1. A lane line identification method based on radar is characterized by comprising the following steps:

determining a plurality of independent identification intervals according to the maximum detection distance of the radar;

screening the effective flight path in each identification interval according to a preset rule;

performing straight line fitting on the effective tracks, and obtaining a set of starting points in each identification interval through the straight line fitted to each effective track in each identification interval;

clustering the set of the starting points of each identification interval to obtain a set of lane center points of each identification interval;

and performing data fitting on the set of lane central points of each identification interval to obtain a lane central line of the target road.

2. The radar-based lane line identification method of claim 1, wherein the step of determining a plurality of independent identification intervals according to the maximum detection range of the radar comprises:

measuring a maximum detection range of the radarCarrying out average division to obtain m independent identification intervals, wherein the length of each identification interval is as follows:

；

wherein, Y_iRepresents the length of the ith identification interval, i is less than or equal to m.

3. The radar-based lane line recognition method according to claim 2, wherein the step of filtering the valid tracks in each recognition interval according to a preset rule comprises:

for each identification interval, determining data of the same target vehicle to form a track of the target vehicle in the identification interval;

calculating the length L of the track of each target vehicle in each identification interval and the standard deviation X relative to the transverse direction of the radar_stdCalculating the number N of points contained in each flight path in each identification interval;

determining each track meeting the following conditions as the effective track:、andthe three are established at the same time;

wherein the content of the first and second substances,is a preset reference length and is used as a reference length,is a preset reference width, and is,is the number of the preset reference points.

4. The radar-based lane line identification method according to claim 3, wherein for each identification interval, the length L of the track of each target vehicle within the identification interval is calculated according to the following formula:

；

wherein the content of the first and second substances,a distance value representing the latest track point of the ith track in the recognition interval with respect to the transverse direction of the radar,a distance value representing the latest track point of the ith track in the identification section relative to the longitudinal direction of the radar,a distance value representing a point at which the ith track first enters the identified section with respect to a lateral direction of the radar,and a distance value representing a point at which the ith track first enters the identification interval with respect to the longitudinal direction of the radar.

5. The radar-based lane line identification method according to claim 3, wherein for each identification section, the standard deviation X of the track of each target vehicle within the identification section with respect to the lateral direction of the radar is calculated according to the following formula_std：

；

Wherein N represents the number of the points in the ith track in the identification interval,denotes a distance value of a k-th point of the ith track with respect to a lateral direction of the radar, μ denotes an average value of distance values of all points in the ith track with respect to the lateral direction of the radar, and k =1, 2.

6. The radar-based lane line identification method of claim 1, wherein said step of fitting a straight line to the valid tracks and obtaining a set of starting points in each identification interval from the straight line fitted to each valid track in the identification interval comprises:

for each effective track, performing straight line fitting on a plurality of data points forming the effective track to obtain the slope a and the intercept b of a fitted straight line, wherein the fitted straight line is represented in the form of the following linear equation:；

based on the linear equation, calculating the starting point of the straight line corresponding to each effective track in each identification interval in the identification intervalCorresponding to X_iWherein, in the step (A),；

aiming at each identification interval, X corresponding to each track in the identification interval_iThe composed data set forms a set of the starting points corresponding to the identification interval;

wherein, for each recognition interval, X_iIndicating the ith trackA distance value of a point when the first time the identification section is entered with respect to a lateral direction of the radar.

7. The radar-based lane line recognition method according to claim 1, wherein the clustering the set of the starting points of each recognition interval to obtain the set of lane center points of each recognition interval comprises:

clustering the set of the starting points corresponding to each identification interval respectively;

judging whether the number K of clusters generated after clustering processing is carried out on the set of each starting point is the same as the number of lanes of the target road and whether the number of the starting points contained in each cluster is larger than a preset threshold value;

for each recognition section, if the number K of the clusters is the same as the number of the lanes of the target road and the number of the starting points contained in each cluster is greater than the preset threshold value, determining that the center of mass of each cluster is the lane center point of the corresponding lane in the recognition section, wherein C_kRepresents the K-th lane center point within the recognition section, K =1,2, … K.

8. The radar-based lane line recognition method of claim 1, wherein the step of performing data fitting on the set of lane center points of each recognition interval to obtain a lane center line of a target road comprises:

all lane center points C in all recognition intervals_kCombining to obtain a two-dimensional matrix, wherein an element C in the two-dimensional matrix_{i_k}Representing the k-th lane central point in the i-th recognition interval;

and respectively performing data fitting on each line of data of the two-dimensional matrix to obtain lane center lines corresponding to all lanes of the target road, wherein a curve obtained by fitting each line of data corresponds to one lane center line.

9. A radar-based lane line recognition apparatus, comprising:

the identification interval generation module is used for determining a plurality of independent identification intervals according to the maximum detection distance of the radar;

the flight path screening module is used for screening the effective flight path in each identification interval according to a preset rule;

the straight line fitting module is used for performing straight line fitting on the effective tracks and obtaining a set of starting points in each identification interval through the straight line fitted to each effective track in each identification interval;

the cluster processing module is used for carrying out cluster processing on the set of the starting points of each identification interval to obtain a set of lane central points of each identification interval;

and the lane line identification module is used for performing data fitting on the set of lane central points of each identification interval to obtain a lane central line of the target road.

10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the lane line identification method according to any one of claims 1 to 8 when executing the program.

11. A non-transitory computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, implements the steps of the lane line identification method according to any one of claims 1 to 8.

Background

Millimeter wave radars are radars that operate in the millimeter wave band (millimeter wave) for detection. The millimeter wave radar is the most basic component in an intelligent traffic system, and can accurately and real-timely acquire road traffic states, wherein the road traffic states generally comprise traffic flow, speed, occupancy, queuing length and the like. If the traffic state of each lane needs to be accurately counted, lane information of the region of interest needs to be configured in advance as prior information in a millimeter wave radar, and the millimeter wave radar combines target information detected by the millimeter wave radar and preset lane information to output the current traffic state after counting.

In the prior art, the coordinate information of each lane is generally obtained by adopting a manual field measurement mode, and the manual field measurement mode must be carried out under the condition of traffic interruption in consideration of factors such as safety and the like. In addition, if the on-site road has a certain curvature, it is difficult to accurately measure the actual lane information by manual on-site measurement.

Disclosure of Invention

The invention provides a lane line identification method and device based on radar, electronic equipment and a storage medium, which are used for solving the problems caused by manual field measurement in the prior art and realizing accurate completion of multi-lane line identification without manual operation.

The invention provides a lane line identification method based on radar, which comprises the following steps:

determining a plurality of independent identification intervals according to the maximum detection distance of the radar;

screening the effective flight path in each identification interval according to a preset rule;

performing straight line fitting on the effective tracks, and obtaining a set of starting points in each identification interval through the straight line fitted to each effective track in each identification interval;

clustering the set of the starting points of each identification interval to obtain a set of lane center points of each identification interval;

and performing data fitting on the set of lane central points of each identification interval to obtain a lane central line of the target road.

According to the radar-based lane line identification method of the present invention, the step of determining a plurality of independent identification sections according to the maximum detection distance of the radar includes:

measuring a maximum detection range of the radarCarrying out average division to obtain m independent identification intervals, wherein the length of each identification interval is as follows:

；

wherein, Y_iRepresents the length of the ith identification interval, i is less than or equal to m.

According to the radar-based lane line identification method, the step of screening the effective track in each identification interval according to a preset rule comprises the following steps:

for each identification interval, determining data of the same target vehicle to form a track of the target vehicle in the identification interval;

calculating the length L of the track of each target vehicle in each identification interval and the standard deviation X relative to the transverse direction of the radar_stdCalculating the number N of points contained in each flight path in each identification interval;

determining each track meeting the following conditions as the effective track:、andthe three are established at the same time;

wherein the content of the first and second substances,is a preset reference length and is used as a reference length,is a preset reference width, and is,is the number of the preset reference points.

According to the radar-based lane line identification method, aiming at each identification interval, the length L of the track of each target vehicle in the identification interval is calculated according to the following formula:

；

wherein the content of the first and second substances,a distance value representing the latest track point of the ith track in the recognition interval with respect to the transverse direction of the radar,a distance value representing the latest track point of the ith track in the identification section relative to the longitudinal direction of the radar,a distance value representing a point at which the ith track first enters the identified section with respect to a lateral direction of the radar,and a distance value representing a point at which the ith track first enters the identification interval with respect to the longitudinal direction of the radar.

According to the radar-based lane line identification method, aiming at each identification interval, the standard deviation X of the track of each target vehicle in the identification interval relative to the transverse direction of the radar is calculated according to the following formula_std：

；

Wherein N represents the number of the points in the ith track in the identification interval,denotes a distance value of a k-th point of the ith track with respect to a lateral direction of the radar, μ denotes an average value of distance values of all points in the ith track with respect to the lateral direction of the radar, and k =1, 2.

According to the radar-based lane line identification method, the step of fitting the effective tracks with straight lines and obtaining a set of starting points in each identification interval through the straight lines fitted to each effective track in each identification interval comprises the following steps:

for each effective track, performing straight line fitting on data points forming the effective track to obtain the slope a and the intercept b of a fitted straight line, wherein the fitted straight line is represented in the form of the following linear equation:；

based on the linear equation, calculating the starting point of the straight line corresponding to each effective track in each identification interval in the identification intervalCorresponding to X_iWherein, in the step (A),；

aiming at each identification interval, X corresponding to each track in the identification interval_iNumber of compositionsForming a set of the starting points corresponding to the identification interval according to the set;

wherein, for each recognition interval, X_iAnd a distance value representing a point at which the ith track first enters the identification interval with respect to the lateral direction of the radar.

According to the radar-based lane line identification method, the step of clustering the set of the starting points of each identification section to obtain the set of lane center points of each identification section comprises the following steps:

clustering the set of the starting points corresponding to each identification interval respectively;

judging whether the number K of clusters generated after clustering processing is carried out on the set of each starting point is the same as the number of lanes of the target road and whether the number of the starting points contained in each cluster is larger than a preset threshold value;

for each recognition section, if the number K of the clusters is the same as the number of the lanes of the target road and the number of the starting points contained in each cluster is greater than the preset threshold value, determining that the center of mass of each cluster is the lane center point of the corresponding lane in the recognition section, wherein C_kRepresents the K-th lane center point within the recognition section, K =1,2, … K.

According to the radar-based lane line identification method of the present invention, the step of performing data fitting on the set of lane center points of each identification section to obtain the lane center line of the target road includes:

all lane center points C in all recognition intervals_kCombining to obtain a two-dimensional matrix, wherein an element C in the two-dimensional matrix_{i_k}Representing the k-th lane central point in the i-th recognition interval;

and respectively performing data fitting on each line of data of the two-dimensional matrix to obtain lane center lines corresponding to all lanes of the target road, wherein a curve obtained by fitting each line of data corresponds to one lane center line.

The invention also provides a lane line recognition device based on radar, which comprises:

the identification interval generation module is used for determining a plurality of independent identification intervals according to the maximum detection distance of the radar;

the flight path screening module is used for screening the effective flight path in each identification interval according to a preset rule;

the straight line fitting module is used for performing straight line fitting on the effective tracks and obtaining a set of starting points in each identification interval through the straight line fitted to each effective track in each identification interval;

the cluster processing module is used for carrying out cluster processing on the set of the starting points of each identification interval to obtain a set of lane central points of each identification interval;

and the lane line identification module is used for performing data fitting on the set of lane central points of each identification interval to obtain a lane central line of the target road.

The invention also provides an electronic device, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the program to realize the steps of the lane line identification method.

The invention also provides a non-transitory computer-readable storage medium having stored thereon a computer program which, when executed by a processor, carries out the steps of the lane line identification method according to any one of the preceding claims.

According to the radar-based lane line identification method, device, electronic equipment and storage medium, a plurality of independent identification intervals are generated through the maximum detection distance based on the radar, and the lane center points of each identification interval are identified and then subjected to data fitting, so that complete multi-lane lines are automatically identified. The invention can automatically and accurately identify the multiple lane lines without obstructing the traffic, and does not need manual operation.

Drawings

In order to more clearly illustrate the technical solutions of the present invention or the prior art, the drawings needed for the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.

FIG. 1 is a schematic flow chart of a radar-based lane line identification method provided by the present invention;

FIG. 2 is a schematic diagram of the identification space partitioning of the present invention;

FIG. 3 is a schematic flow chart of selecting valid target data according to the present invention;

FIG. 4 is a schematic flow chart of identifying a set of track data provided by the present invention;

FIG. 5 is a schematic flow chart of a clustering process provided by the present invention;

FIG. 6 is a schematic flow chart of the method for identifying multiple lane lines provided by the present invention;

FIG. 7 is a schematic view of a multiple lane line provided by the present invention;

fig. 8 is a schematic structural diagram of a radar-based lane line recognition apparatus provided in the present invention;

fig. 9 is a schematic structural diagram of an electronic device provided by the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is obvious that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

The terms "first," "second," and the like in the description and in the claims, and in the drawings described above, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It will be appreciated that the data so used may be interchanged under appropriate circumstances such that the embodiments described herein may be practiced otherwise than as specifically illustrated or described herein.

The technical terms to which the present invention relates are described below:

the millimeter wave radar is a radar whose working frequency band is in the millimeter wave band, and the principle of ranging is the same as that of a general radar, that is, radio waves (radar waves) are transmitted out, then echoes are received, and position data of a target is measured according to a time difference between receiving and transmitting, and the millimeter wave radar is radio waves whose frequency is in the millimeter wave band.

The millimeter wave radar has a narrower wave speed (generally, milliradian magnitude), so that the angle resolution and angle measurement accuracy of the radar can be improved, and the millimeter wave radar is favorable for resisting electronic interference, clutter interference, multipath reflection interference and the like. And because of its high operating frequency, can get large signal bandwidth (such as gigahertz magnitude) and Doppler's frequency shift, help to improve the measurement accuracy and resolving power of the distance and speed and can analyze the target characteristic. Furthermore, the millimeter wave radar can be applied to the fields of airplanes, satellites, intelligent traffic systems and the like due to the small antenna aperture, the small element size and the small device size.

The invention provides a lane line identification method, a device, electronic equipment and a storage medium based on radar, aiming at solving the problem that the actual lane information is difficult to be accurately measured by adopting the manual field measurement of the lane coordinate information in the prior art, and can automatically and accurately identify a plurality of vehicle conductors by processing the data output by the millimeter wave radar,

the radar-based lane line recognition method, apparatus, electronic device, and storage medium of the present invention are described below with reference to fig. 1 to 9.

Fig. 1 is a schematic flow chart of a method for identifying a lane line based on radar according to the present invention, as shown in fig. 1. The invention discloses a radar-based lane line identification method, which comprises the following steps:

step 101, determining a plurality of independent identification intervals according to the maximum detection distance of the radar.

Alternatively, the radar may be a millimeter wave radar, and may also be data of other radars, but the invention is not limited thereto.

The radar detects the distance by emitting microwave pulses, the electromagnetic wave pulses emitted by the radar are reflected by an object in the propagation process, the radar measures the time of the electromagnetic wave passing back and forth after receiving the reflected waves, and the distance from the radar to the object can be determined according to the propagation speed of the electromagnetic wave.

And 102, screening the effective track in each identification interval according to a preset rule.

The effective track data refers to data which is beneficial to lane line identification, most automobiles run according to straight lines in a short interval range, and target tracks detected by the radar are basically parallel to the lane lines, and the data can be used for lane line identification.

And when the lane change exists in the identification section of part of automobiles, the standard deviation of the X direction of the track of target data detected by the radar is increased, the track is crossed with the lane line, and the data can influence the result of the lane line identification. In addition, the radar may have problems of false alarm, track interruption and the like in the target detection process, false target data or data in a part of identification intervals after the track interruption are not long enough can affect the result of lane line identification, and the data needs to be removed in the identification process.

Step 103, performing straight line fitting on the effective tracks, and obtaining a set of starting points in each identification interval through the straight line fitted to each effective track in each identification interval.

Optionally, the set includes the number of track points corresponding to each identification block section, and each track is composed of track points of the same target.

And 104, clustering the set of the starting points of each identification interval to obtain a set of lane central points of each identification interval.

The purpose of the clustering processing is to obtain the center point of each recognition interval, namely the lane center point, and the lane center points of all recognition intervals are called as a lane center point set.

And 105, performing data fitting on the set of lane central points of each identification interval to obtain a lane central line of the target road.

The above steps 101 to 105 will be described in detail below.

FIG. 2 is a schematic diagram of the identification space division of the present invention, as shown in FIG. 2. In the foregoing step 101, the step of determining a plurality of independent identification intervals according to the maximum detection distance of the radar includes:

the maximum detection distance of the radar isPerforming an average division, i.e. maximum detection range of the radarAveragely dividing the identification interval into m independent identification intervals, and obtaining the length of each identification interval as follows:

；

wherein, Y_iRepresents the length of the ith identification interval, i is less than or equal to m.

The purpose of dividing the maximum detection range of the radar into m identification intervals is to enable a large proportion of target vehicles to travel in a straight line within the subdivided identification interval. And in each identification interval, data of the target vehicle driving track detected by the radar is contained.

Wherein, m in the m divided identification intervals can be set according to actual conditions, and the invention is not limited to the specific value of m.

The data detected by the radar is output according to a period, for example, the period is 100ms, and the output time is 0, 100ms, 200ms, 300ms, … …. Where the output time is 100ms and 200ms, 200ms and 300ms, etc. are considered as outputs of adjacent cycles.

And when the track of the target vehicle is established, the radar system assigns an ID number as an identifier, and the track ID of a target vehicle is consistent from the detection of the target vehicle by the radar to the disappearance of the target vehicle (namely, the target vehicle cannot be detected by the radar). The radar then outputs the track (i.e., motion trajectory) of the target vehicle during its travel.

Fig. 3 is a schematic flow chart of selecting valid target data according to the present invention, as shown in fig. 3. In the step 102, the step of screening the effective track in each identification interval according to a preset rule includes:

step 301, determining data of the same target vehicle for each identification interval to form a track of the target vehicle in the identification interval.

The data of the same target vehicle refers to track point data of targets periodically output by the radar, and the track point data of the same target vehicle has the same ID number.

Optionally, in the method disclosed by the present invention, valid target data is selected from the target data according to the mathematical criteria of steps 302-303.

Step 302, calculating the length L of the track of each target vehicle in each identification interval and the standard deviation X relative to the transverse direction of the radar_stdAnd calculating the number N of points contained in each flight path in each identification interval.

Optionally, for each recognition interval, calculating the length L of the track of each target vehicle in the recognition interval according to the following formula:

；

wherein the content of the first and second substances,a distance value representing the latest track point of the ith track in the recognition interval with respect to the transverse direction of the radar,a distance value representing the latest track point of the ith track in the identification section relative to the longitudinal direction of the radar,a distance value representing a point at which the ith track first enters the identified section with respect to a lateral direction of the radar,and a distance value representing a point at which the ith track first enters the identification interval with respect to the longitudinal direction of the radar.

Optionally, for each recognition interval, the standard deviation X of the track of each target vehicle in the recognition interval relative to the transverse direction of the radar is calculated according to the following formula_std：

；

Wherein N represents the number of the points in the ith track in the identification interval,denotes a distance value of a k-th point of the ith track with respect to a lateral direction of the radar, μ denotes an average value of distance values of all points in the ith track with respect to the lateral direction of the radar, and k =1, 2.

Step 303, determining each flight path meeting the following conditions as the effective flight path:、andthe three are established at the same time.

Wherein the content of the first and second substances,is a preset reference length and is used as a reference length,is a preset reference width, and is,for a preset ginsengThe number of test points.

For example,it may be set to a length of two-thirds of the recognition interval,may be set to one third of the standard lane width.

And step 304, caching the selected effective track.

FIG. 4 is a flow chart illustrating the process of identifying a track data set according to the present invention, as shown in FIG. 4. In step 103, the step of fitting a straight line to the effective track and obtaining a set of starting points in each identification interval by using the straight line fitted to each effective track in each identification interval includes:

step 401, for each effective track, performing straight line fitting on a plurality of data points constituting the effective track to obtain a slope a and an intercept b of a fitted straight line, where the fitted straight line is represented in the form of a following linear equation:。

alternatively, a least squares algorithm may be used for straight line fitting. The least square method is a mathematical tool widely applied in the fields of various disciplines of data processing such as error estimation, uncertainty, system identification and prediction, forecast and the like. For example, the least squares equation:

the formula of the fitting straight line is:；

wherein, the slope of the fitting straight line is:

；

after calculating the slope, according toAnd the determined slope k, and calculating the intercept b by using a undetermined coefficient method.

Step 402, based on the linear equation, calculating the starting point of the straight line corresponding to each effective track in each identification interval in the identification intervalCorresponding to X_iWherein, in the step (A),。

wherein, for each recognition interval, X_iAnd the transverse distance value of the point of the ith flight path entering the identification interval for the first time relative to the radar is represented.

Step 403, aiming at each identification interval, X corresponding to each track in the identification interval_iThe formed data set forms a set of the starting points corresponding to the identification interval.

Fig. 5 is a schematic flow chart of the clustering process provided by the present invention, as shown in fig. 5. In the step 104, the clustering the set of the starting points of each recognition interval to obtain a set of lane center points of each recognition interval includes:

and step 501, clustering the sets of the starting points corresponding to each identification interval respectively.

Step 502, judging whether the number K of clusters generated after clustering processing is performed on the set of each starting point is the same as the number of lanes of the target road and whether the number of the starting points included in each cluster is greater than a preset threshold value.

Alternatively, the Clustering process may be performed using a DBSCAN (Density-Based Clustering of Applications with Noise) Density Clustering algorithm.

The DBSCAN algorithm is a typical density-based clustering method that defines clusters as the largest set of density-connected points, can divide areas with sufficient density into clusters, and can find clusters of arbitrary shape in noisy spatial data sets.

Step 503, for each recognition interval, if the number K of the clusters is the same as the number of lanes of the target road and the number of starting points included in each cluster is greater than the preset threshold, determining that the centroid of each cluster is the lane center point of the corresponding lane in the recognition interval.

Wherein, C_kRepresents the K-th lane center point within the recognition section, K =1,2, … K.

And if the number K of the clusters is the same as the number of the lanes of the target road and the number of the starting points contained in each cluster is less than or equal to the preset threshold value, continuing to wait for the data of a new target vehicle for judgment.

Optionally, the preset threshold represents the number of valid target tracks in a lane, for example, the preset threshold is 10, the preset threshold may be obtained according to statistical analysis of actual data, and when the data of each cluster is greater than 10, the position of the center point of the lane tends to be stable.

In the same manner as described above, the lane of the next recognition section is recognized.

Fig. 6 is a schematic flow chart of the method for identifying multiple lane lines, as shown in fig. 6. In the step 105, the step of performing data fitting on the set of lane center points of each recognition section to obtain a lane center line of the target road includes:

601, all lane center points C in all recognition intervals_kCombining to obtain a two-dimensional matrix, wherein an element C in the two-dimensional matrix_{i_k}Indicating the k-th lane center point in the i-th recognition interval.

Optionally, the lane identification process for each identification section is as described in steps 501 to 502 above. When the lane identification process of all the identification sections is finished, the lane center point of each identification section is determinedAre combined to obtain a two-dimensional matrix C_{i_k}。

And step 602, performing data fitting on each line of data of the two-dimensional matrix respectively to obtain lane center lines corresponding to all lanes of the target road, wherein a curve obtained by fitting each line of data corresponds to one lane center line.

Optionally, a third-order spline (cubic spline) algorithm may be used for data fitting, where the third-order spline algorithm is implemented by sequentially connecting n data points by using a segmented cubic function, where the cubic function needs to satisfy: at each data point, the junction of each segment is smooth. This can be done by solving equations.

For two-dimensional matrix C_{i_k}The data of each row of the vehicle is fitted by using a third-order spline algorithm to obtain the lane center line of each lane (as shown in fig. 7).

In summary, the present invention solves the problem that it is difficult to accurately measure actual lane information manually in the prior art by dividing a plurality of identification sections, then selecting effective target data of the identification sections, and performing straight line fitting, cluster analysis, data fitting, etc. on the effective target data to finally obtain a complete lane center line of each lane.

In the following description of the radar-based lane line recognition apparatus according to the present invention, the radar-based lane line recognition apparatus described below and the radar-based lane line recognition method described above may be referred to in correspondence with each other.

Fig. 8 is a schematic structural view of a radar-based lane line recognition apparatus according to the present invention, as shown in fig. 8. A radar-based lane line identification device 800 comprises an identification interval generation module 810, a track screening module 820, a straight line fitting module 830, a clustering processing module 840 and a lane line identification module 850. Wherein the content of the first and second substances,

an identification interval generating module 810, configured to determine a plurality of independent identification intervals according to the maximum detection distance of the radar.

And a track screening module 820, configured to screen an effective track in each of the identification intervals according to a preset rule.

And a straight line fitting module 830, configured to perform straight line fitting on the effective tracks, and obtain a set of starting points in each identification interval according to the straight line fitted to each effective track in each identification interval.

The clustering module 840 is configured to perform clustering on the set of starting points of each identification interval to obtain a set of lane center points of each identification interval.

And the lane line recognition module 850 is configured to perform data fitting on the set of lane center points of each recognition interval to obtain a lane center line of the target road.

Optionally, the identification interval generating module 810 is further configured to:

measuring a maximum detection range of the radarCarrying out average division to obtain m independent identification intervals, wherein the length of each identification interval is as follows:

；

wherein, Y_iRepresents the length of the ith identification interval, i is less than or equal to m.

Optionally, the track filtering module 820 is further configured to:

for each identification interval, determining data of the same target vehicle to form a track of the target vehicle in the identification interval;

calculating the length L of the track of each target vehicle in each identification interval and the standard deviation X relative to the transverse direction of the radar_stdCalculating the number N of the point tracks contained in each flight path in each identification interval;

determining each track meeting the following conditions as the effective track:、andthe three are established at the same time;

wherein the content of the first and second substances,is a preset reference length and is used as a reference length,is a preset reference width, and is,is the number of the preset reference points.

Optionally, for each recognition interval, calculating the length L of the track of each target vehicle in the recognition interval according to the following formula:

；

wherein the content of the first and second substances,a distance value representing the latest track point of the ith track in the recognition interval with respect to the transverse direction of the radar,a distance value representing the latest track point of the ith track in the identification section relative to the longitudinal direction of the radar,a distance value representing a point at which the ith track first enters the identified section with respect to a lateral direction of the radar,and a distance value representing a point at which the ith track first enters the identification interval with respect to the longitudinal direction of the radar.

Optionally, for each recognition interval, it is calculated according to the following formulaCalculating the standard deviation X of the track of each target vehicle in the identification interval relative to the transverse direction of the radar_std：

；

Wherein N represents the number of the points in the ith track in the identification interval,denotes a distance value of a k-th point of the ith track with respect to a lateral direction of the radar, μ denotes an average value of distance values of all points in the ith track with respect to the lateral direction of the radar, and k =1, 2.

Optionally, the line fitting module 830 is further configured to:

for each effective track, performing straight line fitting on data points forming the effective track to obtain the slope a and the intercept b of a fitted straight line, wherein the fitted straight line is represented in the form of the following linear equation:；

based on the linear equation, calculating the starting point of the straight line corresponding to each effective track in each identification interval in the identification intervalCorresponding to X_iWherein, in the step (A),；

aiming at each identification interval, X corresponding to each track in the identification interval_iThe composed data set forms a set of the starting points corresponding to the identification interval;

wherein, for each recognition interval, X_iAnd a distance value representing a point at which the ith track first enters the identification interval with respect to the lateral direction of the radar.

Optionally, the cluster processing module 840 is further configured to:

clustering the set of the starting points corresponding to each identification interval respectively;

judging whether the number K of clusters generated after clustering processing is carried out on the set of each starting point is the same as the number of lanes of the target road and whether the number of the starting points contained in each cluster is larger than a preset threshold value;

for each recognition section, if the number K of the clusters is the same as the number of the lanes of the target road and the number of the starting points contained in each cluster is greater than the preset threshold value, determining that the center of mass of each cluster is the lane center point of the corresponding lane in the recognition section, wherein C_kRepresents the K-th lane center point within the recognition section, K =1,2, … K.

Optionally, the lane line identification module 850 is further configured to:

all lane center points C in all recognition intervals_kCombining to obtain a two-dimensional matrix, wherein an element C in the two-dimensional matrix_{i_k}Representing the k-th lane central point in the i-th recognition interval;

and respectively performing data fitting on each line of data of the two-dimensional matrix to obtain lane center lines corresponding to all lanes of the target road, wherein a curve obtained by fitting each line of data corresponds to one lane center line.

Other aspects of the radar-based lane line identification device disclosed by the invention are the same as or similar to the radar-based lane line identification method described above, and are not repeated herein.

Fig. 9 illustrates a physical structure diagram of an electronic device, and as shown in fig. 9, the electronic device may include: a processor (processor)910, a communication Interface (Communications Interface)920, a memory (memory)930, and a communication bus 940, wherein the processor 910, the communication Interface 920, and the memory 930 communicate with each other via the communication bus 940. Processor 910 may invoke logic instructions in memory 930 to perform the radar-based lane line identification method described previously.

Furthermore, the logic instructions in the memory 930 may be implemented in software functional units and stored in a computer readable storage medium when the logic instructions are sold or used as independent products. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.

In yet another aspect, the present invention also provides a non-transitory computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements any of the radar-based lane line identification methods described above.

The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.

Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.

Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.