Method and device for determining target category

Abstract

The invention discloses a target category determination method and device. The method comprises the steps that N frames of point cloud data containing a first target are acquired, N is an integer larger than 1, and the N frames of point cloud data are acquired through a vehicle-mounted radar; based on the N frames of point cloud data, determining a first height of the first target in each frame of point cloud data to obtain N first heights; determining a mean value and a variance of the N first heights; under the condition that the first target meets the first preset condition, the category of the first target is determined to be a first category, the first category represents that the height of the first target is higher than or equal to the target height and does not affect normal driving of the vehicle, and the vehicle is the vehicle where the vehicle-mounted radar is located; the first preset condition is that the difference value between each first height and the variance is greater than a first threshold value and the variance is greater than a second threshold value. Therefore, the size and the cost of the radar are not increased, and whether the target in the height direction can pass through or not can be accurately judged.

Description

Technical Field

[0001] The present application belongs to the field of vehicle electronic technology, and in particular, relates to a method and device for determining a target category. Background Art

[0002] Existing vehicle-mounted radars can detect and track surrounding vehicles, pedestrians, obstacles and other targets, making it easier to realize intelligent driving functions such as collision avoidance, automatic parking, blind spot monitoring, lane departure warning, etc. In addition, vehicle-mounted radars can also combine information obtained from other sensors, navigation systems and other technologies to realize more intelligent driving assistance functions such as adaptive cruise control, traffic sign recognition, and intelligent speed limit.

[0003] In order to achieve these functions, vehicle-mounted radar needs to have high angular accuracy and resolution. However, due to the limitations of radar size and cost, the number of antennas of a single radar is limited, and most transmitting antennas are used for horizontal angle measurement, resulting in a small number of antennas for pitch angle measurement, and poor pitch angle accuracy and resolution. Therefore, it is impossible to accurately determine whether the target in the height direction is a target that the vehicle can pass through, that is, whether it affects the normal driving of the vehicle.

[0004] Therefore, there is a need for a method that does not increase the size and cost of the radar while accurately determining whether a target in the height direction is traversable. Summary of the invention

[0005] The embodiments of the present application provide a method and device for determining a target category, which can accurately determine whether a target in a height direction is traversable without increasing the size and cost of the radar.

[0006] In a first aspect, an embodiment of the present application provides a method for determining a target category, the method comprising:

[0007] Acquire N frames of point cloud data containing the first target, where N is an integer greater than 1, and the N frames of point cloud data are collected by a vehicle-mounted radar;

[0008] Based on N frames of point cloud data, determine a first height of the first target in each frame of point cloud data to obtain N first heights;

[0009] Determine the mean and variance of the N first heights;

[0010] When the first target meets the first preset condition, the category of the first target is determined to be the first category, the first category indicates that the height of the first target is higher than or equal to the target height and does not affect the normal driving of the vehicle, the vehicle is the vehicle where the vehicle-mounted radar is located, and the first preset condition is that the difference between each first height and the variance is greater than the first threshold and the variance is greater than the second threshold.

[0011] In a second aspect, an embodiment of the present application provides a device for determining a target category, the device comprising:

[0012] A first acquisition module is used to acquire N frames of point cloud data containing a first target, where N is an integer greater than 1, and the N frames of point cloud data are collected by a vehicle-mounted radar;

[0013] A first determination module is used to determine a first height of a first target in each frame of point cloud data based on N frames of point cloud data, and obtain N first heights;

[0014] A second determination module, used to determine the mean and variance of N first heights;

[0015] The third determination module is used to determine that the category of the first target is the first category when the first target meets the first preset condition. The first category indicates that the height of the first target is higher than or equal to the target height and does not affect the normal driving of the vehicle. The vehicle is the vehicle where the vehicle-mounted radar is located. The first preset condition is that the difference between each first height and the variance is greater than the first threshold and the variance is greater than the second threshold.

[0016] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a processor and a memory storing computer program instructions, wherein when the processor executes the computer program instructions, the method for determining a target category as shown in any one of the embodiments of the first aspect is implemented.

[0017] In a fourth aspect, an embodiment of the present application provides a computer storage medium having computer program instructions stored thereon, and when the computer program instructions are executed by a processor, the method for determining a target category shown in any one of the embodiments of the first aspect is implemented.

[0018] In a fifth aspect, an embodiment of the present application provides a computer program product. When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device executes the method for determining a target category shown in any one of the embodiments of the first aspect.

[0019] The method and device for determining the target category of the embodiment of the present application can obtain N frames of point cloud data containing the first target collected by the vehicle-mounted radar, determine the first height of the first target in each frame of point cloud data based on the N frames of point cloud data, obtain N first heights, and then determine the mean and variance of the N first heights. When the difference between each first height and the variance is greater than the first threshold and the variance is greater than the second threshold, the category of the first target is determined to be the first category, and the first category indicates that the height of the first target is higher than or equal to the target height and does not affect the normal driving of the vehicle. In this way, it is possible to accurately determine whether the target affects the normal driving of the vehicle based on the variance and mean of the target's height in multiple frames of point cloud data, without increasing the number of radar antennas, thereby achieving the goal of not increasing the volume and cost of the radar, and accurately determining whether the target in the height direction is traversable. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solution of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0021] Figure 1 This is one of the flow charts of a method for determining a target category provided by an embodiment of the present application;

[0022] Figure 2 is a schematic diagram of a coordinate system provided by an embodiment of the present application;

[0023] Figure 3 is a schematic diagram of a pitch angle provided by an embodiment of the present application;

[0024] Figure 4 is a schematic diagram of a horizontal angle provided by an embodiment of the present application;

[0025] Figure 5 is a schematic diagram of a traversable discrimination module provided by an embodiment of the present application;

[0026] Figure 6 This is a second flowchart of a method for determining a target category provided by an embodiment of the present application;

[0027] Figure 7 is a schematic diagram of the structure of a device for determining a target category provided by an embodiment of the present application;

[0028] Figure 8 It is a structural schematic diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0029] The features and exemplary embodiments of various aspects of the present application will be described in detail below. In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application, rather than to limit the present application. For those skilled in the art, the present application can be implemented without the need for some of these specific details. The following description of the embodiments is only to provide a better understanding of the present application by illustrating the examples of the present application.

[0030] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "include..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0031] As mentioned in the background technology, the existing vehicle-mounted millimeter-wave radar can detect and track surrounding vehicles, pedestrians, obstacles and other objects, and is used to realize intelligent driving functions, such as anti-collision function, automatic parking function, blind spot monitoring function, lane departure warning function, etc. In addition, the vehicle-mounted millimeter-wave radar can also combine the information obtained by other sensors, navigation systems and other technologies to realize more intelligent driving assistance functions, such as adaptive cruise, traffic sign recognition, intelligent speed limit, etc. However, in order to realize these functions, the vehicle-mounted millimeter-wave radar needs to have higher angular accuracy and resolution, but due to the size, technology and cost limitations of the radar, the number of antennas of a single radar is limited (generally 3T4R or 2T4R), and most of the radar transmitting antennas are used as horizontal angle measurements, resulting in a small number of antennas for pitch angle measurement, poor pitch angle accuracy and resolution, and inaccurate height recognition of traversable objects such as high-altitude targets and ground targets, thereby causing false alarms of the intelligent driving function. Currently, the common methods to improve radar target penetration discrimination are mainly to increase the number of elevation angle measurement antennas or to compensate with other sensors. For the method of increasing the number of elevation angle measurement antennas, since the antenna spacing is fixed, the radar size will become larger, which is not conducive to the layout and installation of the radar on the vehicle, and the cost is relatively high. For the method of adding compensation with other sensors, additional sensor information is required, which increases the cost of radar target penetration discrimination and increases the complexity of the radar system and its design difficulty.

[0032] Based on this, an embodiment of the present application provides a method for determining a target category. The method for determining a target category provided by an embodiment of the present application is introduced below.

[0033] Figure 1 A schematic flow chart of a method for determining a target category provided in an embodiment of the present application is shown.

[0034] like Figure 1 As shown, the target category determination method may be executed by a target category determination device, and the target category determination method may include the following steps:

[0035] S110, acquiring N frames of point cloud data including the first target;

[0036] S120, based on N frames of point cloud data, determining a first height of the first target in each frame of point cloud data to obtain N first heights;

[0037] S130, determining the mean and variance of N first heights;

[0038] S140: When the first target satisfies a first preset condition, determine that the category of the first target is a first category.

[0039] Thus, N frames of point cloud data containing the first target collected by the vehicle-mounted radar can be obtained. Based on the N frames of point cloud data, the first height of the first target in each frame of point cloud data is determined to obtain N first heights, and then the mean and variance of the N first heights are determined. When the difference between each first height and the variance is greater than the first threshold and the variance is greater than the second threshold, the category of the first target is determined to be the first category, and the first category indicates that the height of the first target is higher than or equal to the target height and does not affect the normal driving of the vehicle. In this way, it is possible to accurately judge whether the target affects the normal driving of the vehicle based on the variance and mean of the height of the target in multiple frames of point cloud data, without increasing the number of radar antennas, thereby achieving the goal of not increasing the volume and cost of the radar, and accurately judging whether the target in the height direction is traversable.

[0040] In S110 , N may be an integer greater than 1, and the N frames of point cloud data may be collected by a vehicle-mounted radar. For example, the vehicle-mounted radar may be a vehicle-mounted millimeter-wave radar.

[0041] The first target may be any target in the first frame of point cloud data of N frames of point cloud data.

[0042] In some implementations, in order to more accurately determine the category of the target, S110 may include:

[0043] Acquire the first point cloud data;

[0044] Determine a second horizontal coordinate, a second vertical coordinate, and a second height of each object in the first point cloud data;

[0045] Removing the target satisfying the sixth preset condition from the first point cloud data to obtain second point cloud data;

[0046] The second point cloud data is used as the first frame of point cloud data, and N frames of point cloud data including the first target are obtained.

[0047] Here, the second point cloud data may include the first object.

[0048] The sixth preset condition may be that the second horizontal coordinate of the target is greater than the tenth threshold, the second vertical coordinate is greater than the eleventh threshold, and the target meets the stable tracking condition.

[0049] The tenth threshold value can be set according to actual needs and is not limited here. For example, the tenth threshold value can be 1.5 times the lane width. The second horizontal coordinate being greater than the tenth threshold value can indicate that the target is far away from the vehicle in the left and right directions of the vehicle and will not affect the normal driving of the vehicle.

[0050] The eleventh threshold value may be the pitch misalignment distance of the vehicle-mounted radar, which may be set according to actual needs and is not limited here. If the second ordinate is greater than the eleventh threshold value, it may indicate that the target is far away from the vehicle in the front and rear direction of the vehicle, and the point cloud data of the target collected by the vehicle-mounted radar may be inaccurate.

[0051] The stable tracking condition may be that the target has M position changes that meet the seventh preset condition, and M may be a positive integer.

[0052] The seventh preset condition may be that a change in the horizontal coordinate of the target in two adjacent frames of point cloud data is less than a twelfth threshold, a change in the vertical coordinate is less than a thirteenth threshold, and a change in height is less than a fourteenth threshold.

[0053] The twelfth threshold, the thirteenth threshold and the fourteenth threshold can all be set according to actual needs and are not limited here.

[0054] For example, if the position of the target in two adjacent frames of point cloud data (x i,n ,y i,n ,z i,n ) and (x i,n+1 ,y i,n+1 ,z i,n+1 ) meets the associated threshold requirement (Δx i ,Δy i ,Δz i )=(|x i,n+1 -x i,n |,|y i,n+1 -y i,n |,|z i,n+1 -zi,n |)<(σ x ,σ y ,σ z ), it is recorded as a valid association. If M valid associations are met, the stable tracking condition is met.

[0055] Among them, i is the number of the target, n is the number of the point cloud data frame, and x i,n is the second horizontal coordinate of target i in the point cloud data of the nth frame, y i,n is the second ordinate of target i in the point cloud data of the nth frame, z i,n is the second height of target i in the point cloud data of the nth frame, x i,n+1 The second horizontal coordinate of target i in the point cloud data of the n+1th frame, y i,n+1 is the second ordinate of target i in the point cloud data of the n+1th frame, z i,n+1 Δx is the second height of target i in the point cloud data of the n+1th frame. i is the change in the horizontal coordinate of target i, Δy i is the change in the ordinate of target i, Δz i is the height change of target i. x is the twelfth threshold, σ y is the thirteenth threshold, σ z is the fourteenth threshold.

[0056] Specifically, if a target meets the sixth preset condition, it indicates that the accuracy of the point cloud data of the target is low and there is no need to determine the category, or the target is far away from the vehicle and will not affect the normal driving of the vehicle, so there is no need to determine the category. Therefore, the targets that meet the sixth preset condition can be removed, and the category detection can be performed on the remaining targets. Any target among the remaining targets can be used as the first target to determine the category of the first target.

[0057] For the first target, after determining the first frame of point cloud data, N-1 frames including the first target point cloud data can be taken backward to obtain N frames of point cloud data containing the first target. In the point cloud data after the first frame of point cloud data, not every frame necessarily contains the first target, and the point cloud data that does not contain the first target can be removed. Therefore, the N frames of point cloud data containing the first target are not necessarily continuous N frames.

[0058] In this way, by removing the targets that meet the sixth preset condition, the first target whose category is to be determined can be obtained, and the point cloud data required to determine the category of the first target can be accurately acquired, so that the category of the target can be determined more accurately.

[0059] Regarding S120 , the first height may be a z-axis coordinate of the first object.

[0060] In some implementations, in order to more accurately determine the first height of the first target in each frame of point cloud data, the above-mentioned determination of the first height of the first target in each frame of point cloud data may include:

[0061] The following steps are performed for each frame of point cloud data to obtain the first height of the first target in each frame of point cloud data:

[0062] Determine a first distance between the first target and the origin of the point cloud coordinate system and a pitch angle of the first target based on the point cloud data;

[0063] A first height of the first target in the point cloud data is determined based on the first distance and the pitch angle.

[0064] Here, if Figure 2 As shown, the point cloud coordinate system can be such that the center of the vehicle-mounted radar is the origin, the vehicle's moving direction is the positive direction of the y-axis, the horizontal direction perpendicular to the vehicle's moving direction and to the left is the positive direction of the x-axis, and the vertical direction perpendicular to the vehicle's moving direction and upward is the positive direction of the z-axis.

[0065] The pitch angle may be an angle between the first projection and the y-axis in the point cloud coordinate system, and the first projection may be a projection of a line connecting the first target and the origin on the yz plane in the point cloud coordinate system.

[0066] like Figure 3 As shown, the first distance between the first target and the origin of the point cloud coordinate system is R1, and the pitch angle of the first target is θ yz .

[0067] Exemplarily, the calculation formula of the first height z may be: z=R1*sinθ yz .

[0068] In this way, the first height of the first target in each frame of point cloud data can be determined more accurately through the above process.

[0069] The method for determining the height of each target in each frame of point cloud data may be the same as the method for determining the first height.

[0070] In S130, the calculation formula of the mean value Mean_z of N first heights may be:

[0071]

[0072] The calculation formula of the variance Var_z of N first heights can be:

[0073]

[0074] Regarding S140 , the first preset condition may be that the difference between each first height and the variance is greater than a first threshold and the variance is greater than a second threshold.

[0075] The first threshold may be a variance judgment threshold, which may be set according to actual needs and is not limited here.

[0076] The second threshold may be a high altitude mean judgment threshold, which may be set according to actual needs and is not limited here.

[0077] The first category may indicate that the height of the first target is higher than or equal to the target height and does not affect the normal driving of the vehicle. The target height can be set according to actual needs and is not limited here. The first category may indicate that the first target is a high-altitude traversable target.

[0078] The vehicle may be a vehicle where the vehicle-mounted radar is located.

[0079] In some embodiments, in order to accurately determine the category of the first target, the method may further include:

[0080] In the case where the first target does not meet the first preset condition, based on N frames of point cloud data, determine the first ordinate of the first target in each frame of point cloud data, obtain the first vehicle speed corresponding to each frame of point cloud data of the first target, and based on the N frames of point cloud data, determine the difference between the abscissa of the first target in the nth frame of point cloud data and the abscissa of the first target in the n+1th frame of point cloud data, and obtain N-1 lateral displacement differences, n=1, 2...N;

[0081] When the first target meets the second preset condition, determining that the first target is a pedestrian;

[0082] When the first target does not satisfy the second preset condition, it is determined that the first target is not a pedestrian.

[0083] Here, if the first target does not meet the first preset condition, it can be indicated that the category of the first target is not the first category, and at this time, it can be further determined whether the first target is a pedestrian. If the first target meets the second preset condition, it can be determined that the first target is a pedestrian; if the first target does not meet the second preset condition, it can be determined that the first target is not a pedestrian.

[0084] The second preset condition may be that each first vehicle speed is less than a third threshold, each first ordinate is less than a fourth threshold, N is greater than a fifth threshold, and each lateral displacement difference is greater than a sixth threshold.

[0085] The third threshold may be a vehicle low speed threshold, which may be set according to actual needs and is not limited here.

[0086] The fourth threshold may be a pedestrian collision minimum threshold, which may be set according to actual needs and is not limited here.

[0087] The fifth threshold may be a target information stability threshold, which may be set according to actual needs and is not limited here.

[0088] The sixth threshold may be a monotonicity determination threshold, which may be set according to actual needs and is not limited here.

[0089] The fact that each first vehicle speed is less than the third threshold value may indicate that the vehicle is not traveling on a highway or other place where pedestrians will not appear, so the first target may be a pedestrian. In addition, the first vehicle speed may be limited to be greater than 0, indicating that the vehicle does not need to determine the category of the first target when stopping during driving.

[0090] Each of the first ordinates being less than the fourth threshold may indicate that the vehicle is likely to collide with the first target.

[0091] N being greater than the fifth threshold may indicate that the data of the target collected by the radar is stable and reliable.

[0092] The fact that each lateral displacement difference is greater than the sixth threshold value may indicate that the lateral displacement difference of the target is monotonic, and may indicate that the first target continues to move in one direction in the lateral direction.

[0093] In this way, through the above process, when the first target does not meet the first preset condition, that is, does not belong to the first category, it can continue to be determined whether the first target is a pedestrian, so as to accurately determine the category of the first target.

[0094] In some implementations, in order to improve the efficiency of target category determination, the method may further include:

[0095] When the first target does not meet the first preset condition, based on the N frames of point cloud data, determining a first ordinate of the first target in each frame of point cloud data, and obtaining a first vehicle speed corresponding to each frame of point cloud data of the first target;

[0096] When the first target satisfies the third preset condition, based on N frames of point cloud data, determine the difference between the horizontal coordinate of the first target in the nth frame of point cloud data and the horizontal coordinate of the first target in the n+1th frame of point cloud data, and obtain N-1 lateral displacement differences, n=1, 2...N;

[0097] When the first target satisfies a fourth preset condition, determining that the first target is a pedestrian;

[0098] When the first target does not satisfy the fourth preset condition, it is determined that the first target is not a pedestrian.

[0099] Here, the first target does not meet the first preset condition, which may indicate that the category of the first target is not the first category, and at this time, it may be further determined whether it is necessary to determine whether the first target is a pedestrian. If the first target meets the third preset condition, it is necessary; if the first target does not meet the third preset condition, it is not necessary.

[0100] If not, it is not necessary to determine whether the first target is a pedestrian. If necessary, it is possible to determine whether the first target meets the fourth preset condition. If so, it is possible to determine that the first target is a pedestrian; if not, it is possible to determine that the first target is not a pedestrian.

[0101] The third preset condition may be that each first vehicle speed is less than the third threshold, each first ordinate is less than the fourth threshold, and N is greater than the fifth threshold.

[0102] The fourth preset condition may be that each lateral displacement difference is greater than a sixth threshold.

[0103] For example, the third preset condition may be V car,n <thres_V min And y i,n <thres_R ped And N>thres_T. Among them, V car,n is the first vehicle speed, thres_V min is the third threshold, y i,n is the first ordinate, thres_R ped is the fourth threshold, and thres_T is the fifth threshold.

[0104] The fourth threshold value may be Δx n,n+1 =x i,n -x i,n+1 >thres x . Among them, Δx n,n+1 is the lateral displacement difference, x i,n is the horizontal coordinate of target i in the nth frame point cloud data, x i,n+1 is the horizontal coordinate of target i in the point cloud data of the n+1th frame, thres x is the sixth threshold.

[0105] In this way, it can be determined first whether it is necessary to determine whether the first target is a pedestrian. If it is necessary, it can avoid invalid judgment and waste of time.

[0106] In some implementations, in order to more accurately determine the first ordinate of the first target in each frame of point cloud data, the above-mentioned determination of the first ordinate of the first target in each frame of point cloud data may include:

[0107] The following steps are performed for each frame of point cloud data to obtain the first vertical coordinate of the first target in each frame of point cloud data:

[0108] determining a second distance between the first target and the origin of the point cloud coordinate system and a horizontal angle of the first target based on the point cloud data;

[0109] A first longitudinal coordinate of the first target in the point cloud data is determined based on the second distance and the horizontal angle.

[0110] Here, the horizontal angle may be the angle between the second projection and the y-axis in the point cloud coordinate system, and the second projection may be the projection of the line connecting the first target and the origin on the xy plane in the point cloud coordinate system.

[0111] like Figure 4 As shown, the second distance between the first target and the origin of the point cloud coordinate system is R2, and the horizontal angle of the first target is θ xy .

[0112] For example, the calculation formula of the first ordinate y may be: y=R2*cosθ xy .

[0113] In this way, the first vertical coordinate of the first target in each frame of point cloud data can be determined more accurately through the above process.

[0114] The method for determining the ordinate of each target in each frame of point cloud data may be the same as the method for determining the first ordinate.

[0115] In some implementations, determining the horizontal coordinate of the first target in the nth frame of point cloud data may include:

[0116] Determine a third distance between the first target and the origin of the point cloud coordinate system and a horizontal angle of the first target based on the nth frame of point cloud data;

[0117] The horizontal coordinate of the first target in the nth frame of point cloud data is determined based on the third distance and the horizontal angle.

[0118] Here, the horizontal angle may be the angle between the second projection and the y-axis in the point cloud coordinate system, and the second projection may be the projection of the line connecting the first target and the origin on the xy plane in the point cloud coordinate system.

[0119] For example, the calculation formula of the horizontal coordinate x of the first target in the nth frame of point cloud data can be: x=R3*sinθ xy . Among them, R3 is the third distance.

[0120] In this way, the horizontal coordinate of the first target in the nth frame of point cloud data can be determined more accurately through the above process.

[0121] The method for determining the horizontal coordinate of each target in each frame of point cloud data can be the same as the method for determining the horizontal coordinate of the first target in the nth frame of point cloud data. For example, the method for determining the horizontal coordinate of the first target in the n+1th frame of point cloud data can be the same as the method for determining the horizontal coordinate of the first target in the nth frame of point cloud data, which will not be repeated here.

[0122] In some implementations, the vehicle-mounted radar system may be initialized to initialize various parameters of the vehicle-mounted radar system. Then the vehicle-mounted radar system is started to collect the distance between the target and the radar, that is, the distance R between the target and the origin. i , and the horizontal angle θ of the target xy,i and pitch angle θ yz,i .

[0123] In some embodiments, in order to accurately determine the category of the target, the method may further include:

[0124] When the first target does not satisfy the first preset condition, obtaining a first intensity and a reflection cross-sectional area (RCS) of the radio wave reflected by the first target to the vehicle-mounted radar;

[0125] When the first target satisfies the fifth preset condition, determining that the category of the first target is the second category;

[0126] In a case where the first object does not satisfy the fifth preset condition, it is determined that the category of the first object is not the second category.

[0127] Here, if the first target does not meet the first preset condition, it can be indicated that the category of the first target is not the first category, and at this time, it can be further determined whether the category of the first target is the second category. If the first target meets the fifth preset condition, it can be determined that the category of the first target is the second category; if the first target does not meet the fifth preset condition, it can be determined that the category of the first target is not the second category.

[0128] The fifth preset condition may be that the first intensity is greater than the seventh threshold, or the fifth preset condition may be that the RCS is greater than the eighth threshold and the mean is greater than the ninth threshold.

[0129] The seventh threshold may be a strength judgment threshold, which may be set according to actual needs and is not limited here.

[0130] The eighth threshold can be set according to actual needs and is not limited here.

[0131] The ninth threshold may be a ground mean judgment threshold, which may be set according to actual needs and is not limited here.

[0132] Exemplarily, if any of the following conditions is met, it can be determined that the category of the first target is the second category:

[0133] (1) Strength i >thres_Strength, where Strength i is the first strength, and thres_Strength is the seventh threshold.

[0134] (2)RCS i >thres_RCS and Mean_z>thres_mean_L, where RCS i is RCS, thres_RCS is the eighth threshold, Mean_z is the mean, and thres_mean_L is the ninth threshold.

[0135] The second category may indicate that the height of the first target is lower than or equal to the target height and affects the normal driving of the vehicle, and the second category may indicate that the first target is a ground non-passable target.

[0136] In this way, through the above process, when the first target does not meet the first preset condition, that is, does not belong to the first category, it can continue to be judged whether the category of the first target is the second category, so as to accurately determine the category of the first target.

[0137] In some embodiments, in order to accurately determine the category of the target, the method may further include:

[0138] When the first object is not a pedestrian and the category of the first object is not the second category, the category of the first object is determined to be the third category.

[0139] Here, if the category of the first target is neither the first category nor the second category, and the first target is not a pedestrian, it can be determined that the category of the first target is the third category.

[0140] The third category may represent that the height of the first target is lower than or equal to the target height and does not affect the normal driving of the vehicle, and the third category may represent that the first target is a ground traversable target.

[0141] In this way, through the above process, the elimination method can be used to determine whether the category of the first target is the third category, so as to accurately determine the category of the first target.

[0142] In some embodiments, the category of the first target can be determined by a traversable discrimination module, such as Figure 5 As shown, the input of the traversable discrimination module may include the horizontal coordinate, vertical coordinate, altitude, intensity, RCS and vehicle speed of the first target, and the output of the traversable discrimination module may be the category of the first target.

[0143] Specifically, the parameters of the traversable discrimination module can be initialized to establish the high-altitude target statistical matrix h i and accumulator h_count i , ground non-traversable target statistical matrix m i and accumulator m_count i , ground traversable target statistical matrix l i and accumulator l_count i .

[0144] After initialization, h i =[0,0,…,0] i×10 , h_count i =0,m i =[0,0,…,0] i×10 , m_count i =0, l i =[0,0,…,0] i×10 , l_count i =0.

[0145] Then, the first frame point cloud data is stored. For different targets in the first frame point cloud data, statistical matrices of different heights are stored according to the height of the target. Specifically, if the height z of target i is i >H l,h , then store it in h i In the above equation, we get h i =[z i ,0,…,0] i×10 ,h_count i =1, and the height of the target i in the subsequent point cloud data is stored in h i If the height z of target i is i ≤H l,h And z i ≥H m,l , then store in m i In the equation, we get m i =[z i ,0,…,0] i×10 ,m_count i =1, and the height of the target i in the subsequent point cloud data is stored in the m i If the height z of target i is i <H m,l , then store in l i In the i =[z i ,0,…,0] i×10 ,l_count i =1, and the height of the target i in the subsequent point cloud data is stored in the l iAmong them, H l,h H is the minimum height of the high-altitude target that the vehicle can pass through, m,l The maximum height of a ground target that can be traversed by the vehicle.

[0146] The accumulator h_count corresponding to each statistical matrix i ,m_count i ,l_count i Accumulated according to the number of frames n.

[0147] If each statistical matrix can store the height corresponding to N frames of point cloud data at most, then when storing the height corresponding to the N+kth frame of point cloud data, the value in the statistical matrix can be shifted to the left by k values, and the new value is recorded in the last digit, as shown below with h i Take the operation as an example: h i =[z i,1 ,z i,2 ,…,z i,N ] i×10 , after storing the height corresponding to the point cloud data of the N+1th frame, it becomes [z i,2 ,z i,3 ,…,z i,N+1 ] i×10 .

[0148] When a target disappears, the statistical matrix corresponding to the target can be cleared.

[0149] In order to better describe the entire solution, based on the above embodiments, a specific example is given. Figure 6 As shown, the method for determining the target category may include S601-S612, which will be explained in detail below.

[0150] S601, obtaining N frames of point cloud data containing a first target.

[0151] S602, determining a first height of a first target in each frame of point cloud data, and obtaining N first heights.

[0152] S603, determine the mean and variance of the N first heights.

[0153] S604: Determine whether the first target satisfies a first preset condition.

[0154] If yes, execute S605; if no, execute S606.

[0155] S605: Determine that the category of the first target is the first category.

[0156] S606, determining a first ordinate of the first target in each frame of point cloud data, obtaining a first vehicle speed corresponding to each frame of point cloud data of the first target, and determining N-1 lateral displacement differences of the first target.

[0157] S607: Determine whether the first target satisfies a second preset condition.

[0158] If yes, execute S608; if no, execute S609.

[0159] S608: Determine that the first target is a pedestrian.

[0160] S609: Acquire a first intensity and RCS of the first target.

[0161] S610: Determine whether the first target satisfies a fifth preset condition.

[0162] If yes, execute S611; if no, execute S612.

[0163] S611: Determine that the category of the first target is the second category.

[0164] S612: Determine that the category of the first target is the third category.

[0165] The specific process of each step can be found in the above embodiment and will not be described again here.

[0166] In an embodiment of the present application, a very small number of elevation angle measurement antennas can be used to statistically analyze the height information measured by the radar for targets at different positions, perform threshold judgments on the variance and mean, and combine the analysis of the target's movement trend, echo intensity and RCS value to determine the target's traversable properties.

[0167] Based on the same inventive concept, the embodiment of the present application also provides a device for determining a target category. Figure 7 The target category determination device provided in the embodiment of the present application is described in detail.

[0168] Figure 7 A schematic diagram of the structure of a device for determining a target category provided by an embodiment of the present application is shown.

[0169] like Figure 7 As shown, the target category determination device may include:

[0170] A first acquisition module 701 is used to acquire N frames of point cloud data containing a first target, where N is an integer greater than 1, and the N frames of point cloud data are collected by a vehicle-mounted radar;

[0171] A first determination module 702 is used to determine a first height of a first target in each frame of point cloud data based on N frames of point cloud data, and obtain N first heights;

[0172] The second determination module 703 is used to determine the mean and variance of the N first heights;

[0173] The third determination module 704 is used to determine that the category of the first target is the first category when the first target meets the first preset condition. The first category indicates that the height of the first target is higher than or equal to the target height and does not affect the normal driving of the vehicle. The vehicle is the vehicle where the vehicle-mounted radar is located. The first preset condition is that the difference between each first height and the variance is greater than the first threshold and the variance is greater than the second threshold.

[0174] Thus, N frames of point cloud data containing the first target collected by the vehicle-mounted radar can be obtained. Based on the N frames of point cloud data, the first height of the first target in each frame of point cloud data is determined to obtain N first heights, and then the mean and variance of the N first heights are determined. When the difference between each first height and the variance is greater than the first threshold and the variance is greater than the second threshold, the category of the first target is determined to be the first category, and the first category indicates that the height of the first target is higher than or equal to the target height and does not affect the normal driving of the vehicle. In this way, it is possible to accurately judge whether the target affects the normal driving of the vehicle based on the variance and mean of the height of the target in multiple frames of point cloud data, without increasing the number of radar antennas, thereby achieving the goal of not increasing the volume and cost of the radar, and accurately judging whether the target in the height direction is traversable.

[0175] In some embodiments, in order to accurately determine the category of the first target, the device may further include:

[0176] a fourth determination module, for determining, based on N frames of point cloud data, a first ordinate of the first target in each frame of point cloud data, obtaining a first vehicle speed corresponding to each frame of point cloud data of the first target, and determining, based on the N frames of point cloud data, a difference between a horizontal coordinate of the first target in the nth frame of point cloud data and a horizontal coordinate of the first target in the n+1th frame of point cloud data, to obtain N-1 lateral displacement differences, where n=1, 2…N;

[0177] a fifth determination module, configured to determine that the first target is a pedestrian when the first target satisfies a second preset condition, wherein the second preset condition is that each first vehicle speed is less than a third threshold, each first ordinate is less than a fourth threshold, N is greater than a fifth threshold, and each lateral displacement difference is greater than a sixth threshold;

[0178] The sixth determination module is used to determine that the first target is not a pedestrian when the first target does not meet the second preset condition.

[0179] In some embodiments, in order to improve the efficiency of target category determination, the device may further include:

[0180] a seventh determination module, configured to determine, based on N frames of point cloud data, a first ordinate of the first target in each frame of point cloud data, and obtain a first vehicle speed corresponding to each frame of point cloud data of the first target when the first target does not meet the first preset condition;

[0181] an eighth determination module, for determining, based on N frames of point cloud data, a difference between a horizontal coordinate of the first target in the nth frame of point cloud data and a horizontal coordinate of the first target in the n+1th frame of point cloud data, to obtain N-1 lateral displacement differences, n=1, 2…N, when the first target satisfies a third preset condition, and each first vertical coordinate is less than a fourth threshold, and N is greater than a fifth threshold;

[0182] A ninth determination module, configured to determine that the first target is a pedestrian if the first target satisfies a fourth preset condition, wherein the fourth preset condition is that each lateral displacement difference is greater than a sixth threshold;

[0183] The tenth determination module is used to determine that the first target is not a pedestrian when the first target does not meet the fourth preset condition.

[0184] In some embodiments, in order to accurately determine the category of the target, the device may further include:

[0185] A second acquisition module is used to acquire a first intensity and a reflection cross-sectional area RCS of the radio wave reflected by the first target to the vehicle-mounted radar when the first target does not meet the first preset condition;

[0186] an eleventh determination module, configured to determine that the category of the first target is a second category when the first target satisfies a fifth preset condition, the second category indicating that the height of the first target is lower than or equal to the target height and affects normal driving of the vehicle, the fifth preset condition being that the first intensity is greater than a seventh threshold, or the fifth preset condition being that the RCS is greater than an eighth threshold and the mean is greater than a ninth threshold;

[0187] A twelfth determining module is configured to determine that the category of the first target is not the second category when the first target does not satisfy the fifth preset condition.

[0188] In some embodiments, in order to accurately determine the category of the target, the device may further include:

[0189] The thirteenth determination module is used to determine that the category of the first target is a third category when the first target is not a pedestrian and the category of the first target is not the second category, and the third category indicates that the height of the first target is lower than or equal to the target height and does not affect the normal driving of the vehicle.

[0190] In some implementations, in order to more accurately determine the category of the target, the first acquisition module 701 may include:

[0191] A first acquisition submodule, used to acquire first point cloud data;

[0192] A determination submodule, used to determine a second horizontal coordinate, a second vertical coordinate, and a second height of each target in the first point cloud data;

[0193] a removal submodule, used to remove the target satisfying the sixth preset condition from the first point cloud data to obtain the second point cloud data, the sixth preset condition being that the second horizontal coordinate of the target is greater than the tenth threshold, the second vertical coordinate is greater than the eleventh threshold, and the target satisfies the stable tracking condition, the stable tracking condition being that the target has M position changes satisfying the seventh preset condition, M is a positive integer, the seventh preset condition being that the horizontal coordinate change amount of the target in two adjacent frames of point cloud data is less than the twelfth threshold, the vertical coordinate change amount is less than the thirteenth threshold, and the height change amount is less than the fourteenth threshold, and the second point cloud data includes the first target;

[0194] The second acquisition submodule is used to acquire N frames of point cloud data including the first target using the second point cloud data as the first frame of point cloud data.

[0195] In some implementations, in order to more accurately determine the first target, the first determination module 702 may be specifically configured to:

[0196] The following steps are performed for each frame of point cloud data to obtain the first height of the first target in each frame of point cloud data:

[0197] Determine a first distance between the first target and the origin of the point cloud coordinate system and a pitch angle of the first target based on the point cloud data, wherein the pitch angle is an angle between the first projection and the y-axis in the point cloud coordinate system, and the first projection is a projection of a line connecting the first target and the origin on a yz plane in the point cloud coordinate system;

[0198] A first height of the first target in the point cloud data is determined based on the first distance and the pitch angle.

[0199] In some implementations, in order to more accurately determine the first ordinate of the first target in each frame of point cloud data, the fourth determination module or the seventh determination module may be specifically used to:

[0200] The following steps are performed for each frame of point cloud data to obtain the first vertical coordinate of the first target in each frame of point cloud data:

[0201] Determine a second distance between the first target and the origin of the point cloud coordinate system and a horizontal angle of the first target based on the point cloud data, the horizontal angle being the angle between the second projection and the y-axis in the point cloud coordinate system, and the second projection being the projection of a line connecting the first target and the origin on an xy plane in the point cloud coordinate system;

[0202] A first longitudinal coordinate of the first target in the point cloud data is determined based on the second distance and the horizontal angle.

[0203] In some implementations, in order to more accurately determine the horizontal coordinate of the first target in the nth frame of point cloud data, the fourth determination module or the eighth determination module may be specifically used to:

[0204] Determine a third distance between the first target and the origin of the point cloud coordinate system and a horizontal angle of the first target based on the nth frame of point cloud data, the horizontal angle being the angle between the second projection and the y-axis in the point cloud coordinate system, the second projection being the projection of a line connecting the first target and the origin on an xy plane in the point cloud coordinate system;

[0205] The horizontal coordinate of the first target in the nth frame of point cloud data is determined based on the third distance and the horizontal angle.

[0206] Figure 8 A schematic structural diagram of an electronic device provided by an embodiment of the present application is shown.

[0207] like Figure 8 As shown, the electronic device 8 is a structural diagram of an exemplary hardware architecture of an electronic device that can implement the target category determination method and target category determination device in the embodiments of the present application. The electronic device may refer to the electronic device in the embodiments of the present application.

[0208] The electronic device 8 may include a processor 801 and a memory 802 storing computer program instructions.

[0209] Specifically, the processor 801 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0210] The memory 802 may include a large capacity memory for data or instructions. By way of example and not limitation, the memory 802 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a tape, or a universal serial bus (USB) drive or a combination of two or more of these. Where appropriate, the memory 802 may include a removable or non-removable (or fixed) medium. Where appropriate, the memory 802 may be inside or outside the integrated gateway disaster recovery device. In a specific embodiment, the memory 802 is a non-volatile solid-state memory. In a specific embodiment, the memory 802 may include a read-only memory (ROM), a random access memory (RAM), a disk storage medium device, an optical storage medium device, a flash memory device, an electrical, optical, or other physical / tangible memory storage device. Therefore, typically, the memory 802 includes one or more tangible (non-transitory) computer-readable storage media (e.g., a memory device) encoded with software including computer executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to one aspect of the present application.

[0211] The processor 801 implements any one of the methods for determining a target category in the above embodiments by reading and executing computer program instructions stored in the memory 802 .

[0212] In one example, the electronic device may further include a communication interface 803 and a bus 804. Figure 8 As shown, the processor 801, the memory 802, and the communication interface 803 are connected via a bus 804 and communicate with each other.

[0213] The communication interface 803 is mainly used to implement communication between various modules, devices, units and / or equipment in the embodiments of the present application.

[0214] Bus 804 includes hardware, software or both, and the parts of electronic equipment are coupled to each other.For example, but not limitation, bus may include accelerated graphics port (AGP) or other graphics bus, enhanced industrial standard architecture (EISA) bus, front side bus (FSB), hypertransport (HT) interconnection, industrial standard architecture (ISA) bus, infinite bandwidth interconnection, low pin count (LPC) bus, memory bus, micro channel architecture (MCA) bus, peripheral component interconnection (PCI) bus, PCI-Express (PCI-X) bus, serial advanced technology attachment (SATA) bus, video electronics standard association local (VLB) bus or other suitable bus or two or more of these combinations. In appropriate cases, bus 804 may include one or more buses. Although the present application embodiment describes and shows a specific bus, the application considers any suitable bus or interconnection.

[0215] The electronic device can execute the target category determination method in the embodiment of the present application, thereby realizing the combination Figures 1 to 7 A method and apparatus for determining a target category are described.

[0216] In addition, in combination with the target category determination method in the above embodiment, the present application embodiment can provide a computer storage medium for implementation. The computer storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, any target category determination method in the above embodiment is implemented.

[0217] It should be clear that the present application is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present application is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present application.

[0218] The functional blocks shown in the above-described block diagram can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present application are programs or code segments that are used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link by a data signal carried in a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency (RF) links, etc. The code segment can be downloaded via a computer network such as the Internet, an intranet, etc.

[0219] It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, this application is not limited to the order of the above steps, that is, the steps can be performed in the order mentioned in the embodiment, or in a different order from the embodiment, or several steps can be performed simultaneously.

[0220] The above reference is according to the method of the embodiment of the present application, the flow chart of the device (system) and the computer program product and / or the block diagram described various aspects of the present application.It should be understood that each square box in the flow chart and / or the block diagram and the combination of each square box in the flow chart and / or the block diagram can be realized by computer program instructions.These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer or other programmable data processing device to produce a machine so that these instructions executed by the processor of the computer or other programmable data processing device enable the realization of the function / action specified in one or more square boxes of the flow chart and / or the block diagram.Such a processor can be but is not limited to a general-purpose processor, a special-purpose processor, a special application processor or a field programmable logic circuit.It can also be understood that each square box in the block diagram and / or the flow chart and the combination of the square boxes in the block diagram and / or the flow chart can also be realized by the dedicated hardware that performs the specified function or action, or can be realized by the combination of dedicated hardware and computer instructions.

[0221] The above is only a specific implementation of the present application. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the protection scope of the present application is not limited to this. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in this application, and these modifications or replacements should be included in the protection scope of this application.

Claims

1. A method for determining a target category, characterized in that: The method comprises: Acquire N frames of point cloud data containing the first target, where N is an integer greater than 1, and the N frames of point cloud data are collected by a vehicle-mounted radar; Based on the N frames of point cloud data, determining a first height of the first target in each frame of point cloud data to obtain N first heights; Determine the mean and variance of the N first heights; When the first target satisfies a first preset condition, the category of the first target is determined to be a first category, the first category characterizing that the height of the first target is higher than or equal to the target height and does not affect the normal driving of the vehicle, the vehicle is the vehicle where the vehicle-mounted radar is located, and the first preset condition is that the difference between each of the first heights and the variance is greater than a first threshold and the variance is greater than a second threshold.

2. The method for determining a target category according to claim 1, characterized in that: The method further comprises: In the case that the first target does not meet the first preset condition, based on the N frames of point cloud data, determine the first ordinate of the first target in each frame of point cloud data, obtain the first vehicle speed corresponding to each frame of point cloud data of the first target, and based on the N frames of point cloud data, determine the difference between the abscissa of the first target in the nth frame of point cloud data and the abscissa of the first target in the n+1th frame of point cloud data, and obtain N-1 lateral displacement differences, n=1, 2...N; When the first target satisfies a second preset condition, determining that the first target is a pedestrian, the second preset condition being that each of the first vehicle speeds is less than a third threshold, each of the first ordinates is less than a fourth threshold, N is greater than a fifth threshold, and each of the lateral displacement differences is greater than a sixth threshold; When the first target does not satisfy the second preset condition, it is determined that the first target is not a pedestrian.

3. The method for determining a target category according to claim 1, characterized in that: The method further comprises: In the case that the first target does not meet the first preset condition, determining a first ordinate of the first target in each frame of point cloud data based on the N frames of point cloud data, and acquiring a first vehicle speed corresponding to each frame of point cloud data of the first target; In the case where the first target satisfies the third preset condition, based on the N frames of point cloud data, determining the difference between the abscissa of the first target in the nth frame of point cloud data and the abscissa of the first target in the n+1th frame of point cloud data, to obtain N-1 lateral displacement differences, n=1, 2…N, the third preset condition being that each of the first vehicle speeds is less than a third threshold, each of the first ordinates is less than a fourth threshold, and N is greater than a fifth threshold; When the first target satisfies a fourth preset condition, determining that the first target is a pedestrian, wherein the fourth preset condition is that each of the lateral displacement differences is greater than a sixth threshold; When the first target does not satisfy the fourth preset condition, it is determined that the first target is not a pedestrian.

4. The method for determining a target category according to claim 2 or 3, characterized in that: The method further comprises: When the first target does not meet the first preset condition, obtaining a first intensity and a reflection cross-sectional area RCS of the radio wave reflected by the first target to the vehicle-mounted radar; When the first target satisfies the fifth preset condition, determining that the category of the first target is the second category, the second category indicates that the height of the first target is lower than or equal to the target height and affects the normal driving of the vehicle, the fifth preset condition is that the first intensity is greater than the seventh threshold, or the fifth preset condition is that the RCS is greater than the eighth threshold and the mean is greater than the ninth threshold; When the first target does not satisfy the fifth preset condition, it is determined that the category of the first target is not the second category.

5. The method for determining a target category according to claim 4, characterized in that: The method further comprises: When the first target is not a pedestrian and the category of the first target is not the second category, the category of the first target is determined to be a third category, and the third category indicates that the height of the first target is lower than or equal to the target height and does not affect normal driving of the vehicle.

6. The method for determining a target category according to claim 1, characterized in that: The step of acquiring N frames of point cloud data including the first target includes: Acquire the first point cloud data; Determine a second horizontal coordinate, a second vertical coordinate, and a second height of each object in the first point cloud data; Removing a target satisfying a sixth preset condition from the first point cloud data to obtain second point cloud data, the sixth preset condition being that a second horizontal coordinate of the target is greater than a tenth threshold, a second vertical coordinate is greater than an eleventh threshold, and the target satisfies a stable tracking condition, the stable tracking condition being that the target has M position changes satisfying a seventh preset condition, M being a positive integer, the seventh preset condition being that a change in the horizontal coordinate of the target in two adjacent frames of point cloud data is less than a twelfth threshold, a change in the vertical coordinate is less than a thirteenth threshold, and a change in the height is less than a fourteenth threshold, and the second point cloud data includes the first target; Taking the second point cloud data as the first frame of point cloud data, N frames of point cloud data including the first target are acquired.

7. The method for determining a target category according to claim 1, characterized in that: Determining a first height of the first target in each frame of point cloud data includes: The following steps are performed for each frame of point cloud data to obtain a first height of the first target in each frame of point cloud data: Determine a first distance between the first target and the origin of the point cloud coordinate system and a pitch angle of the first target based on the point cloud data, wherein the pitch angle is an angle between the first projection and the y-axis in the point cloud coordinate system, and the first projection is a projection of a line connecting the first target and the origin on a yz plane in the point cloud coordinate system; A first height of the first target in the point cloud data is determined based on the first distance and the pitch angle.

8. The method for determining a target category according to claim 2 or 3, characterized in that: The determining of the first vertical coordinate of the first target in each frame of point cloud data includes: The following steps are performed for each frame of point cloud data to obtain the first vertical coordinate of the first target in each frame of point cloud data: Determine a second distance between the first target and the origin of the point cloud coordinate system and a horizontal angle of the first target based on the point cloud data, wherein the horizontal angle is an angle between the second projection and the y-axis in the point cloud coordinate system, and the second projection is a projection of a line connecting the first target and the origin on an xy plane in the point cloud coordinate system; A first vertical coordinate of the first target in the point cloud data is determined based on the second distance and the horizontal angle.

9. The method for determining a target category according to claim 2 or 3, characterized in that: Determining the horizontal coordinate of the first target in the nth frame of point cloud data includes: Determine a third distance between the first target and the origin of the point cloud coordinate system and a horizontal angle of the first target based on the nth frame of point cloud data, wherein the horizontal angle is an angle between the second projection and the y-axis in the point cloud coordinate system, and the second projection is a projection of a line connecting the first target and the origin on an xy plane in the point cloud coordinate system; The horizontal coordinate of the first target in the nth frame of point cloud data is determined based on the third distance and the horizontal angle.

10. A device for determining a target category, characterized in that: The device comprises: A first acquisition module, used to acquire N frames of point cloud data containing a first target, where N is an integer greater than 1, and the N frames of point cloud data are collected by a vehicle-mounted radar; A first determination module is used to determine, based on the N frames of point cloud data, a first height of the first target in each frame of point cloud data to obtain N first heights; A second determination module, used to determine the mean and variance of the N first heights; The third determination module is used to determine that the category of the first target is a first category when the first target meets a first preset condition, the first category characterizing that the height of the first target is higher than or equal to the target height and does not affect the normal driving of the vehicle, the vehicle is the vehicle where the vehicle-mounted radar is located, and the first preset condition is that the difference between each of the first heights and the variance is greater than a first threshold and the variance is greater than a second threshold.

Images