Radar target tracking method, domain controller and computer program product

Abstract

The invention provides a radar target tracking method, a domain controller and a computer program product, and according to the radar target tracking method provided by the invention, after a radar device collects point cloud data of target point clouds having an association relationship with a historical tracking target, the point cloud data of each target point cloud can be calculated according to the point cloud data of each target point cloud; determining a point cloud coordinate median of the target point cloud in a direction vertical to the historical driving direction of the historical tracking target; according to the difference value between the target point cloud coordinate of the target point cloud in the vertical direction of the historical driving direction and the point cloud coordinate median, determining a target weight value in negative correlation with the difference value for the target point cloud; according to the point cloud data of the target point cloud and the target weight value of the target point cloud, the accuracy of the generated target tracking result for the historical tracking target is improved.

Description

Technical Field

[0001] This invention relates to the field of radar technology, and in particular to a radar target tracking method, a domain controller, and a computer program product. Background Technology

[0002] Currently, due to the superior detection performance of radar equipment, and its excellent performance in environments such as night, rain, and fog, it is able to effectively identify both static and dynamic targets in the surrounding environment. Therefore, people have gradually begun to use radar equipment to detect and track various targets in the surrounding environment. However, in complex environments, the propagation and reflection of radar waves may be interfered with. This interference may lead to inaccurate or biased measurement results from radar equipment, thereby affecting the accuracy of radar target tracking results. Summary of the Invention

[0003] Based on this, the present invention provides a radar target tracking method, a domain controller, and a computer program product. Using this radar target tracking method, the target weight value of each target point cloud can be determined according to the position distribution of the target point cloud that is related to the historical tracked target in the vertical direction of the historical tracked target's historical driving direction. Then, radar target tracking processing can be performed based on the point cloud data and target weight values ​​of the target point cloud, which can improve the accuracy of radar target tracking results.

[0004] On one hand, the present invention provides a radar target tracking method, comprising:

[0005] Acquire point cloud data of targets that are correlated with historically tracked targets, collected by radar equipment;

[0006] Based on the point cloud data of the target point cloud, determine the median point cloud coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historical tracked target;

[0007] The target weight value of the target point cloud is calculated based on the difference between the target point cloud coordinates in the vertical direction of the historical driving direction and the median of the point cloud coordinates; wherein the target weight value is negatively correlated with the difference value.

[0008] Based on the point cloud data of the target point cloud and the target weight value of the target point cloud, a target tracking result is generated for the historical tracking target.

[0009] Furthermore, in some embodiments, the above method further includes:

[0010] Based on the historical speed information of the historical tracked target, the historical driving direction of the historical tracked target relative to the radar device is determined; wherein, the historical driving direction includes either lateral or longitudinal.

[0011] Furthermore, in some embodiments, the above method further includes:

[0012] Determine whether the historical driving direction of the historical tracked target relative to the radar device is lateral, and obtain a first determination result;

[0013] If the first judgment result indicates that the historical driving direction is not lateral, then the azimuth angle of the historical tracking target relative to the radar device is determined based on the historical position information of the historical tracking target.

[0014] Determine whether the absolute value of the azimuth angle is within a preset numerical range to obtain a second determination result; wherein, the starting value of the preset numerical range is greater than 0;

[0015] Determining the median point cloud coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historically tracked target includes:

[0016] If the first judgment result indicates that the historical driving direction is horizontal, or if the second judgment result indicates that the absolute value of the azimuth angle is within a preset value range, then the median point cloud coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historical tracked target are determined.

[0017] Furthermore, in some embodiments, determining the azimuth angle of the historically tracked target relative to the radar device based on the historical position information of the historically tracked target includes:

[0018] If the center point of the radar device is located at the origin of the preset coordinate system, the quotient of the x-coordinate and y-coordinate of the center point of the historical tracked target in the preset coordinate system is used as the independent variable of the arctangent function to calculate the azimuth angle of the historical tracked target relative to the radar device.

[0019] Furthermore, in some embodiments, the above method further includes:

[0020] Based on the point cloud data of each target point cloud, determine the first difference between the maximum and minimum point cloud abscissa of the target point cloud;

[0021] Based on the point cloud data of the target point cloud, determine the second difference between the maximum and minimum point cloud ordinates of the target point cloud;

[0022] If the first difference is greater than the second difference, then the current target corresponding to the target point cloud is determined to be lateral relative to the current driving direction of the radar device;

[0023] If the first difference is less than the second difference, then the current target corresponding to the target point cloud is determined to be longitudinal relative to the current driving direction of the radar device;

[0024] Determine whether the current driving direction of the current target is consistent with the historical driving direction of the historically tracked target, and obtain a third determination result;

[0025] Determining the median point cloud coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historically tracked target includes:

[0026] If the third determination result indicates that the current driving direction of the current target is consistent with the historical driving direction of the historically tracked target, then the median point cloud coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historically tracked target are determined.

[0027] Furthermore, in some embodiments, the above method further includes:

[0028] The fourth judgment result is obtained by determining whether the number of target point clouds that are related to historical tracking targets is greater than or equal to a preset number.

[0029] Determining the median point cloud coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historically tracked target includes:

[0030] If the fourth determination result indicates that the number of target point clouds that are associated with historical tracking targets is greater than or equal to a preset number, then the median point cloud coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historical tracking target are determined.

[0031] Furthermore, in some embodiments, calculating the target weight value of the target point cloud based on the difference between the target point cloud coordinates in the direction perpendicular to the historical driving direction and the median of the point cloud coordinates includes:

[0032] For any of the target point clouds, determine whether the target point cloud coordinates in the direction perpendicular to the historical driving direction are equal to the median of the point cloud coordinates, and obtain the fifth determination result;

[0033] If the fifth judgment result indicates that the target point cloud coordinates are equal to the median of the point cloud coordinates, then the first preset weight is determined as the initial weight value of the target point cloud;

[0034] If the fifth judgment result indicates that the target point cloud coordinates are not equal to the median of the point cloud coordinates, then it is determined whether the difference between the target point cloud coordinates and the median of the point cloud coordinates is greater than the width value of the preset multiple of the historical tracked target, and a sixth judgment result is obtained.

[0035] If the sixth judgment result indicates that the difference value is greater than the width value of the preset multiple of the historical tracked target, then the second preset weight is determined as the initial weight value of the target point cloud;

[0036] If the sixth judgment result indicates that the difference value is less than or equal to the width value of the historical tracked target by a preset multiple, then the quotient of the width value of the historical tracked target and the difference value is determined as the specified weight value of the target point cloud;

[0037] The specified weight values ​​of each target point cloud are normalized to obtain the initial weight values ​​of the target point cloud.

[0038] The initial weight values ​​of each target point cloud are normalized to obtain the target weight values ​​of each target point cloud.

[0039] Furthermore, in some embodiments, if the historical driving direction is horizontal, then the target point cloud coordinates in the direction perpendicular to the historical driving direction are the point cloud ordinates.

[0040] The step of determining the median point cloud coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historically tracked target, based on the point cloud data of the target point cloud, includes:

[0041] Then, the ordinates of the point cloud are sorted in ascending order of the point cloud ordinates, or in descending order of the point cloud ordinates, to obtain a sequence of ordinates.

[0042] The median of the ordinate sequence is determined to obtain the median of the ordinate of the point cloud.

[0043] Furthermore, in some embodiments, if the historical driving direction is longitudinal, then the target point cloud coordinates in the direction perpendicular to the historical driving direction are the abscissas of the point cloud.

[0044] The step of determining the median point cloud coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historically tracked target, based on the point cloud data of the target point cloud, includes:

[0045] The abscissas of each target point cloud are sorted in ascending order of point cloud abscissas, or in descending order of point cloud abscissas, to obtain a sequence of abscissas.

[0046] The median of the x-coordinate sequence is determined to obtain the median x-coordinate of the point cloud.

[0047] Furthermore, in some embodiments, generating a target tracking result for the historical tracking target based on the point cloud data of the target point cloud and the target weight value of the target point cloud includes:

[0048] Target detection processing is performed based on the point cloud data of the target point cloud and the target weight value of the target point cloud to obtain the current target's position information and driving speed information;

[0049] Obtain the historical location information and historical driving speed information of the historically tracked target;

[0050] Based on the current target's location and speed information, and the historical location and speed information of the historical tracked target, a target tracking result is generated to reflect whether the current target and the historical tracked target are the same target.

[0051] Furthermore, in some embodiments, the above method further includes:

[0052] Based on the current target's position information and speed information, determine whether the current target is a lateral moving target in front of the radar device, and obtain the seventh determination result;

[0053] If the seventh determination result indicates that the current target is a lateral moving target in front of the radar device, then the current target is determined as the tracking target of the forward traffic crossing warning function.

[0054] On the other hand, the present invention also provides a method for selecting a tracking target for a forward traffic crossing warning function, comprising:

[0055] Acquire point cloud data of targets that are correlated with historically tracked targets, collected by radar equipment;

[0056] Based on the point cloud data of the target point cloud, determine the median point cloud coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historical tracked target;

[0057] The target weight value of the target point cloud is calculated based on the difference between the target point cloud coordinates in the vertical direction of the historical driving direction and the median of the point cloud coordinates; wherein the target weight value is negatively correlated with the difference value.

[0058] Target detection processing is performed based on the point cloud data of the target point cloud and the target weight value of the target point cloud to obtain the current target's position information and driving speed information;

[0059] If the current target is determined to be a lateral moving target in front of the radar device based on the current target's position information and speed information, then the current target is identified as the tracking target for the forward traffic crossing warning function.

[0060] On the other hand, the present invention also provides a domain controller, comprising: a processor and a memory; wherein the memory stores computer-readable instructions adapted to be loaded by the processor and to execute the steps of the above-described method.

[0061] On the other hand, the present invention also provides a computer program product having at least one instruction stored thereon, wherein the at least one instruction, when executed by a controller, implements the steps of the above method.

[0062] According to the radar target tracking method provided by the present invention, after the radar equipment collects point cloud data of target point clouds that are related to historically tracked targets, the median point cloud coordinates of the target point cloud in the vertical direction of the historical driving direction of the historically tracked target can be determined based on the point cloud data of each target point cloud; and the target weight value of the target point cloud can be determined based on the difference between the target point cloud coordinates in the vertical direction of the historical driving direction and the median point cloud coordinates. Since the target weight value of the target point cloud is negatively correlated with its corresponding difference value, it is more in line with the actual situation and actual needs, thereby reducing the adverse effects of outlier point clouds and point clouds generated under abnormal radar wave propagation or reflection on the target tracking results, so as to improve the accuracy of the generated target tracking results for historically tracked targets based on the point cloud data of the target point cloud and the target weight value of the target point cloud.

[0063] It should be understood that the description in the Summary of the Invention is not intended to limit the key or essential features of the embodiments of the present invention, nor is it intended to restrict the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0064] Figure 1 A schematic flowchart of a radar target tracking method provided in an embodiment of the present invention;

[0065] Figure 2 A schematic diagram illustrating the azimuth angle of a historically tracked target relative to a radar device, provided as an embodiment of the present invention;

[0066] Figure 3 This is a schematic diagram illustrating the distribution of a target point cloud that is associated with historically tracked targets, provided as an embodiment of the present invention.

[0067] Figure 4 This is a schematic diagram illustrating the distribution of a target point cloud that is associated with a historically tracked target, as provided in an embodiment of the present invention.

[0068] Figure 5 A flowchart illustrating another radar target tracking method provided in an embodiment of the present invention;

[0069] Figure 6 A flowchart illustrating a method for selecting a tracking target for a forward traffic crossing warning function provided in an embodiment of the present invention;

[0070] Figure 7 This is a schematic diagram of the structure of a domain controller provided in an embodiment of the present invention. Detailed Implementation

[0071] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0072] In the description of one or more embodiments of the present invention, the term "comprising" and similar terms should be understood as open-ended inclusion, i.e., "including but not limited to". The term "based on" should be understood as "at least partially based on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The terms "first", "second", etc., may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0073] Currently, the technology of deploying radar in vehicles to detect and track surrounding targets is becoming increasingly widespread. For example, in automatic parking scenarios, radar-based target detection and tracking of surrounding obstacles helps vehicles automatically avoid obstacles and achieve safe parking. In driving scenarios, radar-based target detection and tracking of vehicles traveling laterally ahead reduces the risk of collisions between the user's vehicle and such vehicles. Alternatively, during adaptive cruise control, radar-based target detection and tracking of vehicles traveling longitudinally ahead allows for accurate selection of the target vehicle for the adaptive cruise control, enabling automatic speed control to maintain a safe distance and simplifying driving operations while ensuring vehicle safety.

[0074] However, current radar target tracking processing typically determines the target weight of each point cloud based on its radar cross section (RCS) parameters. This means that the target weight is determined by factors such as the geometric cross-sectional area of ​​the detected target, radar reflectivity, and directionality. Instead, the target weight is not determined based on the positional correlation between each point cloud and historically tracked targets. Consequently, when radar wave propagation or reflection is abnormal, some point clouds with weak positional correlation to historically tracked targets may have a greater impact on the target tracking results, while some point clouds with strong positional correlation to historically tracked targets may have a smaller impact, leading to poor accuracy in radar target tracking.

[0075] Therefore, improving the accuracy of radar target tracking results has become an urgent technical problem to be solved.

[0076] This invention proposes a radar target tracking method. After the radar equipment acquires point cloud data of target point clouds that are related to historically tracked targets, the method can determine the median coordinates of the target point clouds in the vertical direction perpendicular to the historical driving direction of the historically tracked targets, based on the point cloud data of each target point cloud. Furthermore, it determines the target weight value of the target point cloud based on the difference between the target point cloud coordinates in the vertical direction perpendicular to the historical driving direction and the median coordinates. Since the target weight value of the target point cloud is negatively correlated with its corresponding difference value, it better reflects actual conditions and needs, thereby reducing the adverse effects of outlier point clouds and point clouds generated under abnormal radar wave propagation or reflection on the target tracking results. This improves the accuracy of the generated target tracking results for historically tracked targets by combining the point cloud data and the target weight value of the target point clouds. When this radar target tracking method is applied to vehicle operation scenarios, it helps ensure the normal operation of various functions implemented by vehicles equipped with radar equipment based on radar target tracking results.

[0077] Please see Figure 1 This is a schematic flowchart illustrating a radar target tracking method provided in an embodiment of the present invention. From a programming perspective, the executing entity of this process can be a program for performing radar target tracking. Alternatively, the executing entity can also be a device, vehicle, or domain controller equipped with a program for performing radar target tracking; no specific limitation is made thereto.

[0078] The following is about Figure 1 The process shown will be described in detail. The radar target tracking method may specifically include the following steps:

[0079] Step S102: Obtain point cloud data of target point clouds that are related to historically tracked targets, collected by radar equipment.

[0080] In this embodiment of the invention, the radar device can periodically detect the surrounding environment to obtain point cloud data of the surrounding environment at various times. If a historically tracked target is detected based on the point cloud data of the surrounding environment at the previous time, after obtaining the point cloud data of the surrounding environment at the current time, the target point cloud data that is related to the historically tracked target can be determined from the point cloud data of the surrounding environment at the current time based on the historical position information of the historically tracked target at the previous time. Then, the current target to which these target point clouds belong is identified as the historically tracked target, thereby realizing radar target detection and tracking.

[0081] In practical applications, based on the historical location information of the tracked target at the previous moment, the possible current location of the historical tracked target can be determined, and the surrounding environmental point cloud within that area can be identified as a target point cloud associated with the historical tracked target. Furthermore, based on the speed and direction of the surrounding environmental point cloud collected at the current moment, as well as the historical speed and direction of each historical tracked target, the similarity between the surrounding environmental point cloud and each historical tracked target can be calculated. This helps determine whether the surrounding environmental point cloud belongs to the target point cloud associated with the historical tracked target, thus improving the accuracy of identifying target point clouds associated with historical tracked targets. Of course, other methods can also be used to determine target point clouds associated with historical tracked targets; no specific limitations are imposed on this approach.

[0082] In practical applications, the type of radar equipment can be set according to actual needs. For example, it can be a millimeter-wave radar or other radar equipment that can collect and generate point cloud data, etc. There is no specific limitation on this.

[0083] Step S104: Based on the point cloud data of the target point cloud, determine the median point cloud coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historical tracked target.

[0084] In this embodiment of the invention, the median coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historical tracked target can be considered as the midpoint of the target point cloud cluster in the direction perpendicular to the historical driving direction of the historical tracked target. If the coordinates of a target point cloud in the direction perpendicular to the historical driving direction of the historical tracked target are close to the median coordinates, it can generally be considered that the target point cloud is closely related to the historical tracked target; otherwise, it can be considered that the target point cloud is less related to the historical tracked target. Therefore, when it is necessary to combine the position information of the target point cloud to assign target weight values ​​to the target point cloud for radar target tracking processing, the median coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historical tracked target can be calculated.

[0085] Step S106: Calculate the target weight value of the target point cloud based on the difference between the target point cloud coordinates in the vertical direction of the historical driving direction and the median of the point cloud coordinates; wherein the target weight value is negatively correlated with the difference value.

[0086] In this embodiment of the invention, when the difference between the target point cloud coordinates and the median coordinates in the direction perpendicular to the historical driving direction is small, it indicates that the target point cloud is relatively close to the center of the target point cloud cluster, and its influence on radar target detection should be relatively large. Therefore, a higher target weight can be assigned to this target point cloud. Similarly, when the difference between the target point cloud coordinates and the median coordinates in the direction perpendicular to the historical driving direction is large, it indicates that the target point cloud is relatively far from the center of the target point cloud cluster, and its influence on radar target detection should be relatively small. Therefore, a lower target weight can be assigned to this target point cloud.

[0087] Since step S106 makes the target weight value of the target point cloud negatively correlated with its corresponding difference value, target point clouds with large deviations in the vertical direction of the historical travel direction of the tracked target will be disregarded or have minimal impact during radar target detection and tracking. This helps ensure the rationality and accuracy of the determined target weight value of the target point cloud. Furthermore, it can reduce the error caused by target point clouds generated by radar wave propagation and reflection anomalies in the target tracking results, thereby improving the accuracy and stability of the target tracking results.

[0088] Step S108: Generate a target tracking result for the historical tracking target based on the point cloud data of the target point cloud and the target weight value of the target point cloud.

[0089] In this embodiment of the invention, the point cloud data of the target point cloud collected at the current moment and the target weight value of the target point cloud, as well as the relevant data of historically tracked targets, can be combined to identify whether the current target and the historically tracked targets to which these target point clouds belong are the same target, thereby realizing radar target tracking in a convenient and fast manner.

[0090] Figure 1 The proposed method determines the target weight value of a point cloud based on the difference between the target point cloud coordinates and the median coordinates in the direction perpendicular to the historical travel direction of the tracked target. Furthermore, it ensures a negative correlation between the target weight value and its corresponding difference value, better reflecting actual conditions and needs. This avoids assigning excessive weights to outliers and prevents inaccurate weights due to radar wave propagation or reflection anomalies when using radar cross-section parameters as weights. This allows for the improvement of the accuracy of generated target tracking results for historically tracked targets by using accurate target weight values ​​from the point cloud.

[0091] In one feasible implementation, Figure 1 The method described herein may further include:

[0092] Based on the historical speed information of the historical tracked target, the historical driving direction of the historical tracked target relative to the radar device is determined; wherein, the historical driving direction includes either lateral or longitudinal.

[0093] In this embodiment of the invention, to simplify the scheme, the historical tracking target's historical travel direction relative to the radar equipment at the previous moment can be broadly divided into two types: lateral and longitudinal. If it is longitudinal, it means that the historical tracking target and the radar equipment are roughly in the same direction of travel, while if it is lateral, it means that the historical tracking target and the radar equipment's direction of travel may easily intersect.

[0094] For example, if we assume that the direction of travel of the radar device is the vertical axis of a two-dimensional preset coordinate system, then when the historical tracking target is laterally relative to the historical travel direction of the radar device, the historical travel direction of the historical tracking target can be considered to be biased towards the horizontal axis of the preset coordinate system; while when the historical tracking target is longitudinally relative to the historical travel direction of the radar device, the historical travel direction of the historical tracking target can be considered to be biased towards the vertical axis of the preset coordinate system.

[0095] In practical applications, if we assume the radar device's travel direction is the longitudinal axis of a preset coordinate system, we can identify whether the target's velocity component along the longitudinal axis is less than its velocity component along the transverse axis based on its historical speed information. If so, the target's historical travel direction relative to the radar device can be determined to be lateral; otherwise, it can be determined to be longitudinal. Alternatively, we can determine the difference in travel direction angle between the target's historical speed information and the radar device's historical speed information, thus determining whether the target's historical travel direction relative to the radar device is lateral or longitudinal. Currently, other methods can also be used to determine whether the target's historical travel direction relative to the radar device is lateral or longitudinal, offering greater flexibility, and no specific limitations are imposed.

[0096] It is understandable that, in vehicle operation scenarios, radar equipment is usually mounted on vehicles. Therefore, the (historical) driving direction of the aforementioned historical tracked target relative to the radar equipment can actually refer to the (historical) driving direction of the aforementioned historical tracked target relative to the vehicle equipped with the radar equipment. This will not be elaborated further.

[0097] In one feasible implementation, Figure 1 The method described herein may further include:

[0098] Determine whether the historical tracking target is lateral to the historical driving direction of the radar device to obtain a first determination result.

[0099] If the first determination result indicates that the historical driving direction is not lateral, then the azimuth angle of the historical tracking target relative to the radar device is determined based on the historical position information of the historical tracking target.

[0100] Determine whether the absolute value of the azimuth angle is within a preset numerical range to obtain a second determination result; wherein the starting value of the preset numerical range is greater than 0.

[0101] Correspondingly, determining the median coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historically tracked target may include:

[0102] If the first judgment result indicates that the historical driving direction is horizontal, or if the second judgment result indicates that the absolute value of the azimuth angle is within a preset value range, then the median point cloud coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historical tracked target are determined.

[0103] In this embodiment of the invention, when the historical tracking target is laterally positioned relative to the radar device's historical driving direction, the radar device can typically collect relatively comprehensive point cloud data of the entire historical tracking target, thereby ensuring the accuracy of the subsequently generated radar target tracking results. However, when the historical tracking target is longitudinally positioned relative to the radar device, if the azimuth angle of the historical tracking target relative to the radar device is small, the point cloud data collected by the radar device for the historical tracking target is usually concentrated in the front / rear bumper area of ​​the historical tracking target, resulting in poor comprehensiveness and thus failing to guarantee the accuracy of the subsequently generated radar target tracking results.

[0104] Based on this, subsequent radar target tracking processing can be allowed when it is determined that the historical travel direction of the tracked target relative to the radar equipment is lateral. Alternatively, subsequent radar target tracking processing can be allowed only when the historical travel direction of the tracked target relative to the radar equipment is longitudinal, but the absolute value of the azimuth angle between them is within a preset range. The preset range can be set according to actual needs, for example, it can be [20°, 160°], [45°, 135°], etc., thus ensuring that even when the historical travel direction of the tracked target relative to the radar equipment is longitudinal, comprehensive point cloud data can still be collected for all positions of the tracked target before subsequent radar target tracking processing is allowed, which helps ensure the accuracy of the subsequently generated radar target tracking results.

[0105] In one feasible implementation, determining the azimuth angle of the historically tracked target relative to the radar device based on the historical position information of the historically tracked target may include:

[0106] If the center point of the radar device is located at the origin of the preset coordinate system, the quotient of the x-coordinate and y-coordinate of the center point of the historical tracked target in the preset coordinate system is used as the independent variable of the arctangent function to calculate the azimuth angle of the historical tracked target relative to the radar device.

[0107] In this embodiment of the invention, the azimuth angle can also be called the position angle; it can represent the horizontal angle between the north direction line from the preset point and the target direction line in a clockwise or counterclockwise direction. To facilitate the calculation of the azimuth angle of historically tracked targets relative to the radar equipment, the center point of the radar equipment can be located at the origin of the preset coordinate system. After determining the x and y coordinates of the center point of the historically tracked target in the preset coordinate system, the y-axis of the preset coordinate system can be used as the "north direction line of the preset point", and the line connecting the origin (i.e., the center point of the radar equipment) to the center point of the historically tracked target can be used as the target direction line. Thus, the angle value obtained by calculating arctan (the quotient of the x and y coordinates of the center point of the historically tracked target) can be used as the azimuth angle of the historically tracked target relative to the radar equipment, which is convenient and quick.

[0108] Figure 2 A schematic diagram illustrating the azimuth angle of a historically tracked target relative to a radar device, provided as an embodiment of the present invention, is shown below. Figure 2 As shown. Assume the horizontal axis of the preset coordinate system is as indicated by arrow 21, the vertical axis by arrow 22, and the center point of the radar device 25 is located at the origin of the preset coordinate system. After determining the horizontal and vertical coordinates of the center point 24 of the historically tracked target 23, the quotient of the horizontal and vertical coordinates of the center point 24 of the historically tracked target 23 can be used as the independent variable of the arctangent function for calculation, yielding the azimuth angle 26 / 27 of the historically tracked target relative to the radar device.

[0109] Understandably, in vehicle operation scenarios, radar equipment can be mounted on vehicles, and the coordinates of the vehicle's center point can usually be easily obtained. Furthermore, the distance between the radar equipment's installation location and the vehicle's center point is typically much smaller than the distance between the vehicle and the target ahead. Therefore, the coordinates of the vehicle's center point can be considered as the coordinates of the radar equipment's center point. Based on this, the vehicle's center point can be used as the origin of a preset coordinate system, and the quotient of the x-coordinate and y-coordinate of the center point of a historically tracked target in the preset coordinate system can be used as the independent variable of the arctangent function. This allows the calculated azimuth angle of the historically tracked target relative to the vehicle to be used as the azimuth angle of the historically tracked target relative to the radar equipment, simplifying the azimuth angle calculation process.

[0110] In one feasible implementation, Figure 1 The method may further include:

[0111] Based on the point cloud data of each target point cloud, determine the first difference between the maximum and minimum point cloud abscissas of the target point cloud.

[0112] Based on the point cloud data of the target point cloud, determine the second difference between the maximum and minimum point cloud ordinates of the target point cloud.

[0113] If the first difference is greater than the second difference, then the current target corresponding to the target point cloud is determined to be lateral relative to the current driving direction of the radar device.

[0114] If the first difference is less than the second difference, then the current target corresponding to the target point cloud is determined to be longitudinal relative to the current driving direction of the radar device.

[0115] A third judgment result is obtained by determining whether the current driving direction of the current target is consistent with the historical driving direction of the historically tracked target.

[0116] Correspondingly, determining the median coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historically tracked target may include:

[0117] If the third determination result indicates that the current driving direction of the current target is consistent with the historical driving direction of the historically tracked target, then the median point cloud coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historically tracked target are determined.

[0118] In the embodiments of this specification, when the historically tracked target is lateral relative to the historical driving direction of the radar device, or when the historically tracked target is longitudinal relative to the historical driving direction of the radar device, but the absolute value of the azimuth angle between the two is within a preset range, the position of the target point cloud associated with the historically tracked target can roughly describe the shape of the current target; and since a vehicle can generally be regarded as a rectangle in a two-dimensional coordinate system, and the long side direction of the rectangle corresponding to the vehicle can be regarded as the driving direction of the vehicle, the shape of the current target corresponding to each target point cloud can be determined based on the position information of each target point cloud, thereby determining the driving direction of the current target (e.g., lateral or longitudinal).

[0119] Because the interval between the previous moment and the current moment is usually small, and a vehicle typically cannot change from lateral to longitudinal movement within such a short time, if the current target's direction of travel is inconsistent with the historical direction of travel of a previously tracked target, the two are likely not the same target. Therefore, to avoid generating erroneous radar target tracking results, subsequent radar target tracking processing steps can be discontinued. However, if the current target's direction of travel is consistent with the historical direction of travel of a previously tracked target, subsequent radar target tracking processing steps can continue, which helps improve the accuracy of the radar target tracking results.

[0120] In practical applications, if we assume the radar device's travel direction (e.g., the travel direction of a vehicle carrying the radar device) is the vertical axis of a preset coordinate system, then the first difference between the maximum and minimum x-coordinates of the target point cloud can be used as the side lengths of the bottom and top edges of the current target corresponding to the target point cloud. Similarly, the second difference between the maximum and minimum y-coordinates of the target point cloud can be used as the side lengths of the two sides of the current target corresponding to the target point cloud. If the first difference is greater than the second difference, it indicates that the length of the top / bottom edge of the current target corresponding to the target point cloud is greater than the length of the sides. In this case, the travel direction of the current target corresponding to the target point cloud can be approximately the horizontal axis of the preset coordinate system, thus determining that the current target corresponding to the target point cloud is lateral relative to the radar device's current travel direction. Likewise, if the first difference is less than the second difference, it indicates that the length of the top / bottom edge of the current target corresponding to the target point cloud is less than the length of the sides. In this case, the travel direction of the current target corresponding to the target point cloud can be approximately the vertical axis of the preset coordinate system, thus determining that the current target corresponding to the target point cloud is longitudinal relative to the radar device's current travel direction.

[0121] Figure 3 This invention provides a schematic diagram of the distribution of target point clouds that are associated with historically tracked targets, as shown in the embodiment of the invention. Figure 3 As shown, assuming the horizontal axis of the preset coordinate system is as indicated by arrow 31 and the vertical axis is as indicated by arrow 32, and the driving direction of the radar equipment (e.g., the driving direction of the vehicle carrying the radar equipment) is the vertical axis; the target point cloud associated with historically tracked targets can include seven points, from point 33 to point 39. The horizontal coordinate of the largest point cloud corresponding to the target point cloud can be the horizontal coordinate of point 37, and the horizontal coordinate of the smallest point cloud corresponding to the target point cloud can be the horizontal coordinate of point 33; the vertical coordinate of the largest point cloud corresponding to the target point cloud can be the vertical coordinate of point 39, and the vertical coordinate of the smallest point cloud corresponding to the target point cloud can be the vertical coordinate of point 33. Calculations show that the first difference between the horizontal coordinates of point 37 and point 33 is greater than the second difference between the vertical coordinates of point 39 and point 33. Therefore, it can be determined that the current target corresponding to the target point cloud is lateral relative to the current driving direction of the radar equipment.

[0122] Figure 4 This is a schematic diagram illustrating the distribution of target point clouds that are associated with historically tracked targets, as provided in an embodiment of the present invention. Figure 4As shown, assuming the horizontal axis of the preset coordinate system is as indicated by arrow 41 and the vertical axis is as indicated by arrow 42, and the driving direction of the radar equipment (e.g., the driving direction of the vehicle carrying the radar equipment) is the vertical axis; the target point cloud associated with historically tracked targets can include seven points, from point 43 to point 49. The horizontal coordinate of the largest point cloud corresponding to the target point cloud can be the horizontal coordinate of point 49, and the horizontal coordinate of the smallest point cloud corresponding to the target point cloud can be the horizontal coordinate of point 43; the vertical coordinate of the largest point cloud corresponding to the target point cloud can be the vertical coordinate of point 47, and the vertical coordinate of the smallest point cloud corresponding to the target point cloud can be the vertical coordinate of point 48. Calculations show that the first difference between the horizontal coordinates of point 49 and point 43 is less than the second difference between the vertical coordinates of point 47 and point 48. Therefore, it can be determined that the current target corresponding to the target point cloud is vertical relative to the current driving direction of the radar equipment. Further details are omitted here.

[0123] In one feasible implementation, Figure 1 The method may further include:

[0124] The fourth judgment result is obtained by determining whether the number of target point clouds that are related to historical tracking targets is greater than or equal to a preset number.

[0125] Correspondingly, determining the median coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historically tracked target may include:

[0126] If the fourth determination result indicates that the number of target point clouds that are associated with historical tracking targets is greater than or equal to a preset number, then the median point cloud coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historical tracking target are determined.

[0127] In this embodiment of the specification, when the number of target point clouds that are related to historically tracked targets acquired at the current moment is too small, it is often impossible to accurately identify whether the current target corresponding to the target point cloud and the historically tracked target are the same target. Therefore, it is possible to prohibit the continued execution of subsequent radar target tracking processing. However, when the number of target point clouds that are related to historically tracked targets acquired at the current moment reaches a preset number, the subsequent radar target tracking processing is then executed, which helps to ensure the accuracy of the generated radar target tracking results. The preset number can be set according to actual needs. For example, the preset number can be 5, 10, 15, etc., and there is no specific limitation on it.

[0128] In one feasible implementation, calculating the target weight value of the target point cloud based on the difference between the target point cloud coordinates in the direction perpendicular to the historical driving direction and the median of the point cloud coordinates may include:

[0129] For any of the target point clouds, determine whether the target point cloud coordinates in the direction perpendicular to the historical driving direction are equal to the median of the point cloud coordinates, and obtain the fifth determination result.

[0130] If the fifth judgment result indicates that the target point cloud coordinates are equal to the median of the point cloud coordinates, then the first preset weight is determined as the initial weight value of the target point cloud.

[0131] If the fifth judgment result indicates that the target point cloud coordinates are not equal to the median of the point cloud coordinates, then it is determined whether the difference between the target point cloud coordinates and the median of the point cloud coordinates is greater than the width value of a preset multiple of the historical tracked target, and a sixth judgment result is obtained.

[0132] If the sixth judgment result indicates that the difference value is greater than the width value of the preset multiple of the historical tracking target, then the second preset weight is determined as the initial weight value of the target point cloud.

[0133] If the sixth determination result indicates that the difference value is less than or equal to the width value of the historical tracked target by a preset multiple, then the quotient of the width value of the historical tracked target and the difference value is determined as the specified weight value of the target point cloud.

[0134] The specified weight values ​​of each target point cloud are normalized to obtain the initial weight values ​​of the target point cloud.

[0135] The initial weight values ​​of each target point cloud are normalized to obtain the target weight values ​​of each target point cloud.

[0136] In this embodiment of the specification, in order to simplify the weight calculation process, a first preset weight can be assigned to the target point cloud to which the median of the point cloud coordinates belongs, as an initial weight value.

[0137] Because when the difference between the target point cloud coordinates and the median coordinates of the corresponding point cloud in the direction perpendicular to the historical travel direction of the tracked target is too large, the influence weight of the target point cloud on the radar target tracking result should generally be greatly reduced. Therefore, when the difference value corresponding to the target point cloud is greater than a preset multiple of the width value of the historical tracked target, a second preset weight can be assigned to the target point cloud as an initial weight value. The width value of the historical tracked target can be the length of the shorter side of the virtual model of the historical tracked target established during radar target detection and tracking. In practical applications, the width value of the historical tracked target often changes with the distance and azimuth angle between the historical tracked target and the radar equipment. Therefore, the preset multiple width value of the historical tracked target used at different times may vary. The preset multiple can be set according to actual needs, for example, it can be 1.5 times, 2 times, etc., without specific limitation.

[0138] In practical applications, the second preset weight is usually less than the first preset weight; for example, the second preset weight and the second preset weight can be set to 0 and 1 respectively. Of course, they can also be set to other values ​​according to actual needs, and there is no specific limitation on this.

[0139] For other target point clouds (i.e., target point clouds whose corresponding difference value is greater than 0 and less than or equal to the width value of a preset multiple of the historical tracked target), the difference value corresponding to this part of the target point cloud can be linearly negatively correlated with the initial weight value to be assigned, which is beneficial to improving the rationality and accuracy of the weights assigned to each target point cloud.

[0140] In practical applications, to facilitate subsequent radar target tracking processing, the initial weight values ​​assigned to each target point cloud can be normalized to obtain the target weight values ​​for each target point cloud. This will not be elaborated upon further.

[0141] In one feasible implementation, if the historical driving direction of the historically tracked target is horizontal, then the target point cloud coordinates in the vertical direction of the historical driving direction can be the point cloud ordinates.

[0142] Correspondingly, determining the median point cloud coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historically tracked target, based on the point cloud data of the target point cloud, may include:

[0143] Then, the ordinates of the point cloud are sorted in ascending order or descending order to obtain a sequence of ordinates.

[0144] The median of the ordinate sequence is determined to obtain the median of the ordinate of the point cloud.

[0145] To facilitate understanding of the process of determining the median coordinates of the point cloud when a historically tracked target moves laterally relative to radar equipment, this is combined with... Figure 3 Provide an explanation. For example... Figure 3 As shown, it is assumed that the radar equipment's travel direction (e.g., the travel direction of the vehicle carrying the radar equipment) is the longitudinal axis direction indicated by arrow 32, and the historically tracked target is traveling laterally relative to the radar equipment. Furthermore, the target point cloud associated with the historically tracked target may include seven points, from point 33 to point 39. In this case, if the longitudinal coordinates of each target point cloud are sorted in ascending order, the resulting longitudinal coordinate sequence can be the longitudinal coordinates of points 33, 34, 35, 36, 37, 38, and 39. Since the longitudinal coordinate of point 36 is the median of the longitudinal coordinate sequence, the longitudinal coordinate of point 36 can be used as the median of the point cloud longitudinal coordinates.

[0146] In one feasible implementation, if the historical driving direction of the historically tracked target is longitudinal, then the target point cloud coordinates in the direction perpendicular to the historical driving direction can be the abscissa of the point cloud.

[0147] Correspondingly, determining the median point cloud coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historically tracked target, based on the point cloud data of the target point cloud, may include:

[0148] The abscissas of each target point cloud are sorted in ascending order or descending order to obtain a sequence of abscissas.

[0149] The median of the x-coordinate sequence is determined to obtain the median x-coordinate of the point cloud.

[0150] To facilitate understanding of the process of determining the median coordinates of the point cloud when a historically tracked target moves longitudinally relative to the radar equipment, this is combined with... Figure 4 Provide an explanation. For example... Figure 4As shown, it is assumed that the radar equipment's travel direction (e.g., the travel direction of the vehicle carrying the radar equipment) is the longitudinal axis direction indicated by arrow 42, and the historically tracked target travels longitudinally relative to the radar equipment. Furthermore, the target point cloud associated with the historically tracked target may include seven points, from point 43 to point 49. In this case, if the abscissas of each target point cloud are sorted in ascending order, the resulting abscissa sequence can be the abscissas of points 43, 44, 45, 46, 47, 48, and 49. Since the abscissa of point 46 is the median of the abscissa sequence, the abscissa of point 46 can be used as the median of the point cloud abscissas.

[0151] In one feasible implementation, generating a target tracking result for the historical tracking target based on the point cloud data of the target point cloud and the target weight value of the target point cloud may include:

[0152] Target detection processing is performed based on the point cloud data of the target point cloud and the target weight value of the target point cloud to obtain the current target's position information and driving speed information.

[0153] Obtain the historical location information and historical driving speed information of the historical tracking target.

[0154] Based on the current target's location and speed information, and the historical location and speed information of the historical tracked target, a target tracking result is generated to reflect whether the current target and the historical tracked target are the same target.

[0155] In the embodiments of this specification, the point cloud data of the target point cloud may include point cloud position information (e.g., point cloud x-coordinate, point cloud y-coordinate, etc.), point cloud speed magnitude and direction information, etc. Based on this, Kalman filtering, extended Kalman filtering, or other target detection algorithms / models, target tracking algorithms / models, etc., can be used to perform target detection processing based on the point cloud data of the target point cloud and the target weight value of the target point cloud, so as to obtain accurate position information and speed information (which may include speed magnitude information and speed direction information) of the current target. This can improve the accuracy of the subsequently generated target tracking results used to reflect whether the current target and historically tracked targets are the same target.

[0156] In one feasible implementation, Figure 1 The method may further include:

[0157] Based on the current target's position and speed information, determine whether the current target is a lateral moving target in front of the radar device, and obtain the seventh determination result.

[0158] If the seventh determination result indicates that the current target is a lateral moving target in front of the radar device, then the current target is determined as the tracking target of the forward traffic crossing warning function.

[0159] In this embodiment, the Front Cross Traffic Alert (FCTA) function aims to monitor other road users, such as pedestrians and vehicles, approaching laterally in front of the vehicle in real time using sensors on the exterior of the vehicle. It then issues warnings when a collision risk is suspected, prompting the driver to take timely action to ensure driving safety. Therefore, if a radar device is mounted on a preset vehicle, and the radar target tracking process identifies the current target as a vehicle traveling laterally in front of the radar device, this current target can be designated as the tracking target of the FCTA function at the preset vehicle, thus contributing to the safety of the preset vehicle's driving process.

[0160] Because the accuracy of the target weights determined by the radar cross-section (RCS) parameters of each point cloud in existing radar target tracking schemes is poor, it affects the accuracy of the calculated position and speed information of the current target, and thus affects the accuracy of the radar target tracking results. This not only easily leads to the interruption of target tracking in the forward traffic crossing warning function, but also easily leads to errors in the generated heading information of the current target due to the poor accuracy of the calculated speed information of the current target. This makes it impossible to set the forward traffic crossing warning function of the vehicle with radar equipment as the tracking target of the vehicle with radar equipment, and thus cannot reduce the collision risk between the preset vehicle and the vehicle with cross traffic.

[0161] Based on this Figure 1 The Chinese method and related embodiments improve the rationality and accuracy of the target weights determined in the target point cloud, thereby enhancing the accuracy of the calculated current target's position and speed information. This not only ensures the accuracy of radar target tracking results and reduces the likelihood of target tracking interruptions in the forward traffic crossing warning function, but also ensures the reliability of the determined current target's heading information based on the accurate speed information of the current target. This allows for the timely setting of laterally moving vehicles ahead as tracking targets for the forward traffic crossing warning function at preset vehicles equipped with radar equipment, thus reducing the risk of collisions between preset vehicles and vehicles ahead and ensuring the safety of the preset vehicles during their operation.

[0162] For ease of understanding, Figure 5 This is a schematic flowchart illustrating a relatively complete radar target tracking method provided in an embodiment of the present invention. Figure 5As shown,

[0163] Step S501: Determine the historical driving direction of the historical tracked target relative to the radar equipment based on the historical driving speed information of the historical tracked target; wherein, the historical driving direction may include either the lateral or longitudinal direction.

[0164] Step S502: Determine whether the historical tracking target is lateral relative to the historical driving direction of the radar equipment, and obtain the first judgment result.

[0165] If the first judgment result indicates that the historical driving direction is horizontal, then the process can proceed to step S503: determine whether the number of target point clouds that are related to the historical tracking target is greater than or equal to a preset number, and obtain the fourth judgment result.

[0166] If the first judgment result indicates that the historical driving direction is not lateral, then the process can proceed to step S504: determine the azimuth angle of the historical tracking target relative to the radar equipment based on the historical position information of the historical tracking target.

[0167] Step S505: Determine whether the absolute value of the azimuth angle is within a preset numerical range to obtain a second determination result; wherein, the starting value of the preset numerical range is greater than 0.

[0168] If the second determination result indicates that the absolute value of the azimuth angle is outside the preset value range, then the process can be skipped to step S520 and the process ends.

[0169] If the second judgment result indicates that the absolute value of the azimuth angle is within the preset value range, then the process can proceed to step S503: determine whether the number of target point clouds that are related to historical tracking targets is greater than or equal to a preset number, and obtain the fourth judgment result.

[0170] If the fourth judgment result indicates that the number of target point clouds is less than the preset number, then the process can be skipped to step S520 and the process ends.

[0171] If the fourth judgment result indicates that the number of target point clouds is greater than or equal to the preset number, then step S506 can be executed: based on the point cloud data of each target point cloud, determine the first difference between the maximum point cloud horizontal coordinate and the minimum point cloud horizontal coordinate of the target point cloud.

[0172] Step S507: Based on the point cloud data of the target point cloud, determine the second difference between the maximum and minimum point cloud ordinates of the target point cloud.

[0173] Step S508: If the first difference is greater than the second difference, then the current target corresponding to the target point cloud is determined to be horizontal relative to the current driving direction of the radar device; if the first difference is less than the second difference, then the current target corresponding to the target point cloud is determined to be vertical relative to the current driving direction of the radar device.

[0174] Step S509: Determine whether the current driving direction of the current target is consistent with the historical driving direction of the historically tracked target, and obtain a third determination result.

[0175] If the third judgment result indicates that the current driving direction of the current target is inconsistent with the historical driving direction of the historically tracked target, then the process can be skipped to step S520 and the process ends.

[0176] If the third judgment result indicates that the current driving direction of the current target is consistent with the historical driving direction of the historical tracked target, then step S510 can be executed: determine the median point cloud coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historical tracked target.

[0177] Step S511: For any of the target point clouds, determine whether the target point cloud coordinates in the direction perpendicular to the historical driving direction are equal to the median of the point cloud coordinates, and obtain the fifth determination result.

[0178] If the fifth judgment result indicates that the target point cloud coordinates are equal to the median of the point cloud coordinates, then step S512 can be executed: the first preset weight is determined as the initial weight value of the target point cloud.

[0179] If the fifth judgment result indicates that the target point cloud coordinates are not equal to the median of the point cloud coordinates, then step S513 can be executed: determine whether the difference between the target point cloud coordinates and the median of the point cloud coordinates is greater than the width value of the preset multiple of the historical tracked target, and obtain the sixth judgment result.

[0180] If the sixth judgment result indicates that the difference value is greater than the width value of the preset multiple of the historical tracking target, then step S514 can be executed: the second preset weight is determined as the initial weight value of the target point cloud.

[0181] If the sixth judgment result indicates that the difference value is less than or equal to the width value of the historical tracked target by a preset multiple, then step S515 can be executed: the quotient of the width value of the historical tracked target and the difference value is determined as the specified weight value of the target point cloud.

[0182] Step S516: Normalize the specified weight values ​​of each target point cloud to obtain the initial weight values ​​of the target point cloud.

[0183] Step S517: Normalize the initial weight values ​​of each target point cloud to obtain the target weight values ​​of each target point cloud.

[0184] Step S518: Perform target detection processing based on the point cloud data of the target point cloud and the target weight value of the target point cloud to obtain the current target's position information and driving speed information.

[0185] Step S519: Based on the current target's position information and speed information, generate a target tracking result reflecting whether the current target and the historically tracked target are the same target. And / or, based on the current target's position information and speed information, when the current target is identified as a vehicle traveling laterally in front of the radar equipment, determine the current target as a tracking target for the forward traffic crossing warning function at a preset vehicle equipped with radar equipment.

[0186] Step S520: End.

[0187] Figure 6 This is a flowchart illustrating a method for selecting a tracking target for a forward traffic crossing warning function, provided in an embodiment of the present invention. From a programming perspective, the executing entity of this process can be a program used to implement the forward traffic crossing warning function. Alternatively, the executing entity can also be a vehicle or domain controller equipped with a program for implementing the forward traffic crossing warning function; no specific limitation is made thereto.

[0188] The following is about Figure 6 The process shown will be explained in detail. The method for selecting the tracking target of the forward traffic crossing warning function may specifically include the following steps:

[0189] Step S602: Acquire point cloud data of target point clouds that are related to historically tracked targets, collected by radar equipment.

[0190] Step S604: Based on the point cloud data of the target point cloud, determine the median point cloud coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historical tracked target.

[0191] Step S606: Calculate the target weight value of the target point cloud based on the difference between the target point cloud coordinates in the vertical direction of the historical driving direction and the median of the point cloud coordinates; wherein the target weight value is negatively correlated with the difference value.

[0192] Step S608: Perform target detection processing based on the point cloud data of the target point cloud and the target weight value of the target point cloud to obtain the current target's position information and driving speed information.

[0193] Step S610: If the current target is determined to be a lateral moving target in front of the radar device based on the current target's position information and driving speed information, then the current target is determined as the tracking target of the forward traffic crossing warning function.

[0194] In the embodiments of this specification, the technical principle of calculating the target weight value of the target point cloud in steps 602-606 is similar to... Figure 1 The technical principles underlying the calculation of the target weight value of the target point cloud in the methods and embodiments can be consistent; however, the technical principles underlying the target detection based on the point cloud data and target weight value in steps 608-608, and the technical principles underlying the screening of the tracking targets for the forward traffic crossing warning function, are consistent with... Figure 1 The technical principles of target detection based on point cloud data and target weight values ​​in the embodiments of the Chinese method are consistent with those of the tracking targets for the forward traffic crossing warning function; therefore, it can be referenced. Figure 1 To understand the content of the embodiments of the method Figure 6 The implementation principle of the method described herein will not be elaborated here.

[0195] Figure 6 The method described above improves the rationality and accuracy of the target weights in the determined target point cloud, thereby enhancing the accuracy of the calculated current target's position and speed information. This not only ensures the accuracy of radar target tracking results and reduces the interruption of target tracking in the forward traffic crossing warning function, but also ensures the reliability of the determined current target's heading information based on the accurate speed information of the current target. This allows for the timely setting of laterally moving vehicles ahead as tracking targets for the forward traffic crossing warning function at preset vehicles equipped with radar equipment, which helps reduce the collision risk between preset vehicles and vehicles ahead, ensuring the safety of the preset vehicles during their operation.

[0196] The present invention also provides a computer program product on which at least one instruction can be stored. When the at least one instruction is executed by the controller, it can realize the radar target tracking method or the target selection method of the forward traffic crossing warning function as described in the above embodiments. For the specific execution process, please refer to the specific description in the above embodiments, which will not be repeated here.

[0197] In one embodiment, the present invention also provides Figure 7 The diagram shows the structure of a domain controller. Figure 7At the hardware level, the domain controller may include a processor 71 and a memory 75, and may also include an internal bus 72, a network interface 73, memory 74, and other hardware required for services. The domain controller can be installed in a vehicle. The processor 71 can read corresponding computer-readable instructions from the memory 75 into memory and then execute them to implement the radar target tracking method or the target selection method for the forward traffic crossing warning function described above. The specific execution process can be found in the detailed descriptions of the above embodiments, and will not be repeated here.

[0198] Finally, the various embodiments in this invention are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on describing the differences from other embodiments. In particular, embodiments such as domain controllers and computer program products are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0199] The above description is merely an embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of the present invention should be included within the scope of the claims of the present invention.

Claims

1. A radar target tracking method, comprising: Acquire point cloud data of targets that are correlated with historically tracked targets, collected by radar equipment; Based on the point cloud data of the target point cloud, determine the median point cloud coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historical tracked target; The target weight value of the target point cloud is calculated based on the difference between the target point cloud coordinates in the vertical direction of the historical driving direction and the median of the point cloud coordinates; wherein the target weight value is negatively correlated with the difference value. Based on the point cloud data of the target point cloud and the target weight value of the target point cloud, a target tracking result is generated for the historical tracking target.

2. The method according to claim 1, further comprising: Based on the historical speed information of the historical tracked target, the historical driving direction of the historical tracked target relative to the radar device is determined; wherein, the historical driving direction includes either lateral or longitudinal.

3. The method according to claim 2, further comprising: Determine whether the historical driving direction of the historical tracked target relative to the radar device is lateral, and obtain a first determination result; If the first determination result indicates that the historical driving direction is not lateral, then the azimuth angle of the historical tracking target relative to the radar device is determined based on the historical position information of the historical tracking target. Determine whether the absolute value of the azimuth angle is within a preset numerical range to obtain a second determination result; wherein, the starting value of the preset numerical range is greater than 0; Determining the median point cloud coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historically tracked target includes: If the first judgment result indicates that the historical driving direction is horizontal, or if the second judgment result indicates that the absolute value of the azimuth angle is within a preset value range, then the median point cloud coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historical tracked target are determined.

4. The method according to claim 3, wherein determining the azimuth angle of the historically tracked target relative to the radar device based on the historical position information of the historically tracked target includes: If the center point of the radar device is located at the origin of the preset coordinate system, the quotient of the x-coordinate and y-coordinate of the center point of the historical tracked target in the preset coordinate system is used as the independent variable of the arctangent function to calculate the azimuth angle of the historical tracked target relative to the radar device.

5. The method according to claim 1, further comprising: Based on the point cloud data of each target point cloud, determine the first difference between the maximum and minimum point cloud abscissa of the target point cloud; Based on the point cloud data of the target point cloud, determine the second difference between the maximum and minimum point cloud ordinates of the target point cloud; If the first difference is greater than the second difference, then the current target corresponding to the target point cloud is determined to be lateral relative to the current driving direction of the radar device; If the first difference is less than the second difference, then the current target corresponding to the target point cloud is determined to be longitudinal relative to the current driving direction of the radar device; Determine whether the current driving direction of the current target is consistent with the historical driving direction of the historically tracked target, and obtain a third determination result; Determining the median point cloud coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historically tracked target includes: If the third determination result indicates that the current driving direction of the current target is consistent with the historical driving direction of the historically tracked target, then the median point cloud coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historically tracked target are determined.

6. The method according to claim 1, further comprising: The fourth judgment result is obtained by determining whether the number of target point clouds that are related to historical tracking targets is greater than or equal to a preset number. Determining the median point cloud coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historically tracked target includes: If the fourth determination result indicates that the number of target point clouds that are associated with historical tracking targets is greater than or equal to a preset number, then the median point cloud coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historical tracking target are determined.

7. The method according to claim 1, wherein calculating the target weight value of the target point cloud based on the difference between the target point cloud coordinates in the direction perpendicular to the historical driving direction and the median of the point cloud coordinates comprises: For any of the target point clouds, determine whether the target point cloud coordinates in the direction perpendicular to the historical driving direction are equal to the median of the point cloud coordinates, and obtain the fifth determination result; If the fifth judgment result indicates that the target point cloud coordinates are equal to the median of the point cloud coordinates, then the first preset weight is determined as the initial weight value of the target point cloud; If the fifth judgment result indicates that the target point cloud coordinates are not equal to the median of the point cloud coordinates, then it is determined whether the difference between the target point cloud coordinates and the median of the point cloud coordinates is greater than the width value of the preset multiple of the historical tracked target, and a sixth judgment result is obtained. If the sixth judgment result indicates that the difference value is greater than the width value of the preset multiple of the historical tracked target, then the second preset weight is determined as the initial weight value of the target point cloud; The second preset weight is greater than the first preset weight; If the sixth judgment result indicates that the difference value is less than or equal to the width value of the historical tracked target by a preset multiple, then the quotient of the width value of the historical tracked target and the difference value is determined as the specified weight value of the target point cloud; The specified weight values ​​of each target point cloud are normalized to obtain the initial weight values ​​of the target point cloud. The initial weight values ​​of each target point cloud are normalized to obtain the target weight values ​​of each target point cloud.

8. The method according to claim 7, wherein if the historical driving direction is horizontal, the target point cloud coordinates in the direction perpendicular to the historical driving direction are the point cloud ordinates; The step of determining the median point cloud coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historically tracked target, based on the point cloud data of the target point cloud, includes: Then, the ordinates of the point cloud are sorted in ascending order of the point cloud ordinates, or in descending order of the point cloud ordinates, to obtain a sequence of ordinates. The median of the ordinate sequence is determined to obtain the median of the ordinate of the point cloud.

9. The method according to claim 7, wherein if the historical driving direction is longitudinal, the target point cloud coordinates in the direction perpendicular to the historical driving direction are the abscissas of the point cloud; The step of determining the median point cloud coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historically tracked target, based on the point cloud data of the target point cloud, includes: The abscissas of each target point cloud are sorted in ascending order of point cloud abscissas, or in descending order of point cloud abscissas, to obtain a sequence of abscissas. The median of the x-coordinate sequence is determined to obtain the median x-coordinate of the point cloud.

10. The method according to claim 1, wherein generating a target tracking result for the historical tracking target based on the point cloud data of the target point cloud and the target weight value of the target point cloud includes: Target detection processing is performed based on the point cloud data of the target point cloud and the target weight value of the target point cloud to obtain the current target's position information and driving speed information; Obtain the historical location information and historical driving speed information of the historically tracked target; Based on the current target's location and speed information, and the historical location and speed information of the historical tracked target, a target tracking result is generated to reflect whether the current target and the historical tracked target are the same target.

11. The method of claim 10, further comprising: Based on the current target's position information and speed information, determine whether the current target is a lateral moving target in front of the radar device, and obtain the seventh determination result; If the seventh determination result indicates that the current target is a lateral moving target in front of the radar device, then the current target is determined as the tracking target of the forward traffic crossing warning function.

12. A method for selecting a tracking target for a forward traffic crossing warning function, comprising: Acquire point cloud data of targets that are correlated with historically tracked targets, collected by radar equipment; Based on the point cloud data of the target point cloud, determine the median point cloud coordinates of the target point cloud in the direction perpendicular to the historical driving direction of the historical tracked target; The target weight value of the target point cloud is calculated based on the difference between the target point cloud coordinates in the vertical direction of the historical driving direction and the median of the point cloud coordinates; wherein the target weight value is negatively correlated with the difference value. Target detection processing is performed based on the point cloud data of the target point cloud and the target weight value of the target point cloud to obtain the current target's position information and driving speed information; If the current target is determined to be a lateral moving target in front of the radar device based on the current target's position information and speed information, then the current target is identified as the tracking target for the forward traffic crossing warning function.

13. A domain controller, comprising: A processor and a memory; wherein the memory stores computer-readable instructions adapted to be loaded by the processor and to perform the steps of the method as claimed in any one of claims 1 to 12.

14. A computer program product having at least one instruction stored thereon, wherein the at least one instruction, when executed by a controller, implements the steps of the method according to any one of claims 1 to 12.

Images