embodiment of fig. 2 are achieved, which are detailed in the foregoing description and are not repeated herein.

The above processing method, apparatus, processing device and computer readable storage medium for millimeter wave radar tracking result provided in the present application are described in detail, and specific examples are applied herein to illustrate the principles and embodiments of the present application, where the above description of the examples is only for helping to understand the method and core ideas of the present application; meanwhile, those skilled in the art will have variations in the specific embodiments and application scope in light of the ideas of the present application, and the present description should not be construed as limiting the present application in view of the above.

Claims

1. A method for processing millimeter wave radar tracking results, the method comprising:

based on different clustering clusters related to the second millimeter wave radar tracking result, carrying out cluster merging processing on the second millimeter wave radar tracking result to obtain different third millimeter wave radar tracking results corresponding to different clustering clusters;

Sequentially carrying out second aggregation processing based on time sequence on the different third millimeter wave radar tracking results with single frame properties and the track set existing at present to obtain different fourth millimeter wave radar tracking results;

taking the degree of distancing as a reference, carrying out matching processing between the different fourth millimeter wave radar tracking results and the current track set to obtain a matching result;

and updating the current track set based on the matching result to obtain a fifth millimeter wave radar tracking result as output.

2. The method of claim 1, wherein the performing a first clustering process on the first millimeter wave radar tracking result to obtain a second millimeter wave radar tracking result comprises:

and performing first clustering processing on the first millimeter wave radar tracking result through a clustering algorithm DBSCAN with noise and based on density to obtain the second millimeter wave radar tracking result.

3. The method of claim 1, wherein the performing a first clustering process on the first millimeter wave radar tracking result to obtain a second millimeter wave radar tracking result comprises:

Performing first clustering processing on the first millimeter wave radar tracking result to obtain the second millimeter wave radar tracking result, and in the clustering process, calculating the measurement distance between clusters by the following formula:

，

4. The method of claim 1, wherein the first millimeter wave radar tracking result specifically includes a tracking identifier id, a longitudinal distance_dist, a transverse distance_dist, a longitudinal speed_vel, a transverse speed_vel, a width, a height, a length, and an external polygon, and the second millimeter wave radar tracking result adds a cluster identifier cluster_id to indicate a cluster to which the object belongs based on the first millimeter wave radar tracking result.

5. The method according to claim 4, wherein the cluster merging processing performed on the millimeter wave radar tracking result of each cluster in the second millimeter wave radar tracking result includes the following processing contents:

merging objects { obj1, obj2, & gt, obj k } of millimeter wave radar tracking results belonging to the same cluster into a new object obj, averaging the longitudinal velocity longitude_vel and the transverse velocity transverse_vel of the millimeter wave radar tracking results belonging to the same cluster, wherein the longitudinal distance longitude_dist, the transverse distance longitude_dist, the width, the height and the length are determined by a scheme selected from the following four external polygonal updating schemes:

taking all objects { obj1, obj2,.,. Obj k } to the center of the x-direction and y-direction, the width, the height, and the length are unchanged;

taking the circumscribed polygon of all objects { obj1, obj2,.. The circumscribed polygon of obj k } as a unit, taking a minimum circumscribed rectangle, wherein the direction is a vertical direction, the angle=90 degrees, and the width and the length are the width and the length of the minimum circumscribed rectangle;

taking the circumscribed polygon of all objects { obj1, obj2,.. The circumscribed polygon of obj k } as a unit, taking a minimum rotation circumscribed rectangle, wherein the direction is the direction of the rotation rectangle, and the width and the length are the width and the length of the minimum rotation circumscribed rectangle.

6. The method of claim 5, wherein the timing-based second aggregation process comprises the following:

comparing a cluster identifier cluster_id set a= { id_c1, id_c2, & gt } of the current third millimeter wave radar tracking result with the current track set t= { T1, T2, & gt, tM }, and if the intersection ratio of the cluster identifier cluster_id set a= { id_c1, id_c2, & gt and the current track set t= { T1, T2, & gt, tM }, incorporating the current third millimeter wave radar tracking result into a set to be clustered c= { a };

After comparing all the third millimeter wave radar tracking results, if the elements in the last set C to be clustered are equal to two, adopting Euclidean distance of a center point to compare the distance between the two third millimeter wave radar tracking results in the set, and if the distance exceeds a corresponding threshold value, not clustering;

if the elements in the last to-be-clustered set C are less than two, exiting the current processing flow of the second clustering processing based on the time sequence;

if the elements in the last set C to be clustered are more than two, combining the third millimeter wave radar tracking results in the last set C to be clustered to form a new millimeter wave radar tracking result.

7. The method according to claim 5, wherein the matching processing between the different fourth millimeter wave radar tracking results and the currently existing track set with the distance degree as a reference is performed to obtain a matching result, which includes:

calculating the different fourth millimeter wave radar tracking results c= { C1, C2..the cN } and the currently existing track set t= { T1, T2..the tM } Cost matrix Cost, wherein the matrix size of the Cost matrix Cost is N x M, cost (i, j) represents the distance degree between the ith cluster and the jth track, and the calculating of the Cost matrix Cost comprises the following processing contents:

Comparing the cluster identification of the current fourth millimeter wave radar tracking result with the current track set, and if the intersection ratio of the cluster identification and the current track set is higher than a corresponding threshold value, performing Cost (i, j) =0;

if the distance similarity and the speed similarity are lower than the corresponding threshold values, calculating the distance similarity and the speed similarity of the two, and carrying out weighted average on the distance similarity and the speed similarity by the following formula to obtain a final Cost (i, j):wherein->Is weight(s)>Representing the distance similarity->Representing the speed similarity;

based on the Cost matrix Cost, matching the different fourth millimeter wave radar tracking results c= { C1, C2,..cn } and the currently existing track set t= { T1, T2,..tm } with a hungarian algorithm to obtain matched clusters and tracks { (C1, T1), (C2, T2),...

8. A processing apparatus for millimeter wave radar tracking results, the apparatus comprising:

the second aggregation unit is used for sequentially carrying out second aggregation processing based on time sequence on the different third millimeter wave radar tracking results with single frame properties and the track set existing at present to obtain different fourth millimeter wave radar tracking results;

the matching unit is used for carrying out matching processing between the different fourth millimeter wave radar tracking results and the current track set by taking the degree of distancing as a reference to obtain a matching result;

And the updating unit is used for updating the track set existing at present based on the matching result to obtain a fifth millimeter wave radar tracking result as output.

9. A processing device comprising a processor and a memory, the memory having stored therein a computer program, the processor executing the method of any of claims 1 to 7 when invoking the computer program in the memory.

10. A computer readable storage medium storing a plurality of instructions adapted to be loaded by a processor to perform the method of any one of claims 1 to 7.