A radar data processing method, apparatus, electronic device, and storage medium

Abstract

This invention provides a radar data processing method, apparatus, electronic device, and storage medium. The method includes: acquiring radar data of a current frame and radar data of a reference frame; acquiring a first object that exists in the reference frame but not in the current frame, and a second object that does not exist in the reference frame but exists in the current frame; acquiring a following object of the i-th first object in the reference frame; when a following object is acquired, acquiring a stitched object of the i-th first object in the current frame from the second object based on the radar data of the following object in the reference frame and the radar data of the i-th first object in the reference frame; when a stitched object is acquired, determining the radar data of the stitched object in the current frame as the radar data of the i-th first object in the current frame. Therefore, this invention solves the problem of missing detection objects within the detection range due to missing radar data.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a radar data processing method, apparatus, electronic device, and storage medium. Background Technology

[0002] With the explosive growth of the intelligent transportation industry, the application of intersection radar equipment has attracted increasing attention. Radar equipment senses the speed and size of dynamic objects on the road within a certain range based on the reflection characteristics of waves, thereby reconstructing the continuous traffic flow status of the sensing area.

[0003] Currently, due to the large number of radar devices installed at intersections, cost constraints often prevent the large-scale installation of high-performance, high-precision radars at intersections. Moreover, the possibility of large-scale replacement with high-precision radars in a short period of time is not high. Therefore, most radar devices at intersections are medium- or low-precision radar devices.

[0004] Among them, the low frequency of signals emitted by medium and low precision radar equipment results in a similarly low resolution of the returned images, making it easy to lose the detection of tracked objects. At the same time, as the distance of the object to be detected by the radar equipment increases, the transmission and return of radar signals are more susceptible to interference, and the corresponding detection accuracy also decreases. Moreover, the detection range of the radar is affected by a variety of factors such as the installation angle and height of the equipment, weather, obstruction by obstacles, and overlapping of the collected targets.

[0005] Therefore, in the existing technology, when using low-to-medium precision radar equipment to detect objects at intersections, the returned detection data may be incomplete, resulting in the disappearance of the detected objects within the detection range. Summary of the Invention

[0006] This invention provides a radar data processing method, apparatus, electronic device, and storage medium to solve the problem of missing radar data causing the detection object to disappear from the detection range.

[0007] In a first aspect of the present invention, a radar data processing method is provided, the method comprising:

[0008] Acquire radar data of the current frame and radar data of a reference frame, wherein the reference frame is the frame preceding the current frame, and the radar data includes radar data of a radar detecting at least one object.

[0009] Obtain a first object that exists in the reference frame but does not exist in the current frame;

[0010] Obtain the i-th following object of the first object in the reference frame, where i is an integer from 1 to N, and N is the total number of the first objects;

[0011] If the following object is obtained, the stitching object of the i-th first object in the current frame is obtained based on the radar data of the following object in the reference frame and the radar data of the i-th first object in the reference frame.

[0012] When the stitching object is obtained, the radar data of the stitching object in the current frame is determined as the radar data of the i-th first object in the current frame.

[0013] In a second aspect of the present invention, a radar data processing apparatus is also provided, the apparatus comprising:

[0014] A radar data acquisition module is used to acquire radar data of the current frame and radar data of a reference frame, wherein the reference frame is the frame preceding the current frame, and the radar data includes radar data of a radar detecting at least one object.

[0015] The first object acquisition module is used to acquire a first object that exists in the reference frame but does not exist in the current frame;

[0016] The car-following object acquisition module is used to acquire the i-th car-following object of the first object in the reference frame, where i is an integer from 1 to N, and N is the total number of the first objects;

[0017] The splicing object acquisition module is used to, when the following object is acquired, acquire the splicing object of the i-th first object in the current frame based on the radar data of the following object in the reference frame and the radar data of the i-th first object in the reference frame;

[0018] The determination module is used to determine, when the stitching object is obtained, the radar data of the stitching object in the current frame as the radar data of the i-th first object in the current frame.

[0019] In a third aspect of the present invention, an electronic device is also provided, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus.

[0020] Memory, used to store computer programs;

[0021] The processor, when executing the program stored in the memory, implements the radar data processing method described above.

[0022] In a fourth aspect of the present invention, a computer-readable storage medium is also provided, wherein instructions are stored therein, which, when executed on a computer, cause the computer to perform any of the radar data processing methods described above.

[0023] In a fifth aspect of the present invention, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to perform any of the radar data processing methods described above.

[0024] The embodiments of the present invention include at least the following technical effects:

[0025] The radar data processing method provided in this embodiment of the invention can acquire radar data of the current frame and radar data of a reference frame, wherein the reference frame is the frame preceding the current frame, and the radar data includes radar data of a radar detecting at least one object; then, it acquires a first object that exists in the reference frame but does not exist in the current frame, and a second object that does not exist in the reference frame but exists in the current frame; furthermore, it acquires the i-th following object of the first object in the reference frame, where i is an integer from 1 to N, and N is the total number of the first objects; thus, when the following object is acquired, based on the radar data of the following object in the reference frame and the radar data of the i-th first object in the reference frame, it acquires the i-th stitched object of the first object in the current frame from the second object; furthermore, when the stitched object is acquired, it determines the radar data of the stitched object in the current frame as the radar data of the i-th first object in the current frame.

[0026] Therefore, in this embodiment of the invention, when the radar data of the current frame is acquired, radar data prediction and object disappearance trajectory splicing can be performed on objects with missing radar data in the current frame based on the radar data of the current frame and the radar data of the previous frame. That is, the splicing object is determined based on the following object, realizing real-time trajectory splicing of objects with missing radar data. In this way, the processed current frame data is used as the reference frame for the next frame, so that the acquired radar data can be continuously repaired in real time, which solves the problem of the disappearance of the detection object in the detection range due to the lack of radar data to a certain extent. Attached Figure Description

[0027] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.

[0028] Figure 1 This is a schematic diagram of a radar data processing method provided in an embodiment of the present invention;

[0029] Figure 2 This is a schematic diagram illustrating the factors affecting vehicle prediction at intersections provided in an embodiment of the present invention.

[0030] Figure 3This is a schematic diagram of the semi-circular region for determining the object to which interference is applied, provided in an embodiment of the present invention.

[0031] Figure 4 This is a flowchart illustrating a specific implementation of the radar data processing method provided in this embodiment of the invention.

[0032] Figure 5 This is a schematic diagram of the radar trajectory of the vehicle before it disappears, provided in an embodiment of the present invention.

[0033] Figure 6 This is a schematic diagram of radar trajectory prediction for a missing vehicle provided in an embodiment of the present invention;

[0034] Figure 7 This is a schematic diagram of radar trajectory stitching for a vanished vehicle provided in an embodiment of the present invention;

[0035] Figure 8 This is a schematic diagram of the structure of a radar data processing device provided in an embodiment of the present invention;

[0036] Figure 9 This is a block diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0037] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0038] It should be understood that the phrase "one embodiment" or "an embodiment" throughout the specification means that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of the invention. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification do not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments.

[0039] In various embodiments of the present invention, it should be understood that the sequence number of each process described below does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0040] The technical solutions of the present invention will now be described with reference to the accompanying drawings in the embodiments of the present invention.

[0041] like Figure 1The diagram shown is a flowchart of a radar data processing method according to an embodiment of the present invention. The method may include the following steps:

[0042] Step 101: Obtain radar data for the current frame and radar data for the reference frame.

[0043] The reference frame is the frame preceding the current frame, and the radar data includes radar data from a radar detecting at least one object.

[0044] For example, if the current frame is the 10th frame within a certain time period, then the reference frame is the 9th frame; 1 frame is 0.1 seconds.

[0045] In addition, the radar data mentioned above may include the type of object, its location (e.g., latitude and longitude information), speed, and direction of movement (e.g., azimuth information).

[0046] Step 102: Obtain a first object that exists in the reference frame but does not exist in the current frame, and a second object that does not exist in the reference frame but exists in the current frame.

[0047] The presence of an object in a reference frame indicates that radar data exists in that frame; the absence of an object in the current frame indicates that radar data does not exist in the reference frame.

[0048] For example, if a radar device detects two objects, A and B, where A has no radar data recorded in the reference frame but has radar data recorded in the current frame, and B has radar data recorded in the reference frame but no radar data recorded in the current frame, then A is determined to be the second object, and B is determined to be the first object.

[0049] Step 103: Obtain the i-th following object of the first object in the reference frame, where i is an integer from 1 to N, and N is the total number of the first objects.

[0050] In this context, the following object refers to the object that is positioned in front of the first object in the reference frame.

[0051] For example, if the first object is vehicle A, then the following object is any vehicle traveling in a straight line within a certain length in front of vehicle A. Based on the driving angle and position of vehicle A, vehicle B, which is closest to vehicle A among any vehicles, can be identified as the following object.

[0052] Additionally, when N is an integer greater than 1 (i.e., when there are multiple first objects), it is necessary to obtain the trailing object of each first object.

[0053] Step 104: When the following object is obtained, based on the radar data of the following object in the reference frame and the radar data of the i-th first object in the reference frame, obtain the splicing object of the i-th first object in the current frame from the second object.

[0054] Wherein, if the first object exists in the reference frame but not in the current frame, then the first object can be called a "disappeared object"; and if the second object does not exist in the reference frame but exists in the current frame, then the second object can be called a "object to be stitched". In the embodiments of the present invention, the following object of the first disappearing object can be determined first, and then the stitching object of the disappearing object can be selected from the "objects to be stitched" according to the following object.

[0055] Step 105: When the stitching object is obtained, the radar data of the stitching object in the current frame is determined as the radar data of the i-th first object in the current frame.

[0056] After obtaining the stitching object, the radar data of the stitching object in the current frame is determined as the radar data of the i-th first object in the current frame. This means that the identity information (IDentity, ID) included in the radar data of the stitching object is modified to the ID of the i-th first object. For example, if the ID of vehicle A is "1234" and the ID of vehicle B is "1212", then if vehicle A is a stitching object of vehicle B, the ID of vehicle A is modified to "1212".

[0057] In addition, after completing step 105, the current frame is used as the reference frame for the next frame, and steps 101 to 105 are repeated to continuously repair the radar data obtained in real time.

[0058] In addition, radar data can be repaired once at a preset time interval, or when a preset operation is received (such as a click operation on a preset control on the radar data display interface).

[0059] It should be noted that when there are multiple radars, the operations described in steps 101 to 105 above can be performed individually for the radar data of each radar.

[0060] As can be seen from steps 101 to 105 above, the radar data processing method provided in this embodiment of the invention can acquire radar data of the current frame and radar data of a reference frame, wherein the reference frame is the frame preceding the current frame, and the radar data includes radar data of a radar detecting at least one object; then, a first object that exists in the reference frame but does not exist in the current frame, and a second object that does not exist in the reference frame but exists in the current frame are acquired; furthermore, the i-th following object of the first object in the reference frame is acquired, where i is an integer from 1 to N, and N is the total number of the first objects; thus, when the following object is acquired, based on the radar data of the following object in the reference frame and the radar data of the i-th first object in the reference frame, the i-th stitched object of the first object in the current frame is acquired from the second object; furthermore, when the stitched object is acquired, the radar data of the stitched object in the current frame is determined as the radar data of the i-th first object in the current frame.

[0061] Therefore, in this embodiment of the invention, when the radar data of the current frame is acquired, radar data prediction and object disappearance trajectory splicing can be performed on objects with missing radar data in the current frame based on the radar data of the current frame and the radar data of the previous frame. That is, the splicing object is determined based on the following object, realizing real-time trajectory splicing of objects with missing radar data. In this way, the processed current frame data is used as the reference frame for the next frame, so that the acquired radar data can be continuously repaired in real time, which solves the problem of the disappearance of the detection object in the detection range due to the lack of radar data to a certain extent.

[0062] Optionally, obtaining the i-th following object of the first object in the reference frame includes:

[0063] Based on the radar data of the i-th first object in the reference frame, the reference area where the i-th first object is located is determined;

[0064] Obtain a third object that is located within the reference region in the reference frame;

[0065] The object that is closest to the i-th first object and is located in front of the i-th first object among the third objects is determined as the following object of the i-th first object in the reference frame.

[0066] The reference area can be the rectangular area in front of the first object in the reference frame, and the third object refers to the object located in front of the first object in the reference frame.

[0067] In addition, the term "front" as mentioned above refers to the direction in front of the i-th first object along its movement direction.

[0068] As can be seen from the above, in the embodiments of the present invention, a reference region can be defined based on the position of the i-th first object in the reference frame, and the object in the reference region that is closest to the i-th first object and is located in front of the i-th first object can be used as the following object of the i-th first object.

[0069] Optionally, radar data may include the object's position and direction of motion;

[0070] Determining the reference area where the i-th first object is located based on the radar data of the i-th first object in the reference frame includes:

[0071] Determine a target straight line, wherein the target straight line passes through the position of the i-th first object in the reference frame and is parallel to the direction of movement of the i-th first object in the reference frame;

[0072] Based on the target straight line, a first straight line, a second straight line, a third straight line, and a fourth straight line are obtained, and the rectangular area formed by the intersection of the first straight line, the second straight line, the third straight line, and the fourth straight line is determined as the reference area;

[0073] Wherein, the first straight line and the second straight line are located on both sides of the target straight line, and are parallel to the first straight line and the second straight line, and the distance between the first straight line and the target straight line is a first preset value, and the distance between the second straight line and the target straight line is the first preset value;

[0074] The third straight line passes through the position of the i-th first object in the reference frame and is perpendicular to the target straight line. The fourth straight line passes through the first target reference point and is perpendicular to the target straight line. The first target reference point is a point on the target straight line that moves forward along the direction of movement of the i-th first object in the reference frame. The distance between the first target reference point and the third straight line is the second preset value.

[0075] For example, the first object is vehicle A, and the radar data of vehicle A includes the position and direction of movement of vehicle A; taking the direction of movement of vehicle A as the straight line direction, draw a straight line through the position of vehicle A, and this straight line is called the target straight line; along the direction perpendicular to the target straight line, draw two straight lines parallel to the target straight line and located on both sides of the target straight line, and call them the first straight line and the second straight line respectively, and the distance between the first straight line, the second straight line and the target straight line can be a first preset value (such as 1.5 meters); draw a straight line perpendicular to the target straight line and passing through the position of vehicle A, and call it the third straight line; along the direction of movement of vehicle A, draw a fourth straight line parallel to the third straight line, and the distance between the fourth straight line and the third straight line can be a second preset value (such as 100 meters); then the rectangular area surrounded by the first straight line, the second straight line, the third straight line and the fourth straight line is the reference area where vehicle A is located.

[0076] It should be noted that the first and second preset values ​​can be set according to the motion scenario of the first object. For example, in a driving lane, the first and second preset values ​​are determined according to the lane level.

[0077] Optionally, the step of obtaining the stitched object of the i-th first object in the current frame from the second object based on the radar data of the following object in the reference frame and the radar data of the i-th first object in the reference frame includes:

[0078] If the following object is obtained, the target position of the i-th first object in the current frame is predicted based on the radar data of the following object in the reference frame and the radar data of the i-th first object in the reference frame.

[0079] Based on the target position, obtain the i-th splicing object of the first object in the current frame from the second object.

[0080] As can be seen from the above, in the embodiments of the present invention, when a following object is obtained, the target position of the i-th first object in the current frame can be predicted based on the radar data of the following object and the i-th first object in the reference frame, so that the splicing object of the i-th first object can be selected from the above-mentioned second objects based on the target position.

[0081] Optionally, radar data may include the object's position and velocity;

[0082] The step of predicting the target position of the i-th first object in the current frame based on the radar data of the following object in the reference frame and the radar data of the i-th first object in the reference frame when the following object is obtained includes:

[0083] When the car-following object is obtained, the first acceleration of the i-th first object in the reference frame is calculated based on the speed and position of the car-following object in the reference frame, the speed and position of the i-th first object in the reference frame, and the predetermined car-following model.

[0084] Calculate the second acceleration of the i-th first object in the reference frame, wherein the second acceleration is the acceleration generated by the resultant force of the boundary force and the repulsive force acting on the first object in the reference frame;

[0085] Based on the first acceleration and the second acceleration, determine the combined acceleration of the i-th first object in the reference frame;

[0086] Based on the combined acceleration, predict the target position of the i-th first object in the current frame.

[0087] Among them, the car-following model describes the interaction between two adjacent vehicles in a convoy traveling on a single lane where overtaking is restricted; at the same time, the car-following model is applicable to vehicles traveling in convoy on a single lane where overtaking is not allowed, and is a mathematical model for studying the driving state of a following vehicle following the preceding vehicle when vehicles are queuing in a single lane.

[0088] In addition, the velocity and position (i.e., latitude and longitude coordinates) of the following object in the reference frame, and the velocity and position (i.e., latitude and longitude coordinates) of the i-th first object in the reference frame, are substituted into the following car-following model formula to calculate the first acceleration:

[0089]

[0090]

[0091] Where a represents the first acceleration, a' represents the expected acceleration of the current road segment, and v a v represents the velocity of the first object. o The following parameters represent the expected speed of the first object in the current motion scenario, v represents the speed of the following object, Δv represents the speed difference between the following object and the first object, T represents the safe following distance (the ratio of the longitudinal distance between two vehicles to the speed of the following vehicle, generally reflecting the safety of the automatic cruise control function; it can be seen as the time required for the following vehicle to collide with the preceding vehicle if the preceding vehicle brakes to a stop and the following vehicle does not slow down, for example, 1.5 seconds), b represents the customizable expected deceleration (the ratio of the change in speed after braking to the time taken for this change to occur; referring to acceleration, used in car braking, it is usually called braking deceleration), and s. α This represents the distance between the first object and the following object, s0 represents the parking distance of 2 meters, α is a constant with a default value of 4, and β is a constant with a default value of 2.

[0092] It should be noted that the motion state of the first object varies considerably depending on the scenario, such as... Figure 2 As shown, for example, in complex intersections, if a vehicle does not have a clear direction and speed, its direction of travel will be affected by surrounding vehicles, vehicles turning left, and surrounding non-motorized vehicles, thus increasing the uncertainty in predicting the target location. Furthermore, in real-time trajectory prediction, many other complex environmental factors must be considered. Therefore, in complex intersections, since car-following models cannot accurately represent more detailed vehicle behaviors such as turning, social force models are needed to further predict disappearing vehicles.

[0093] The social force model is a traffic flow model that considers the direction of travel, the destination direction, and interference from surrounding motor vehicles, pedestrians, and non-motorized vehicles. It includes the resultant force acting on vehicle i traveling within a complex intersection. Mainly includes driving force Boundary forces and repulsive force

[0094] In addition, boundary forces refer to the forces that prevent drivers from crossing lane lines and other boundary lines, while repulsive forces refer to the forces that prevent drivers from colliding with vehicles in their field of vision.

[0095] It should be noted that the driving force represented by the acceleration in the car-following model can be regarded as the first acceleration. The first acceleration and the second acceleration can be added together as the comprehensive acceleration. Based on the comprehensive acceleration, the target position of the first object in the current frame can be predicted.

[0096] In this embodiment of the invention, the comprehensive acceleration and radar data of the first object in the reference frame are substituted into the target position prediction formula to obtain the latitude and longitude coordinates of the first object in the current frame:

[0097]

[0098]

[0099]

[0100] in, This indicates the displacement of the first object from its position in the reference frame to its position in the current frame. Indicates the velocity of the first object in the reference frame; The vector represents the combined acceleration; x1 represents the predicted x-coordinate of the target position; y1 represents the predicted y-coordinate of the target position; θ1 represents... The direction angle; x0 represents the x-coordinate of the first object in the reference frame; y0 represents the y-coordinate of the first object in the reference frame.

[0101] Optionally, the method further includes:

[0102] If the following object is not obtained, the first acceleration of the i-th first object is set to the target preset value;

[0103] Calculate the second acceleration of the i-th first object in the reference frame, wherein the second acceleration is the acceleration generated by the resultant force of the boundary force and the repulsive force acting on the first object in the reference frame;

[0104] Based on the first acceleration and the second acceleration, determine the combined acceleration of the i-th first object in the reference frame;

[0105] Based on the combined acceleration, predict the target position of the i-th first object in the current frame.

[0106] If no following object is found, it means that the first object is the first object in the current scene. For example, if the first object is vehicle A and the current scene is a queuing area, the movement state of vehicle A is determined by the movement state of vehicles in the adjacent lanes. If the movement direction of vehicle A is red light prohibition, then the speed and acceleration of vehicle A remain at 0. If the movement direction of vehicle A is green light passage, then the acceleration of vehicle A is the maximum acceleration of the current road segment.

[0107] Additionally, if the first object is vehicle A and the current scene is a non-queuing area, then vehicle A's acceleration is 0, and its speed and direction of motion are the same as in the previous frame.

[0108] As can be seen from the above, if no following object is obtained, the first acceleration of the i-th first object is set to the target preset value, and then the first acceleration is combined with the acceleration generated by the resultant force of the boundary force and repulsion force on the first object in the reference frame (i.e., the second acceleration). Then, the target position of the i-th first object in the current frame can be predicted based on the combined acceleration.

[0109] Optionally, calculating the second acceleration of the i-th first object in the reference frame includes:

[0110] Determine the boundary force exerted on the i-th first object in the reference frame;

[0111] Obtain the i-th interference-applying object of the first object in the reference frame;

[0112] Obtain the repulsive force exerted by each of the disturbances on the i-th first object;

[0113] Calculate the resultant force of the boundary forces and each of the repulsive forces;

[0114] Calculate the second acceleration based on the resultant force.

[0115] The force can be obtained by calculating the product of the object's mass and acceleration. In this embodiment of the invention, the mass of the first object can be set to a default value of 1. The resultant force obtained by adding the boundary force and the repulsive force can be regarded as the second acceleration.

[0116] Furthermore, the boundary forces can be obtained from the following boundary force formula.

[0117]

[0118] Among them, A B B B The constant represents the gain coefficient of the boundary force. This indicates the distance between the first object and the boundary line, with the direction pointing from the first object to the boundary line; It is the normal vector perpendicular to the boundary line.

[0119] Furthermore, the repulsive force can be obtained from the following formula.

[0120]

[0121] Where i represents the i-th first object, and j represents the j-th object to which the interference is applied; A V B V is a constant, representing the gain coefficient of the force between the first object and the object applying the disturbance; ω is a constant; d ij Let n be a vector, with its direction pointing from the i-th first object to the j-th object applying the interference, and its size being the distance between the two objects; ij It is a unit vector pointing from the j-th interference object to the i-th first object.

[0122] Optionally, obtaining the interference-applying object of the i-th first object in the reference frame includes:

[0123] Using the second target reference point as the center and a preset length as the radius, a semicircle is obtained in which the boundary line passes through the position of the i-th first object in the reference frame, wherein the second target reference point is a point along the movement direction of the i-th first object;

[0124] The object located within the semi-circular area in the reference frame is identified as the object to which the interference is applied.

[0125] That is, Figure 3As shown, along the movement direction of the i-th first object in the reference frame, a position with a preset distance from the i-th first object is determined. This position is the circular position. Then, with this position as the center and the preset distance as the radius, a semicircle is drawn between the center position and the position of the i-th first object. The objects within this semicircular area are the interference objects applied to the i-th first object.

[0126] Optionally, radar data may include object type, velocity, position, and direction of motion;

[0127] Radar data includes object type, velocity, position, and direction of motion;

[0128] The step of obtaining the i-th splicing object of the first object in the current frame according to the target position includes:

[0129] Obtain objects from the second object that meet preset conditions as candidate splicing objects. The preset conditions include that the object type is the same as the object type of the i-th first object, the deviation angle between the movement direction and the movement direction of the i-th first object is less than a third preset value, and the deviation between the speed and the speed of the i-th first object is less than a fourth preset value.

[0130] Among the candidate splicing objects, the object that is closest to the i-th first object is obtained, and is used as the splicing object of the i-th first object in the current frame.

[0131] When the object is a vehicle, the fourth preset value can take different values ​​depending on whether the second object is a motor vehicle or a non-motor vehicle.

[0132] For example, consider vehicles A, B, C, and D. Vehicle A has type X, direction of motion angle a1, and speed a2; vehicle B has type Y, direction of motion angle b1, and speed b2; vehicle C has type X, direction of motion angle c1, and speed c2; vehicle D has type X, direction of motion angle d1, and speed d2. Vehicles A, C, and D have the same type. If the absolute value of the difference between a1 and c1 is less than a third preset value, and the absolute value of the difference between a2 and c2 is less than a fourth preset value, vehicle C is determined as a candidate splicing object. If the absolute value of the difference between a1 and d1 is less than a third preset value, and the absolute value of the difference between a2 and d2 is less than a fourth preset value, vehicle D is determined as a candidate splicing object. Then, among vehicles C and D, the vehicle closest to vehicle A is selected as the splicing object.

[0133] Optionally, the method further includes:

[0134] When the stitching object is not obtained, the target position is determined as the position of the i-th first object in the radar data of the current frame, and the i-th first object is recorded as having failed to stitch once.

[0135] When the number of times the first object i fails to be successfully spliced ​​reaches a preset number, stop predicting the position of the first object i.

[0136] Specifically, after obtaining the predicted position (i.e., target position) of the first object in the current frame, if no stitching object is obtained, the target position is used as the position of the first object in the current frame; thus, when the current frame is used as the reference frame for the next frame, radar data can be continuously repaired. This process continues until the number of unsuccessful stitching results for the same first object reaches a preset number, at which point the prediction of the first object stops.

[0137] Furthermore, if the i-th first object fails to be stitched successfully X times consecutively, it can be assumed that the predicted duration of the i-th first object lasts for X frames. Therefore, by determining whether the number of consecutive failures of the i-th first object to be stitched successfully exceeds the preset number, it can be determined whether the predicted duration of the i-th first object exceeds the preset duration.

[0138] Therefore, to avoid situations where a missing object in the radar trajectory never reappears, causing a certain object to remain in the prediction indefinitely, which would lead to a discrepancy between the number of objects at any given time and the actual number of objects, resulting in more predicted objects than actually there, an upper limit is set on the number of predictions or the duration. For example, if the predicted trajectory exceeds 5 seconds after an object disappears and no matching object is found, the prediction can be terminated.

[0139] In summary, the specific implementation method of the radar data processing method of this invention can be described as follows:

[0140] like Figure 4 As shown, step H1: acquire radar data collected by a radar device on at least one detection object in the reference frame and the current frame;

[0141] Step H2: Determine the first object from the radar data of the reference frame and the current frame, wherein the first object is an object that exists in the reference frame but does not exist in the current frame;

[0142] Step H3: Determine whether the first object has a following object; if the first object has a following object, calculate the first acceleration of the first object; if the first object does not have a following object, the first acceleration of the first object is a preset value.

[0143] Step H4: Obtain the boundary forces of the first object;

[0144] Step H5: Obtain the repulsive force of the first object;

[0145] Step H6: Based on the first acceleration and the second acceleration generated by the resultant force of the boundary force and repulsive force of the first object, predict the position of the first object in the current frame;

[0146] Step H7: Based on the predicted position of the first object in the current frame, determine whether a stitching object exists; if a stitching object exists, modify the ID of the stitching object to the ID of the first object, thus completing the trajectory prediction of the first object; if no stitching object exists, determine whether the prediction duration for the first object has reached the maximum prediction duration; if the maximum prediction duration has been reached, stop the prediction of the first object; if the maximum prediction duration has not been reached, record the radar data predicted for the first object as the radar data of the first object in the current frame (so that when the current frame is used as the reference frame for the next frame, it is assumed that the first object exists in the reference frame of the next frame), and proceed to the next time frame;

[0147] Step H8: Determine whether the current frame is the last frame. If the current frame is the last frame, stop prediction; if the current frame is not the last frame, execute step H2.

[0148] Specifically, for example, the reference frame data set is named M, and the current frame data set is named N;

[0149] The set of IDs that exist in M ​​but not in N is named D, referring to the missing vehicles;

[0150] The set of IDs that exist in N but not in M ​​is named A, referring to the vehicles that appear again and need to be spliced.

[0151] Scenario 1:

[0152] like Figure 5 As shown, vehicle ①, which exists in M, is not detected in N, meaning vehicle ① disappears in N. Therefore, the trajectory of vehicle ① is predicted in N. At this point, vehicle ① is located inside the intersection, and its direction angle is to the left before disappearing. Since no valid following object is identified for vehicle ①, it continues to travel at its original angle and speed; that is, the following acceleration of vehicle ① is 0 at this moment. Because vehicle ① is a left-turning vehicle, it is constrained by the left-turning vehicle's travel boundary. That is, inside the intersection, the left-turning vehicle will be subjected to a leftward acceleration a2 perpendicular to its current travel direction throughout the entire process to ensure the continuous change of its travel angle and complete the entire turning process. Under the action of a2, the vehicle acquires a vertical velocity v2.

[0153] For vehicle ①, there is vehicle ② within its visible range. Vehicle ② is traveling towards vehicle ①, and its speed is greater than a threshold value. Therefore, vehicle ② exerts a repulsive force on the predicted vehicle ①. This repulsive force is perpendicular to the traveling direction of vehicle ① and towards vehicle ②. Combining the original speed v1, the speed v2 caused by the boundary force, and the speed v3 caused by the repulsive force acceleration, the speed of vehicle ① is v0. Figure 6 As shown, ③ is the prediction result of continuing to travel at the original speed, ⑤ is the prediction result considering the driving force and boundary force, and ④ is the result obtained after considering the repulsive force of vehicle ② on vehicle ① based on ⑤, which is also the result that best matches the actual situation.

[0154] Scenario 2:

[0155] In the current frame, vehicle ⑥ exists in M ​​but not in N, meaning it disappears in N; therefore, trajectory prediction is performed for vehicle ⑥. In the reference frame, vehicle ⑥ has not detected the vehicle in front and is subject to the driving boundary force of the right-turn lane; the predicted trajectory of the vehicle is a right-turning posture, and trajectory prediction is performed. Trajectories for the target vehicle are predicted in the following frames.

[0156] like Figure 7 As shown, when vehicle ⑥ is predicted to be vehicle ⑦ in the current frame, and vehicle ⑧ appears in the next frame, since the distance between the position of vehicle ⑧ and the predicted position of vehicle ⑦ in the next frame is less than 5 meters, and the vehicle types are the same, the difference in driving direction is less than 30 degrees, and the speed difference is less than 5 m / s, the preset conditions for determining the stitching object are met. Therefore, vehicle ⑧ is considered to be the vehicle that reappeared after vehicle ⑥ disappeared. In the next frame, the prediction of vehicle ⑥ is canceled, the trajectories of vehicle ⑥ and vehicle ⑧ are stitched together, and the IDs of all vehicles ⑧ in subsequent frames are renamed to the IDs of vehicle ⑥.

[0157] As described above, the radar data processing method of this invention, based on real-time transmitted radar detection data and a microscopic driving behavior model, predicts the movement trajectories of multiple targets in real time. It identifies the predicted vehicle based on real-time data from two consecutive frames, uses a car-following model to predict the acceleration of the vanished vehicle in the straight line in real time, and obtains the acceleration of the vehicle in other directions based on a social force model, thereby realizing the instantaneous prediction of the trajectory of the vanished vehicle until a vehicle with a similar model, shape, and position to the predicted vehicle is found and reappears. At this point, the prediction stops, the IDs of the two vehicles are unified, and the missing trajectory is completed.

[0158] It should be noted that, for the sake of simplicity, the method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments of the present invention are not limited to the described order of actions, because according to the embodiments of the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions involved are not necessarily essential to the embodiments of the present invention.

[0159] Reference Figure 8 The diagram illustrates a structural block diagram of a radar data processing device according to an embodiment of the present invention. The radar data processing device 800 may include the following modules:

[0160] The radar data acquisition module 801 is used to acquire radar data of the current frame and radar data of a reference frame, wherein the reference frame is the previous frame of the current frame, and the radar data includes radar data of a radar detecting at least one object.

[0161] The first object acquisition module 802 is used to acquire a first object that exists in the reference frame but does not exist in the current frame, and a second object that does not exist in the reference frame but exists in the current frame;

[0162] The following object acquisition module 803 is used to acquire the i-th following object of the first object in the reference frame, where i is an integer from 1 to N, and N is the total number of the first objects;

[0163] The splicing object acquisition module 804 is used to, when the following object is acquired, obtain the splicing object of the i-th first object in the current frame from the second object based on the radar data of the following object in the reference frame and the radar data of the i-th first object in the reference frame;

[0164] The determining module 805 is used to determine the radar data of the splicing object in the current frame as the radar data of the i-th first object in the current frame when the splicing object is obtained.

[0165] Optionally, the carousel object acquisition module 803 includes:

[0166] The reference area determination submodule is used to determine the reference area where the i-th first object is located based on the radar data of the i-th first object in the reference frame.

[0167] The third object acquisition submodule is used to acquire a third object that is located within the reference region in the reference frame;

[0168] The following object determination submodule is used to determine the object among the third objects that is closest to the i-th first object and is located in front of the i-th first object as the following object of the i-th first object in the reference frame.

[0169] Optionally, radar data may include the object's position and direction of motion;

[0170] The reference region determination submodule includes:

[0171] The first determining unit is used to determine the target straight line, wherein the target straight line is the position of the i-th first object in the reference frame and is parallel to the movement direction of the i-th first object in the reference frame;

[0172] The second determining unit is used to obtain a first straight line, a second straight line, a third straight line, and a fourth straight line based on the target straight line, and to determine the rectangular area formed by the intersection of the first straight line, the second straight line, the third straight line, and the fourth straight line as the reference area;

[0173] Wherein, the first straight line and the second straight line are located on both sides of the target straight line, and are parallel to the first straight line and the second straight line, and the distance between the first straight line and the target straight line is a first preset value, and the distance between the second straight line and the target straight line is the first preset value;

[0174] The third straight line passes through the position of the i-th first object in the reference frame and is perpendicular to the target straight line. The fourth straight line passes through the first target reference point and is perpendicular to the target straight line. The first target reference point is a point on the target straight line that moves forward along the direction of movement of the i-th first object in the reference frame. The distance between the first target reference point and the third straight line is the second preset value.

[0175] Optionally, the splicing object acquisition module 804 includes:

[0176] The prediction submodule is used to predict the target position of the i-th first object in the current frame based on the radar data of the following object in the reference frame and the radar data of the i-th first object in the reference frame when the following object is acquired.

[0177] The splicing object determination submodule is used to obtain the i-th splicing object of the first object in the current frame from the second object according to the target position.

[0178] Optionally, radar data may include the object's position and velocity;

[0179] The prediction submodule includes:

[0180] The first acceleration calculation unit is used to calculate the first acceleration of the i-th first object in the reference frame based on the speed and position of the following object in the reference frame, the speed and position of the i-th first object in the reference frame, and a predetermined following model when the following object is obtained.

[0181] The second acceleration calculation unit is used to calculate the second acceleration of the i-th first object in the reference frame, wherein the second acceleration is the acceleration generated by the resultant force of the boundary force and the repulsive force acting on the first object in the reference frame;

[0182] A composite acceleration determination unit is configured to determine the composite acceleration of the i-th first object in the reference frame based on the first acceleration and the second acceleration.

[0183] A target position prediction unit is configured to predict the target position of the i-th first object in the current frame based on the combined acceleration.

[0184] Optionally, the radar data processing device 800 further includes:

[0185] The setting module is used to set the first acceleration of the i-th first object to a target preset value when the following object is not obtained;

[0186] The second acceleration calculation module is used to calculate the second acceleration of the i-th first object in the reference frame, wherein the second acceleration is the acceleration generated by the resultant force of the boundary force and the repulsive force acting on the first object in the reference frame;

[0187] A composite acceleration determination module is used to determine the composite acceleration of the i-th first object in the reference frame based on the first acceleration and the second acceleration;

[0188] The target position prediction module is used to predict the target position of the i-th first object in the current frame based on the comprehensive acceleration.

[0189] Optionally, the second acceleration calculation module includes:

[0190] A boundary force determination submodule is used to determine the boundary force experienced by the i-th first object in the reference frame;

[0191] An interference application object acquisition submodule is used to acquire the i-th interference application object of the first object in the reference frame;

[0192] The repulsion force acquisition submodule is used to acquire the repulsion force of each of the interference-applying objects on the i-th first object;

[0193] The resultant force calculation submodule is used to calculate the resultant force of the boundary forces and each of the repulsive forces;

[0194] The second acceleration calculation submodule is used to calculate the second acceleration based on the resultant force.

[0195] Optionally, the interference application object acquisition submodule includes:

[0196] The third determining unit is used to obtain a semicircle with the second target reference point as the center and a preset length as the radius, through which the boundary line passes the position of the i-th first object in the reference frame, wherein the second target reference point is a point along the movement direction of the i-th first object;

[0197] The fourth determining unit is used to determine the object located within the semi-circular area in the reference frame as the object to which the interference is applied.

[0198] Optionally, radar data may include object type, velocity, position, and direction of motion;

[0199] The splicing object determination submodule includes:

[0200] The first acquisition unit is used to acquire objects in the second object that meet preset conditions as candidate splicing objects. The preset conditions include that the object type is the same as the object type of the i-th first object, and the deviation angle between the movement direction and the movement direction of the i-th first object is less than a third preset value, and the deviation between the speed and the speed of the i-th first object is less than a fourth preset value.

[0201] The second acquisition unit is used to acquire the candidate splicing objects that are closest to the i-th first object, so as to use the i-th first object as the splicing object in the current frame.

[0202] Optionally, the radar data processing device 800 further includes:

[0203] The position determination module is used to determine the target position as the position of the i-th first object in the radar data of the current frame when the stitching object is not obtained, and to record that the i-th first object failed to be stitched once.

[0204] The stop prediction module is used to stop predicting the position of the i-th first object when the number of times the i-th first object fails to be successfully spliced ​​reaches a preset number.

[0205] Therefore, the radar data processing method provided in this embodiment of the invention can acquire radar data of the current frame and radar data of a reference frame, wherein the reference frame is the frame preceding the current frame, and the radar data includes radar data of a radar detecting at least one object; then, it acquires a first object that exists in the reference frame but does not exist in the current frame, and a second object that does not exist in the reference frame but exists in the current frame; furthermore, it acquires the i-th following object of the first object in the reference frame, where i is an integer from 1 to N, and N is the total number of the first objects; thus, when the following object is acquired, based on the radar data of the following object in the reference frame and the radar data of the i-th first object in the reference frame, it acquires the i-th stitched object of the first object in the current frame from the second object; furthermore, when the stitched object is acquired, it determines the radar data of the stitched object in the current frame as the radar data of the i-th first object in the current frame.

[0206] Therefore, in this embodiment of the invention, when the radar data of the current frame is acquired, radar data prediction and object disappearance trajectory splicing can be performed on objects with missing radar data in the current frame based on the radar data of the current frame and the radar data of the previous frame. That is, the splicing object is determined based on the following object, realizing real-time trajectory splicing of objects with missing radar data. In this way, the processed current frame data is used as the reference frame for the next frame, so that the acquired radar data can be continuously repaired in real time, which solves the problem of the disappearance of the detection object in the detection range due to the lack of radar data to a certain extent.

[0207] As the device embodiment is basically similar to the method embodiment, the description is relatively simple, and relevant parts can be found in the description of the method embodiment.

[0208] This invention also provides an electronic device, such as... Figure 9 As shown, it includes a processor 901, a communication interface 902, a memory 903, and a communication bus 904, wherein the processor 901, the communication interface 902, and the memory 903 communicate with each other through the communication bus 904.

[0209] Memory 903 is used to store computer programs;

[0210] The processor 901 is used to execute the program stored in the memory 903 to implement the radar data processing method described above.

[0211] The memory may include random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0212] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0213] In another embodiment of the present invention, a computer-readable storage medium is also provided, which stores instructions that, when executed on a computer, cause the computer to perform any of the radar data processing methods described in the above embodiments.

[0214] In another embodiment of the present invention, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to execute any of the radar data processing methods described in the above embodiments.

[0215] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state disk (SSD)).

[0216] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0217] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments 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.

[0218] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the scope of protection of the present invention.

Claims

1. A radar data processing method, characterized in that, The method includes: Acquire radar data of the current frame and radar data of a reference frame, wherein the reference frame is the frame preceding the current frame, and the radar data includes radar data of a radar detecting at least one object. Obtain a first object that exists in the reference frame but does not exist in the current frame, and a second object that does not exist in the reference frame but exists in the current frame; Obtain the i-th following object of the first object in the reference frame, where i is an integer from 1 to N, and N is the total number of the first objects; When the car-following object is obtained, based on the radar data of the car-following object in the reference frame and the radar data of the i-th first object in the reference frame, the stitched object of the i-th first object in the current frame is obtained from the second object. The stitched object is determined based on the target position of the i-th first object in the current frame. The target position is predicted based on the first acceleration of the i-th first object in the reference frame determined by the car-following model and the second acceleration of the i-th first object in the reference frame determined based on the social force model. The social force model is a traffic flow model that considers the driving direction, target direction, and interference from surrounding motor vehicles, pedestrians, and non-motor vehicles. When the stitching object is obtained, the radar data of the stitching object in the current frame is determined as the radar data of the i-th first object in the current frame.

2. The method according to claim 1, characterized in that, The step of obtaining the i-th carousel object of the first object in the reference frame includes: Based on the radar data of the i-th first object in the reference frame, the reference area where the i-th first object is located is determined; Obtain a third object located within the reference region in the reference frame; The object that is closest to the i-th first object and is located in front of the i-th first object among the third objects is determined as the following object of the i-th first object in the reference frame.

3. The method according to claim 2, characterized in that, Radar data includes the object's position and direction of movement; Determining the reference area where the i-th first object is located based on the radar data of the i-th first object in the reference frame includes: Determine a target straight line, wherein the target straight line passes through the position of the i-th first object in the reference frame and is parallel to the direction of movement of the i-th first object in the reference frame; Based on the target straight line, a first straight line, a second straight line, a third straight line, and a fourth straight line are obtained, and the rectangular area formed by the intersection of the first straight line, the second straight line, the third straight line, and the fourth straight line is determined as the reference area; Wherein, the first straight line and the second straight line are located on both sides of the target straight line, and are parallel to the first straight line and the second straight line, and the distance between the first straight line and the target straight line is a first preset value, and the distance between the second straight line and the target straight line is the first preset value; The third straight line passes through the position of the i-th first object in the reference frame and is perpendicular to the target straight line. The fourth straight line passes through the first target reference point and is perpendicular to the target straight line. The first target reference point is a point on the target straight line that moves forward along the direction of movement of the i-th first object in the reference frame. The distance between the first target reference point and the third straight line is a second preset value.

4. The radar data processing method according to claim 1, characterized in that, The step of obtaining the stitched object of the i-th first object in the current frame from the second object based on the radar data of the following object in the reference frame and the radar data of the i-th first object in the reference frame includes: If the following object is obtained, the target position of the i-th first object in the current frame is predicted based on the radar data of the following object in the reference frame and the radar data of the i-th first object in the reference frame. Based on the target position, obtain the i-th splicing object of the first object in the current frame from the second object.

5. The radar data processing method according to claim 4, characterized in that, Radar data includes the object's position and velocity; The step of predicting the target position of the i-th first object in the current frame based on the radar data of the following object in the reference frame and the radar data of the i-th first object in the reference frame when the following object is obtained includes: When the car-following object is obtained, the first acceleration of the i-th first object in the reference frame is calculated based on the speed and position of the car-following object in the reference frame, the speed and position of the i-th first object in the reference frame, and the predetermined car-following model. Calculate the second acceleration of the i-th first object in the reference frame, wherein the second acceleration is the acceleration generated by the resultant force of the boundary force and the repulsive force acting on the first object in the reference frame; Based on the first acceleration and the second acceleration, determine the combined acceleration of the i-th first object in the reference frame; Based on the combined acceleration, predict the target position of the i-th first object in the current frame.

6. The method according to claim 4, characterized in that, The method further includes: If the following object is not obtained, the first acceleration of the i-th first object is set to the target preset value; Calculate the second acceleration of the i-th first object in the reference frame, wherein the second acceleration is the acceleration generated by the resultant force of the boundary force and the repulsive force acting on the first object in the reference frame; Based on the first acceleration and the second acceleration, determine the combined acceleration of the i-th first object in the reference frame; Based on the combined acceleration, predict the target position of the i-th first object in the current frame.

7. The method according to claim 5 or 6, characterized in that, The calculation of the second acceleration of the i-th first object in the reference frame includes: Determine the boundary force exerted on the i-th first object in the reference frame; Obtain the i-th interference-applying object of the first object in the reference frame; Obtain the repulsive force exerted by each of the disturbances on the i-th first object; Calculate the resultant force of the boundary forces and each of the repulsive forces; Calculate the second acceleration based on the resultant force.

8. The method according to claim 7, characterized in that, Obtaining the i-th interference-applying object of the first object in the reference frame includes: Using the second target reference point as the center and a preset length as the radius, a semicircle is obtained in which the boundary line passes through the position of the i-th first object in the reference frame, wherein the second target reference point is a point along the movement direction of the i-th first object; The object located within the semi-circular area in the reference frame is identified as the object to which the interference is applied.

9. The radar data processing method according to claim 4, characterized in that, Radar data includes object type, velocity, position, and direction of motion; The step of obtaining the i-th splicing object of the first object in the current frame according to the target position includes: Obtain objects from the second object that meet preset conditions as candidate splicing objects. The preset conditions include that the object type is the same as the object type of the i-th first object, the deviation angle between the movement direction and the movement direction of the i-th first object is less than a third preset value, and the deviation between the speed and the speed of the i-th first object is less than a fourth preset value. Among the candidate splicing objects, the object that is closest to the i-th first object is obtained, and is used as the splicing object of the i-th first object in the current frame.

10. The method according to claim 4, characterized in that, The method further includes: When the stitching object is not obtained, the target position is determined as the position of the i-th first object in the radar data of the current frame, and the i-th first object is recorded as having failed to stitch once. When the number of times the first object i fails to be successfully spliced ​​reaches a preset number, stop predicting the position of the first object i.

11. A radar data processing device, characterized in that, The device includes: A radar data acquisition module is used to acquire radar data of the current frame and radar data of a reference frame, wherein the reference frame is the frame preceding the current frame, and the radar data includes radar data of a radar detecting at least one object. The first object acquisition module is used to acquire a first object that exists in the reference frame but does not exist in the current frame; The car-following object acquisition module is used to acquire the i-th car-following object of the first object in the reference frame, where i is an integer from 1 to N, and N is the total number of the first objects; The splicing object acquisition module is used to, when the following object is acquired, acquire the splicing object of the i-th first object in the current frame based on the radar data of the following object in the reference frame and the radar data of the i-th first object in the reference frame. The splicing object is determined based on the target position of the i-th first object in the current frame. The target position is predicted based on the first acceleration of the i-th first object in the reference frame determined by the following model and the second acceleration of the i-th first object in the reference frame determined based on the social force model. The social force model is a traffic flow model that considers the driving direction, target direction, and interference from surrounding motor vehicles, pedestrians, and non-motor vehicles. The determination module is used to determine, when the stitching object is obtained, the radar data of the stitching object in the current frame as the radar data of the i-th first object in the current frame.

12. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; A processor, when executing a program stored in memory, implements the steps of the method described in any one of claims 1-10.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-10.

Images