Course angle calculation method based on target tracking, program product and equipment

Abstract

The invention discloses a course angle calculation method based on target tracking, a program product, radar equipment and vehicle equipment. The method comprises the following steps: acquiring a reference course angle of a track of a target in a current frame; acquiring associated point cloud data of the track in the current frame, and calculating a point cloud surrounding course angle based on the associated point cloud data; calculating a speed course angle of the track in the current frame; calculating a fitting course angle based on the associated point cloud data; determining a target course angle of the track based on the reference course angle, the point cloud surrounding course angle, the speed course angle and the fitting course angle; and obtaining the course angle of the track in the current frame based on the target course angle.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving technology, and in particular to a heading angle calculation method based on target tracking, a computer program product, a radar device and a vehicle device. Background Art

[0002] In the field of vehicle driving technology, autonomous driving perception applications play an increasingly important role. For example, visual perception and laser radar can perceive the environment in front of the vehicle. Millimeter-wave radar is also used in the existing technology to perceive the environment. Compared with visual perception and laser radar, millimeter-wave radar has the advantages of low hardware cost and ignoring the influence of weather and environmental factors. However, in order to obtain higher-precision target position, speed, classification and other related attributes, these are not only related to the front-end hardware design and point cloud perception algorithm, but also closely related to the back-end data processing. When the vehicle is driving automatically, the accuracy of the heading angle is directly related to the accuracy of related attributes, such as precise vehicle control, path planning, vehicle obstacle avoidance, target position positioning and many other attributes. The road conditions during the vehicle's driving process are generally complex, which will affect the calculation of the heading angle, making the calculated heading angle less accurate. Summary of the invention

[0003] In order to solve the existing technical problems, the present invention provides a heading angle calculation method based on target tracking, a computer program product, a radar device and a vehicle device, which can improve the calculation accuracy of the heading angle.

[0004] In a first aspect, a method for calculating a heading angle based on target tracking is provided, comprising: obtaining a reference heading angle of a target track in a current frame; obtaining associated point cloud data of the track in the current frame, and calculating a point cloud enclosed heading angle based on the associated point cloud data; calculating a speed heading angle of the track in the current frame; calculating a fitted heading angle based on the associated point cloud data; determining a target heading angle of the track based on the reference heading angle, the point cloud enclosed heading angle, the speed heading angle and the fitted heading angle; and obtaining a heading angle of the track in the current frame based on the target heading angle.

[0005] In a second aspect, a computer program product is provided, comprising a computer program, wherein when the computer program is executed by a processor, the method for calculating a heading angle based on target tracking as described in any embodiment of the present application is implemented.

[0006] In a third aspect, a radar device is provided, comprising a storage unit and a processing unit, wherein the storage unit stores a computer program, and when the computer program is executed by the processing unit, the processing unit executes a heading angle calculation method based on target tracking as described in any embodiment of the present application.

[0007] In a fourth aspect, a vehicle device is provided, comprising the radar device described in any embodiment of the present application; or comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the heading angle calculation method based on target tracking as described in any embodiment of the present application.

[0008] In the embodiment of the present application, a reference heading angle of the target's track in the current frame is obtained, so that the target heading angle that meets the scene can be selected by referring to the reference heading angle later; the associated point cloud data of the target's track is obtained, and the associated point cloud data is obtained by tracking the point cloud data based on the filtering track algorithm, including the data in the historical frame of the current frame. Therefore, based on the associated point cloud data, an alternative point cloud enclosing heading angle and an alternative fitting heading angle can be calculated, and the speed heading angle of the track can also be calculated based on the motion data of the target. Therefore, based on different data, multiple alternative heading angles can be calculated, and then the reference heading angle is used as a reference benchmark, and the heading angle that meets the motion situation of the current scene is selected from the multiple alternative heading angles as the target heading angle of the target. In this way, the target heading angle of each target can be obtained more accurately during the tracking process, the calculation accuracy of the heading angle is improved, and there is no need to install additional hardware to assist in calculating the heading angle, thereby saving costs. Since the appropriate target heading angle can be selected in time according to the running situation of each target in the current scene during the tracking process, the motion state of each target can be accurately predicted, thereby improving driving safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 A diagram of an application environment of a method for calculating a heading angle based on target tracking in an embodiment;

[0010] Figure 2 is a schematic diagram of a vehicle coordinate system in one embodiment;

[0011] Figure 3 is a schematic diagram of a vehicle coordinate system in another embodiment;

[0012] Figure 4 is a flow chart of a method for calculating a heading angle based on target tracking in one embodiment;

[0013] Figure 5 is a schematic diagram of a heading angle of a track in one embodiment;

[0014] Figure 6 It is a flow chart of calculating the point cloud enclosing heading angle in a heading angle calculation method based on target tracking in one embodiment;

[0015] Figure 7 A flow chart of determining a target heading angle in a heading angle calculation method based on target tracking in an embodiment;

[0016] Figure 8is a schematic diagram of a heading angle calculation device based on target tracking in one embodiment;

[0017] Fig. 9 is a schematic diagram of a vehicle device in an embodiment;

[0018] Fig.10 Schematic diagram of a radar device in one embodiment. DETAILED DESCRIPTION

[0019] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by those skilled in the art of the technical field of the present invention. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the scope of protection of the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0021] In the following description, reference is made to “some embodiments”, which describes a subset of all possible embodiments, but it should be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0022] See also Figure 1 , is an application environment diagram of a heading angle calculation method based on target tracking in one embodiment. The heading angle calculation method based on target tracking is applied to a vehicle device 10, and the vehicle device 10 includes a display terminal 11, a radar device 12 and a controller 13. The radar device 12 is used to collect point cloud data of the front environment during the driving of the vehicle device 10, and the controller 13 generates a map indicating the drivable area based on the point cloud data collected by the radar device 12, so that the vehicle device 10 can accurately drive according to the map. In some application scenarios, the heading angle calculation method based on target tracking can be implemented on the radar device 12, and the radar device 12 can be directly installed in the vehicle device 10 for use, and the radar device 12 communicates with the vehicle device 10.

[0023] The radar device 12 includes but is not limited to a combination of one or more of the following sensors: 4D millimeter wave radar, 3D millimeter wave radar, other types of three-dimensional or four-dimensional radar devices. The radar device 12 is installed at the front center of the vehicle device 10 and faces the front of the vehicle device 10 to collect environmental data in front of the vehicle device 10. For example, the radar device 12 is installed at the center of the front bumper of the vehicle device 10. It is understandable that the vehicle device 10 may also include other types of sensors, including but not limited to multi-spectral vision sensors, environmental perception sensors, and motion posture sensors. The multi-spectral sensor includes but is not limited to a combination of one or more of the following sensors: thermal imaging sensors, visible light image sensors, millimeter wave sensors, lidar sensors, infrared thermal imaging sensors, and depth sensors. Environmental perception sensors include but are not limited to one or more combinations of the following sensors, such as brightness sensors, temperature sensors, and haze sensors. Motion posture sensors include but are not limited to one or more of the following combinations: inertial sensor (Inertial Measurement Unit, IMU), velocity sensor, acceleration sensor, gyroscope sensor, geomagnetic sensor, rotation vector sensor, steering wheel angle sensor, level sensor, tilt sensor, vibration sensor, displacement sensor and gravity sensor, etc.

[0024] The screen is the surface on which the display terminal 11 displays images. The screen is arranged in front of the driver so that the driver can view the images. The display terminal 11 includes but is not limited to displays such as liquid crystal displays, projectors with projector screens, etc. The display terminal 11 can be a display located on the central control of the vehicle device 10, or a projector arranged at the rear of the vehicle device 10 to project images onto the projector screen.

[0025] The controller 13 may be one or more. When there are multiple controllers 13, the multiple controllers may be integrated on one chip or independently set on each chip. The vehicle device 10 is a device installed on any type of mobile body, such as a vehicle, an electric vehicle, a hybrid electric vehicle, a motorcycle, a bicycle, a personal mobile device, an airplane, a drone, a ship or a robot, etc. In some embodiments, the controller 13 and the display terminal 11 may be integrated.

[0026] In the prior art, the Kalman filter algorithm is often used in millimeter-wave radar back-end data processing technology, in which the correlation and matching process between the track and the point cloud and how to obtain equivalent physical measurement information from multiple point cloud targets are technical difficulties in multi-target tracking. A common correlation matching method is to project the point cloud position data onto the heading angle direction of the track, and determine whether the association is successful by judging whether the difference between the point cloud projection and the track attributes meets the track correlation threshold. The method for calculating the heading angle in the prior art is difficult when the target is moving at a low speed due to Slight fluctuations in the function characteristic Vx will cause large jitters in the result of the heading angle. When the track tracking position attribute jitters or the target performs high-maneuverability movements (turns or U-turns), the real-time performance of the heading angle extracted by the motion trajectory is difficult to ensure, and the heading angle extracted by the motion trajectory cannot well describe the target's motion direction. Therefore, the heading angle calculation method based on target tracking provided in the embodiment of the present application solves the above problems.

[0027] In order to facilitate the subsequent description of the embodiments of the present application, the vehicle coordinate system is first described. Figure 2 As shown, it is a schematic diagram of the vehicle coordinate system in one embodiment, the x-axis direction is the longitudinal direction, which is along the center line of the vehicle device 10 and points to the forward direction of the vehicle device. The y-axis direction is the transverse direction, which is perpendicular to the longitudinal direction and located on the same plane, and the y-axis direction points from the right side of the vehicle device to the left side. The z-axis direction is the pitch direction, that is, the direction perpendicular to the x-axis and the y-axis, and points from the bottom side of the vehicle device to the top side. It can be understood that the coordinate origin position can be set at different positions as needed. Figure 3 As shown, Figure 3 Schematic diagram of a vehicle coordinate system in another embodiment, the radar device is arranged at the center of the front bumper of the vehicle device, and the coordinate origin position is arranged at the center of the front bumper.

[0028] See also Figure 4 , is a flow chart of a method for calculating a heading angle based on target tracking provided in an embodiment of the present application. The method for calculating a heading angle based on target tracking comprises the following steps:

[0029] S11. Obtain the reference heading angle of the target track in the current frame.

[0030] In this embodiment, during the driving process of the vehicle, the radar device sends electromagnetic waves to the current scene in front of the vehicle. After the electromagnetic waves encounter the target in the current scene, they are transmitted and reflected within a certain period of time to obtain the corresponding phase difference. These phase differences are converted to obtain the point cloud data of each target in the current scene. After processing the point cloud data frame by frame through the track filtering algorithm, the position, speed, acceleration, etc. of each target in the current scene can be obtained. The heading angle is an important attribute in the track filtering algorithm. The heading angle can be used to estimate the speed, estimate the position, estimate the size of each target in the current scene, estimate the direction of travel of the moving target, etc. Relative to the vehicle coordinate system, the heading angle is the angle between the trajectory of the target and the positive direction of the longitudinal direction.

[0031] In this embodiment, the data processing of the millimeter wave radar is discrete, and the heading angle is calculated approximately every preset interval (such as 100ms), that is, the heading angle is calculated once for each target in each frame. The reference heading angle is used to calculate the heading angle of the target in the current frame. The reference heading angle can be calculated based on the motion data of the target in the historical frame. In this way, a more accurate reference heading angle can be obtained based on the data in the historical frame, which is convenient for subsequent more accurate calculation of the target heading angle of the track. The tracked target can be one or more. If there are multiple targets to be tracked, the heading angle calculation method based on target tracking in the embodiment of the present application can be executed separately to obtain the heading angle of each target.

[0032] In this embodiment, the track is used to describe the motion path of each target. Generally, when tracking a target, a track will be created for a newly tracked target. When the track is just created, it is the initial stage of the track. In each frame, the track filtering algorithm will confirm the track. Once confirmed, the life cycle of the track is accumulated once. In this way, when the life cycle count of the track is greater than the preset count threshold, the newly created track is set as a confirmed track. If the radar does not detect the relevant point cloud of the track, the track will be kept in the previous frame state, that is, the track holding state. The life cycle of the track can be used to determine the existence time of the track. The longer the existence time, the higher the possibility of the existence of the target.

[0033] S12: Obtain associated point cloud data of the track in the current frame, and calculate the point cloud enclosing heading angle based on the associated point cloud data.

[0034] In this embodiment, the associated point cloud data is the point cloud data of the target tracked and matched according to the track filtering algorithm. For example, in the current frame, a car target is captured by the electromagnetic waves of the radar device, then the electromagnetic waves reflected by some positions on the surface of the car target will be obtained, and these electromagnetic waves are processed as point cloud data. The track filtering processing algorithm continuously tracks the target, and the historical information of the car target can be obtained through historical information. The point cloud data in the current frame is matched with the historical information of the tracking, and the process of judging whether it is the same target is the association process, that is, when calculating the heading angle, the association module in the track filtering algorithm has completed the association matching between the point cloud and the historical information of the tracking, so the association module in the track filtering algorithm can be directly used to obtain the associated point cloud data.

[0035] The point cloud bounding heading angle is a heading angle calculated based on the bounding box obtained based on the associated point cloud data.

[0036] S13, calculating the speed and heading angle of the track in the current frame.

[0037] In this embodiment, in the vehicle coordinate system, the speed heading angle is the angle between the speed of the target and the longitudinal direction.

[0038] S14. Calculate a fitting heading angle based on the associated point cloud data.

[0039] In this embodiment, the associated point cloud data includes but is not limited to the azimuth angle, radial velocity relative to the ground, etc. of the point cloud, and the fitted heading angle is a heading angle obtained by fitting the associated point cloud data.

[0040] S15. Determine a target heading angle of the track based on the reference heading angle, the point cloud enclosing heading angle, the speed heading angle and the fitting heading angle.

[0041] In this embodiment, the reference heading angle is calculated based on the data in the historical frame, and the target heading angle is determined with reference to the reference heading angle from a plurality of alternative heading angles, namely, the point cloud enclosed heading angle, the speed heading angle and the fitted heading angle, which can better reflect the motion state of the target in the current scene, thereby being able to more accurately calculate the heading angle of the target's track.

[0042] S16. Based on the target heading angle, obtain the heading angle of the track in the current frame.

[0043] In this embodiment, the target heading angle is normalized to obtain the heading angle of the track in the current frame. For example, after the target heading angle is normalized, the result is limited to the range of [-180°180]. Figure 5 As shown, Figure 5Schematic diagram of the heading angle of a track in an embodiment, where the heading angle of the track in the current frame is θ.

[0044] In the above embodiment, a reference heading angle of the target track in the current frame is obtained, so that the target heading angle that meets the scene can be selected by referring to the reference heading angle later; the associated point cloud data of the target track is obtained, and the associated point cloud data is obtained by tracking the point cloud data based on the filtering track algorithm, including the data in the historical frame of the current frame. Therefore, based on the associated point cloud data, the alternative point cloud enclosing heading angle and the alternative fitting heading angle can be calculated, and the speed heading angle of the track can also be calculated based on the motion data of the target. Therefore, based on different data, multiple alternative heading angles can be calculated, and then the reference heading angle is used as a reference benchmark. From the multiple alternative heading angles, the heading angle that meets the motion situation of the current scene is selected as the target heading angle of the target. In this way, the target heading angle of each target can be obtained more accurately during the tracking process, the calculation accuracy of the heading angle is improved, and there is no need to install additional hardware to assist in calculating the heading angle, thereby saving costs. Since the appropriate target heading angle can be selected in time according to the running situation of each target in the current scene during the tracking process, the motion state of each target can be accurately predicted, thereby improving driving safety.

[0045] In some embodiments, before obtaining the reference heading angle of the track of the target in the current frame, the method further includes:

[0046] Acquire tracking data of the target in the current frame, the tracking data comprising at least one of the following: a tracking attribute indicating whether the target is a moving target, and a ground speed of the track;

[0047] Based on the tracking data, determining whether to update a reference heading angle corresponding to the target;

[0048] When the tracking attribute of the target indicates that the target is a moving target and / or the ground speed of the track is greater than a preset reference speed value, it is determined to update the reference heading angle corresponding to the target.

[0049] In this embodiment, radar equipment can be used to obtain multiple frames of radar data, and target tracking filtering is performed on the multiple frames of radar data to obtain each target and its tracking data. Since there are stationary targets such as guardrails, roadside trees, and flower beds on the road, there are also moving targets such as moving cars and battery vehicles. The heading angle is used to describe the approximate direction of movement of a target, so it is meaningless for stationary targets such as guardrails and trees on the road, so only the heading angles of those moving targets need to be updated.

[0050] In the above embodiment, for each target in the current frame, not every target will update the reference heading angle. By tracking data, screening moving targets, and updating the heading angle of the moving targets, it is convenient to accurately judge the moving direction of the moving targets in the subsequent process, so as to improve driving safety.

[0051] In some embodiments, obtaining a reference heading angle of the track of the target in the current frame includes:

[0052] When a reference heading angle calculation condition is met, obtaining a current heading angle of the track in the current frame and a heading angle of the track in a previous preset frame of the current frame, and calculating a first reference heading angle based on the current heading angle and the heading angle of the track in the previous preset frame;

[0053] Acquire the ground speed of the track in the current frame, and calculate a second reference heading angle according to the ground speed of the track;

[0054] A reference heading angle difference between the first reference heading angle and the second reference heading angle is calculated, and based on the reference heading angle difference, the reference heading angle is determined from the first reference heading angle and the second reference heading angle.

[0055] Optionally, determining the reference heading angle from the first reference heading angle and the second reference heading angle based on the reference heading angle difference includes:

[0056]

[0057]

[0058] Where headingRef represents the reference heading angle, headingRef 1 Indicates the first reference heading angle, headingRef 2 represents the second reference heading angle, xVel represents the component of the ground velocity of the track in the X-axis direction, yVel represents the component of the ground velocity of the track in the Y-axis direction, wherein the Y-axis direction is perpendicular to the X-axis direction and is located in the same plane, and the X-axis direction is along the centerline of the vehicle equipment and points to the forward direction of the vehicle equipment, headingRef diff Indicates the reference heading angle difference, angleThre 1 Indicates the preset reference angle difference threshold.

[0059] In this embodiment, since the radar device is installed on the vehicle device, the radar device collects data once every preset interval, for example, the time interval is 100ms. During this time interval, the vehicle device is also moving. Since the vehicle device may move the steering wheel or other operations, it will cause a rotation of the vehicle coordinate system. Therefore, in the current frame, it is necessary to compensate the heading angle of the track in the previous preset frame (such as the previous frame) of the current frame according to the vehicle information to obtain the angle corresponding to the target in the current vehicle coordinate system, that is, the first reference heading angle. The vehicle information includes but is not limited to the vehicle yaw rate (yawRate) and the like. In an optional implementation, the reference heading angle calculation condition includes but is not limited to at least one of the following: the ground speed of the track is greater than the first speed threshold, the track is in a confirmed state and the life cycle is greater than the preset count threshold. If the reference heading angle calculation condition can be met in the current frame, the state flag of the reference heading angle is updated to valid. If the reference heading angle calculation condition cannot be met in the current frame, the state flag of the reference heading angle is updated to invalid. The state flag of the reference heading angle is updated in real time in each frame to identify whether the reference heading angle in the frame is available.

[0060] In this embodiment, the second reference heading angle is also calculated based on the ground speed of the track, that is, the movement speed of the target, and the difference between the first reference heading angle and the second reference heading angle is compared to obtain the angle difference; when the angle difference is large, it means that the direction of the movement speed of the target here has changed greatly, and it is more accurate to use the second reference heading angle as the reference heading angle; when the angle difference is small, it means that the direction of the movement speed of the target has changed little at this time, and the heading angle of the track in the historical frame can be combined to obtain the reference heading angle in the current frame, that is, the first reference heading angle is used.

[0061] In the above embodiment, a first reference heading angle is calculated based on the heading angle of the track in the historical frame, and a second reference heading angle is calculated based on the moving speed of the target. The moving state of the target in the current frame is determined based on the angle difference between the first reference heading angle and the second reference heading angle. A suitable reference heading angle is selected based on the moving state of the target to better fit the moving scene of the target, thereby accurately predicting the moving trajectory of the target to improve driving safety.

[0062] In some embodiments, Figure 6 As shown, Figure 6 FIG. 1 is a flow chart of calculating the point cloud enclosed heading angle in a heading angle calculation method based on target tracking in an embodiment, wherein step S12 specifically includes:

[0063] S121. When the point cloud enclosing heading angle calculation conditions are met, the current feature angle is obtained within a preset angle range and with a preset angle step size.

[0064] In this embodiment, the point cloud enclosed heading angle calculation conditions include that the number of associated point clouds is greater than the preset number of point clouds, and the ground speed of the track is less than the second speed threshold. The preset angle range is 0 to 90°, for example, the preset angle step is 1°, and a current feature angle is obtained in each cycle, and the following steps are performed until the minimum fitting loss error is determined. If the point cloud enclosed heading angle calculation conditions can be met in the current frame, the state flag of the point cloud enclosed heading angle is updated to valid. If the point cloud enclosed heading angle calculation conditions cannot be met in the current frame, the state flag of the point cloud enclosed heading angle is updated to invalid. The state flag of the point cloud enclosed heading angle is updated in real time in each frame to indicate whether the point cloud enclosed heading angle in the frame is available.

[0065] S122: Rotate the position coordinates of the associated point cloud in the associated point cloud data according to the current feature angle to obtain position coordinate limits of the rotated point cloud bounding box.

[0066] In this embodiment, the two-dimensional coordinate information (xpos, ypos) of the associated point cloud is rotated according to the current feature angle to obtain the position coordinates of the rotated point cloud bounding box. In this way, each point cloud is rotated to form a rotated point cloud bounding box, thereby obtaining the position coordinate limits of the point cloud bounding box, that is, the upper and lower limits of the coordinates (xpos min ,xpos max ,ypos min ,ypos max ). xpos is the horizontal coordinate of the associated point cloud, ypos is the vertical coordinate of the associated point cloud, min is the minimum value of the horizontal coordinate of the point cloud bounding box, xpos max ypos is the maximum horizontal coordinate of the point cloud bounding box min ypos is the minimum ordinate value of the point cloud bounding box max It is the maximum value of the vertical coordinate of the point cloud bounding box.

[0067] S123. Calculate size data of the rotated point cloud bounding box based on the position coordinate limit of the rotated point cloud bounding box.

[0068] In this embodiment, the length and width of the point cloud bounding box

[0069]

[0070] in is the length of the point cloud bounding box, box width is the width of the point cloud bounding box.

[0071] S124. Calculate the fitting loss error corresponding to the current feature angle based on the position coordinates of the associated point cloud and the size data.

[0072] In this embodiment, the expression of the fitting loss error is as follows:

[0073] Loss

[0074] =min{(xpos-xpos min ),(xpos max ―xpos),(ypos―ypos_min),(ypos max ―ypos)}

[0075] Where Loss represents the fitting loss error. The fitting loss error corresponding to the current feature angle is obtained by accumulating the fitting loss errors corresponding to all point clouds, that is, Loss track =∑Loss, where Loss track Represents the fitting loss error corresponding to the current feature angle.

[0076] S125. When the fitting loss error corresponding to the current feature angle is less than the current fitting loss error, the fitting loss error corresponding to the current feature angle is updated to the current fitting loss error, and the current feature angle is continuously acquired in a loop until the minimum fitting loss error is acquired, and the point cloud enclosing heading angle is determined based on the feature angle corresponding to the minimum fitting loss error.

[0077] In an optional embodiment, when a first enclosing heading angle calculation condition is met, the point cloud enclosing heading angle is determined to be the characteristic angle corresponding to the minimum fitting loss error minus 90°, wherein the first enclosing heading angle calculation condition includes that the length of the rotated point cloud enclosing frame calculated based on the characteristic angle corresponding to the minimum fitting loss error is less than the width;

[0078] Under the condition that the second enclosing heading angle calculation condition is met, determining the point cloud enclosing heading angle as the characteristic angle corresponding to the minimum fitting loss error plus 180°, wherein the second enclosing heading angle calculation condition includes: when the angle formed by the characteristic angle corresponding to the minimum fitting loss error and the radial velocity of the associated point cloud is greater than 90°;

[0079] When the first enclosing heading angle calculation condition or the second enclosing heading angle calculation condition is not met, the point cloud enclosing heading angle is determined to be a characteristic angle corresponding to the minimum fitting loss error.

[0080] In this embodiment, the loss function Loss is obtained by trying to calculate at various angles within the preset angle range. trackThe angle corresponding to the minimum is used to obtain the required optimal feature angle, but the optimal feature angle may not necessarily be the point cloud enclosing heading angle, but may also be the complementary angle or complementary angle of the heading angle. Therefore, after obtaining the optimal feature angle, further processing is required to obtain the point cloud enclosing heading angle. The point cloud enclosing heading angle is calculated based on the bounding box formed by the associated point cloud of the target.

[0081] In the above embodiment, by acquiring the associated point cloud data of the target, calculating the fitting loss errors at multiple angles, finding the minimum fitting loss error, and obtaining the angle corresponding to the minimum fitting loss error, the point cloud enclosed heading angle is calculated according to the angle corresponding to the minimum fitting loss error, so that the accuracy of the point cloud enclosed heading angle is higher. In this way, when the point cloud enclosed heading angle is used as an alternative heading angle, the calculation accuracy of the target heading angle can be improved, thereby improving driving safety.

[0082] In some embodiments, calculating the speed and heading angle of the track in the current frame includes:

[0083] When the speed and heading angle calculation conditions are met, obtaining the ground speed of the track in the previous preset frame of the current frame, and obtaining the ground speed of the track in the current frame;

[0084] Calculating a weighted ground speed of the track in the current frame based on the ground speed of the track in the previous preset frame and the ground speed of the track in the current frame;

[0085] The speed and heading angle of the track in the current frame are calculated based on the weighted ground speed of the track in the current frame.

[0086] In an optional implementation, the calculating the speed and heading angle of the track in the current frame based on the weighted ground speed of the track in the current frame includes:

[0087] Based on the weighted ground speed of the track in the current frame, the speed and heading angle of the track in the current frame is calculated using the speed and heading angle calculation formula. The speed and heading angle calculation formula of the track is as follows:

[0088]

[0089]

[0090] Wherein speedHeading represents the speed heading angle of the track, yVelFilter′ represents the component of the weighted ground speed of the track in the Y-axis direction, xVelFilter′ represents the component of the weighted ground speed of the track in the X-axis direction, xVelFiter represents the component of the ground speed of the track in the previous preset frame in the X-axis direction, yVelFiter represents the component of the ground speed of the track in the previous preset frame in the Y-axis direction, xVelAbs represents the component of the ground speed of the track in the current frame in the X-axis direction, yCelAbs represents the component of the ground speed of the track in the current frame in the Y-axis direction, α and β represent coefficients, the Y-axis direction is perpendicular to the X-axis direction and is located in the same plane, and the X-axis direction is along the center line of the vehicle equipment and points to the forward direction of the vehicle equipment.

[0091] In this embodiment, the speed and heading angle calculation condition includes that the confirmation state of the track is in the confirmed state, and the ground speed of the track is greater than the third speed threshold. If the speed and heading angle calculation condition is met in the current frame, the state flag of the speed and heading angle is updated to valid. If the speed and heading angle calculation condition is not met in the current frame, the state flag of the speed and heading angle is updated to invalid. The state flag of the speed and heading angle is updated in real time in each frame to indicate whether the speed and heading angle in the frame is available.

[0092] When the target is moving at a high speed, the speed and heading angle of the track can be used as an alternative heading angle to accurately reflect the target's motion state. Since the radar equipment collects data at a certain time interval, it is necessary to compensate the ground speed of the track in the current frame according to the ground speed of the track in the historical frame, so as to obtain a more accurate weighted ground speed of the track, and then calculate the alternative speed and heading angle in the current frame based on the weighted ground speed of the track.

[0093] In the above embodiment, when the target's moving speed is relatively high, the speed heading angle is obtained as an alternative heading angle, so that the target's heading angle can be calculated more accurately; and the ground speed of the track in the current frame is compensated according to the ground speed of the track in the historical frame, so that the obtained speed heading angle is more accurate.

[0094] In some embodiments, the calculating the fitting heading angle based on the associated point cloud data comprises:

[0095] When the fitting heading angle calculation condition is met, the azimuths of all associated point clouds are acquired based on the associated point cloud data, and a first matrix is ​​generated based on the azimuths of all associated point clouds;

[0096] Based on the associated point cloud data, obtaining radial velocities of all associated point clouds relative to the ground, and generating a second matrix based on the radial velocities of all associated point clouds relative to the ground;

[0097] Calculating a fitting speed of the track based on the first matrix and the second matrix;

[0098] When it is determined that the fitting speed of the track meets a preset fitting speed condition, the fitting heading angle is calculated based on the fitting speed of the track.

[0099] In an optional implementation, the first matrix is ​​represented by A, and the first matrix is ​​as follows:

[0100]

[0101] in Indicates the azimuth of the nth associated point cloud;

[0102] The second matrix is ​​denoted by Y, and the second matrix is ​​as follows:

[0103]

[0104] Where VradVelAbs n represents the radial velocity of the nth associated point cloud over the ground;

[0105] The fitting speed of the track is represented by V, and the fitting speed of the track is as follows:

[0106]

[0107] Wherein Vx represents the component of the fitting speed of the track in the X-axis direction, Vy represents the component of the fitting speed of the track in the Y-axis direction, and the X-axis direction is along the center line of the vehicle equipment and points to the forward direction of the vehicle equipment.

[0108] In this embodiment, the fitting heading angle calculation condition includes that the number of associated point clouds is greater than a preset number or the sum of the number of point clouds of the track in the current frame and the number of point clouds of the track in the previous preset frame is greater than a preset number. If the fitting heading angle calculation condition is met in the current frame, the state flag of the fitting heading angle is updated to valid. If the fitting heading angle calculation condition is not met in the current frame, the state flag of the fitting heading angle is updated to invalid. In an optional implementation, when the number of associated point clouds is less than a preset number, the azimuth and radial velocity of the point cloud of the track in the historical frame are obtained, and the corresponding function values ​​are calculated based on the azimuth and radial velocity of the point cloud of the track in the historical frame, and added to the matrix A and matrix Y respectively. By associating point cloud data or combining point cloud data of the track in the historical frame, the fitting speed of the track is predicted, and an alternative heading angle is calculated based on the fitting speed, so as to facilitate the subsequent accurate calculation of the heading angle of the track.

[0109] In the above embodiment, the fitting speed of the track is predicted based on the associated point cloud data, and an alternative heading angle, namely the fitting heading angle, is calculated based on the fitting speed, so as to facilitate the subsequent accurate calculation of the heading angle of the track. In this way, when the fitting heading angle is used as an alternative heading angle, the calculation accuracy of the target heading angle can be improved, thereby improving driving safety.

[0110] In some embodiments, Figure 7 This is a flow chart of determining a target heading angle in a heading angle calculation method based on target tracking in an embodiment, wherein step S15 specifically includes:

[0111] S151. Based on the reference heading angle, the point cloud enclosing heading angle is compensated to obtain an extended enclosing heading angle, the speed heading angle is compensated to obtain an extended speed heading angle, and the fitting heading angle is compensated to obtain an extended fitting heading angle.

[0112] In this embodiment, the setting range of the heading angle is between [-180°180°], so the polarity directions of the calculated multiple alternative heading angles, i.e., the point cloud enclosed heading angle, the velocity heading angle, and the fitted heading angle may be different from the reference heading angle, so it is necessary to compensate for each alternative heading angle separately. When the reference heading angle is a positive angle, i.e., between [0°180°], the polarity of the multiple alternative heading angles should also be between [0°180°]. When the reference heading angle is a negative angle, i.e., between [-180°0°], the polarity of the multiple alternative heading angles should also be between [-180°0°].

[0113] S152. Calculate the absolute difference between the extended enclosed heading angle and the reference heading angle to obtain a first absolute difference, calculate the absolute difference between the extended speed heading angle and the reference heading angle to obtain a second absolute difference, and calculate the absolute difference between the extended fitting heading angle and the reference heading angle to obtain a third absolute difference.

[0114] In this embodiment, the first absolute difference obbRef diff =|obbHeadingExpand―headingRef|

[0115] The second absolute difference speedRef diff =|speedHeadingExpand―headingRef|

[0116] The third absolute difference leastRef diff =|leastHeadingExpand―headingRef|

[0117] Among them, obbHeadingExpand is the expanded enclosing heading angle, speedHeadingExpand is the expanded speed heading angle, leastHeadingExpand is the expanded fitting heading angle, and headingRef is the reference heading angle.

[0118] S153: When the state indicator indicating that the point cloud encloses the heading angle is valid, determine the target heading angle using a first screening strategy based on the first absolute difference, the second absolute difference, and the third absolute difference.

[0119] In an optional embodiment, when the state indicator of the point cloud enclosing heading angle indicates that it is valid, it includes the following:

[0120] (1) When a first preset condition is met, if the first absolute difference is less than the second absolute difference, the extended encircling heading angle is determined as the target heading angle; if the first absolute difference is greater than or equal to the second absolute difference, the extended speed heading angle is determined as the target heading angle, wherein the first preset condition is that the status indicator indication of the speed heading angle is valid.

[0121] In this embodiment, the target heading angle headingAngle is:

[0122]

[0123] (2) When a second preset condition is met, if the first absolute difference is less than the third absolute difference, the extended enclosed heading angle is determined as the target heading angle; if the first absolute difference is greater than or equal to the third absolute difference, the extended fitted heading angle is determined as the target heading angle, wherein the second preset condition is that the status mark of the speed heading angle indicates invalid, and the status mark of the fitted heading angle indicates valid.

[0124] In this embodiment, the target heading angle headingAngle is:

[0125]

[0126] (3) Under the third preset condition, if the first absolute difference is greater than the preset angle difference, the reference heading angle is determined as the target heading angle; if the first absolute difference is less than or equal to the preset angle difference, the extended encircled heading angle is determined as the target heading angle, and the third preset condition is that the track is in a confirmed state and the track is in a non-stationary state.

[0127] In this embodiment, the target heading angle headingAngle is:

[0128]

[0129] (4) When the first preset condition, the second preset condition, or the third preset condition is not satisfied, the extended encircling heading angle is determined as the target heading angle.

[0130] In this embodiment, the target heading angle headingAngle is:

[0131] headingAngle=obbHeadingExpand

[0132] S154. When the state indicator indicating that the point cloud encloses the heading angle is invalid, determine the target heading angle using a second screening strategy based on the first absolute difference, the second absolute difference, and the third absolute difference.

[0133] In an optional implementation, when the state indicator of the point cloud enclosing heading angle indicates invalid, the method includes the following:

[0134] (1) When the first execution condition is met, if the third absolute difference is less than the second absolute difference, the extended fitting heading angle is determined as the target heading angle; if the third absolute difference is greater than or equal to the second absolute difference, the extended speed heading angle is determined as the target heading angle, wherein the first execution condition is that the status mark indication of the fitting heading angle is valid and the status mark indication of the speed heading angle is valid.

[0135] In this embodiment, the target heading angle headingAngle is:

[0136]

[0137] (2) When a second execution condition is met, the extended fitting heading angle is determined as the target heading angle, wherein the second execution condition is that the status mark of the fitting heading angle indicates that it is valid, and the status mark of the speed heading angle indicates that it is invalid.

[0138] In this embodiment, the target heading angle headingAngle is:

[0139] headingAngle=leastHeadingExpand

[0140] (3) When a third execution condition is met, the extended speed heading angle is determined as the target heading angle; wherein the third execution condition is that the status mark of the fitting heading angle indicates invalid and the status mark of the speed heading angle indicates valid.

[0141] In this embodiment, the target heading angle headingAngle is:

[0142] headingAngle=speedHeadingExpand

[0143] (4) When the fourth execution condition is met, the heading angle of the track in the previous preset frame is obtained, and the target heading angle is predicted based on the heading angle of the track in the previous preset frame, and the fourth execution condition is that the status mark of the fitting heading angle indicates invalid and the status mark of the speed heading angle indicates invalid.

[0144] In this embodiment, the target heading angle headingAngleRad is:

[0145] headingAngleRad=headingAngleRad

[0146] Wherein headingAngleRad represents the heading angle predicted based on the heading angle of the track in the previous preset frame.

[0147] In the above embodiment, based on different data, multiple alternative heading angles can be calculated, and then the reference heading angle is used as a reference benchmark, various scene conditions are judged according to different threshold conditions, and the heading angle matching the scene conditions is selected from the multiple alternative heading angles under different scene conditions as the target heading angle of the target. In this way, the target heading angle of each target can be obtained more accurately during the tracking process, thereby improving the calculation accuracy of the heading angle; and the heading angle is selected only based on the threshold conditions, which makes the calculation more efficient; by providing multiple alternative heading angles for selection, the influence of the target heading angle on the track attributes can be reduced, the calculation accuracy of the heading angle can be improved, and the motion state of each target can be accurately predicted, thereby improving driving safety.

[0148] In some embodiments, the determining the target heading angle of the track based on the reference heading angle, the point cloud enclosed heading angle, the speed heading angle, and the fitted heading angle includes at least one of the following:

[0149] In the case where the moving speed of the target is in a fluctuating state, if the state identification indication of the speed and heading angle is valid, determining the target heading angle based on the speed and heading angle;

[0150] In the case where the target is in a U-turn motion state, if the state identification indication of the fitting heading angle is valid, the target heading angle is determined based on the fitting heading angle.

[0151] In the above embodiments, selecting a suitable heading angle according to the target's motion state, or selecting a heading angle according to the target's motion speed can more accurately reflect the target's motion scene. In this way, when the target is in motion or the trajectory suddenly changes, the suitable heading angle can be determined in real time, and the motion state of each target can be accurately predicted, thereby improving driving safety.

[0152] On the other hand, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the method for calculating a heading angle based on target tracking as described in any embodiment of the present application.

[0153] Among them, in the computer program product, an optional implementation form of the program module architecture of the computer program that implements each step of the target recognition method can be a heading angle calculation device based on target tracking.

[0154] See also Figure 8 An embodiment of the present application provides a heading angle calculation device based on target tracking, including: an acquisition module 81, used to obtain a reference heading angle of the target track in the current frame; the acquisition module 81 is also used to obtain the associated point cloud data of the track in the current frame, and calculate the point cloud enclosed heading angle based on the associated point cloud data; a calculation module 82 is used to calculate the speed heading angle of the track in the current frame; the calculation module 82 is also used to calculate the fitting heading angle based on the associated point cloud data; a determination module 83 is used to determine the target heading angle of the track based on the reference heading angle, the point cloud enclosed heading angle, the speed heading angle and the fitting heading angle; the determination module 83 is also used to obtain the heading angle of the track in the current frame based on the target heading angle.

[0155] Optionally, the acquisition module 81 is further used for:

[0156] Acquire tracking data of the target in the current frame, the tracking data comprising at least one of the following: a tracking attribute indicating whether the target is a moving target, and a ground speed of the track;

[0157] Based on the tracking data, determining whether to update a reference heading angle corresponding to the target;

[0158] When the tracking attribute of the target indicates that the target is a moving target and / or the ground speed of the track is greater than a preset reference speed value, it is determined to update the reference heading angle corresponding to the target.

[0159] Optionally, the acquisition module 81 is further used for:

[0160] When a reference heading angle calculation condition is met, obtaining a current heading angle of the track in the current frame and a heading angle of the track in a previous preset frame of the current frame, and calculating a first reference heading angle based on the current heading angle and the heading angle of the track in the previous preset frame;

[0161] Acquire the ground speed of the track in the current frame, and calculate a second reference heading angle according to the ground speed of the track;

[0162] A reference heading angle difference between the first reference heading angle and the second reference heading angle is calculated, and based on the reference heading angle difference, the reference heading angle is determined from the first reference heading angle and the second reference heading angle.

[0163] Optionally, the acquisition module 81 is further used for:

[0164]

[0165]

[0166] Where headingRef represents the reference heading angle, headingRef 1 Indicates the first reference heading angle, headingRef 2 represents the second reference heading angle, xVel represents the component of the ground velocity of the track in the X-axis direction, yVel represents the component of the ground velocity of the track in the Y-axis direction, wherein the Y-axis direction is perpendicular to the X-axis direction and is located in the same plane, and the X-axis direction is along the center line of the vehicle equipment and points to the forward direction of the vehicle equipment, Indicates the reference heading angle difference, angleThre 1 Indicates the preset reference angle difference threshold.

[0167] Optionally, the calculation module 82 is further used for:

[0168] When the point cloud enclosing heading angle calculation conditions are met, the current feature angle is obtained within the preset angle range and with the preset angle step size;

[0169] Rotating the position coordinates of the associated point cloud in the associated point cloud data according to the current feature angle to obtain the position coordinate limit of the rotated point cloud bounding box;

[0170] Calculating size data of the rotated point cloud bounding box based on the position coordinate limit of the rotated point cloud bounding box;

[0171] Calculating the fitting loss error corresponding to the current feature angle based on the position coordinates of the associated point cloud and the size data;

[0172] When the fitting loss error corresponding to the current feature angle is less than the current fitting loss error, the fitting loss error corresponding to the current feature angle is updated to the current fitting loss error, and the current feature angle is continuously acquired in a loop until the minimum fitting loss error is acquired, and the point cloud enclosing heading angle is determined based on the feature angle corresponding to the minimum fitting loss error.

[0173] Optionally, the calculation module 82 is further used for:

[0174] Under the condition that the first enclosing heading angle calculation condition is met, determining the point cloud enclosing heading angle to be the characteristic angle corresponding to the minimum fitting loss error minus 90°, wherein the first enclosing heading angle calculation condition includes that the length of the rotated point cloud enclosing frame calculated based on the characteristic angle corresponding to the minimum fitting loss error is less than the width;

[0175] Under the condition that the second enclosing heading angle calculation condition is met, determining the point cloud enclosing heading angle as the characteristic angle corresponding to the minimum fitting loss error plus 180°, wherein the second enclosing heading angle calculation condition includes: when the angle formed by the characteristic angle corresponding to the minimum fitting loss error and the radial velocity of the associated point cloud is greater than 90°;

[0176] When the first enclosing heading angle calculation condition or the second enclosing heading angle calculation condition is not met, the point cloud enclosing heading angle is determined to be a characteristic angle corresponding to the minimum fitting loss error.

[0177] Optionally, the calculation module 82 is further used for:

[0178] When the speed and heading angle calculation conditions are met, obtaining the ground speed of the track in the previous preset frame of the current frame, and obtaining the ground speed of the track in the current frame;

[0179] Calculating a weighted ground speed of the track in the current frame based on the ground speed of the track in the previous preset frame and the ground speed of the track in the current frame;

[0180] The speed and heading angle of the track in the current frame are calculated based on the weighted ground speed of the track in the current frame.

[0181] Optionally, the calculation module 82 is further used for:

[0182] Based on the weighted ground speed of the track in the current frame, the speed and heading angle of the track in the current frame is calculated using the speed and heading angle calculation formula. The speed and heading angle calculation formula of the track is as follows:

[0183]

[0184]

[0185] Wherein speedHeading represents the speed heading angle of the track, yVelFilter′ represents the component of the weighted ground speed of the track in the Y-axis direction, xVelFiter′ represents the component of the weighted ground speed of the track in the X-axis direction, xVelFiter represents the component of the ground speed of the track in the previous preset frame in the X-axis direction, yVelFilter represents the component of the ground speed of the track in the previous preset frame in the Y-axis direction, xVelAbs represents the component of the ground speed of the track in the current frame in the X-axis direction, yVelAbs represents the component of the ground speed of the track in the current frame in the Y-axis direction, α and β represent coefficients, the Y-axis direction is perpendicular to the X-axis direction and is located in the same plane, and the X-axis direction is along the center line of the vehicle equipment and points to the forward direction of the vehicle equipment.

[0186] Optionally, the calculation module 82 is further used for:

[0187] When the fitting heading angle calculation condition is met, the azimuths of all associated point clouds are acquired based on the associated point cloud data, and a first matrix is ​​generated based on the azimuths of all associated point clouds;

[0188] Based on the associated point cloud data, obtaining radial velocities of all associated point clouds relative to the ground, and generating a second matrix based on the radial velocities of all associated point clouds relative to the ground;

[0189] Calculating a fitting speed of the track based on the first matrix and the second matrix;

[0190] When it is determined that the fitting speed of the track meets a preset fitting speed condition, the fitting heading angle is calculated based on the fitting speed of the track.

[0191] Optionally, the first matrix is ​​represented by A, and the first matrix is ​​as follows:

[0192]

[0193] in Indicates the azimuth of the nth associated point cloud;

[0194] The second matrix is ​​denoted by Y, and the second matrix is ​​as follows:

[0195]

[0196] Where VradVelAbs n represents the radial velocity of the nth associated point cloud over the ground;

[0197] The fitting speed of the track is represented by V, and the fitting speed of the track is as follows:

[0198]

[0199] Wherein Vx represents the component of the fitting speed of the track in the X-axis direction, Vy represents the component of the fitting speed of the track in the Y-axis direction, and the X-axis direction is along the center line of the vehicle equipment and points to the forward direction of the vehicle equipment.

[0200] Optionally, the determination module 83 is further used for:

[0201] Based on the reference heading angle, the point cloud enclosing heading angle is compensated to obtain an extended enclosing heading angle, the speed heading angle is compensated to obtain an extended speed heading angle, and the fitting heading angle is compensated to obtain an extended fitting heading angle;

[0202] Calculating an absolute difference between the extended enclosing heading angle and the reference heading angle to obtain a first absolute difference, calculating an absolute difference between the extended speed heading angle and the reference heading angle to obtain a second absolute difference, and calculating an absolute difference between the extended fitting heading angle and the reference heading angle to obtain a third absolute difference;

[0203] In a case where the state indicator indicating that the point cloud encloses the heading angle is valid, determining the target heading angle by using a first screening strategy based on the first absolute difference, the second absolute difference, and the third absolute difference;

[0204] When the status indicator of the point cloud enclosed heading angle indicates invalid, the target heading angle is determined using a second screening strategy based on the first absolute difference, the second absolute difference and the third absolute difference.

[0205] Optionally, the determination module 83 is further used for:

[0206] When the state indicator of the point cloud enclosing heading angle indicates that it is valid, it includes the following:

[0207] When a first preset condition is met, if the first absolute difference is less than the second absolute difference, the extended enveloping heading angle is determined as the target heading angle; if the first absolute difference is greater than or equal to the second absolute difference, the extended speed heading angle is determined as the target heading angle, wherein the first preset condition is that the state identification indication of the speed heading angle is valid;

[0208] Under the condition that the second preset condition is met, if the first absolute difference is less than the third absolute difference, the extended enclosed heading angle is determined as the target heading angle, and if the first absolute difference is greater than or equal to the third absolute difference, the extended fitted heading angle is determined as the target heading angle, wherein the second preset condition is that the state identification indication of the speed heading angle is invalid, and the state identification indication of the fitted heading angle is valid;

[0209] Under the third preset condition, if the first absolute difference is greater than the preset angle difference, the reference heading angle is determined as the target heading angle; if the first absolute difference is less than or equal to the preset angle difference, the extended enveloping heading angle is determined as the target heading angle, and the third preset condition is that the track is in a confirmed state and the track is in a non-stationary state;

[0210] When the first preset condition, the second preset condition or the third preset condition is not satisfied, the extended enclosing heading angle is determined as the target heading angle.

[0211] Optionally, the determination module 83 is further used for:

[0212] When the state indicator of the point cloud enclosing heading angle indicates invalid, it includes the following:

[0213] When the first execution condition is met, if the third absolute difference is less than the second absolute difference, the extended fitting heading angle is determined as the target heading angle; if the third absolute difference is greater than or equal to the second absolute difference, the extended speed heading angle is determined as the target heading angle, wherein the first execution condition is that the state identification indication of the fitting heading angle is valid and the state identification indication of the speed heading angle is valid;

[0214] When a second execution condition is met, determining the extended fitting heading angle as the target heading angle, wherein the second execution condition is that the state identification of the fitting heading angle indicates that it is valid, and the state identification of the speed heading angle indicates that it is invalid;

[0215] When a third execution condition is met, the extended speed heading angle is determined as the target heading angle; wherein the third execution condition is that the state identification indication of the fitting heading angle is invalid and the state identification indication of the speed heading angle is valid;

[0216] When the fourth execution condition is met, the heading angle of the track in the previous preset frame is obtained, and the target heading angle is predicted based on the heading angle of the track in the previous preset frame, and the fourth execution condition is that the status mark indication of the fitting heading angle is invalid and the status mark indication of the speed heading angle is invalid.

[0217] Optionally, the determination module 83 is further used for:

[0218] In the case where the moving speed of the target is in a fluctuating state, if the state identification indication of the speed and heading angle is valid, determining the target heading angle based on the speed and heading angle;

[0219] In the case where the target is in a U-turn motion state, if the state identification indication of the fitting heading angle is valid, the target heading angle is determined based on the fitting heading angle.

[0220] It can be understood by those skilled in the art that Figure 8 The structure of the heading angle calculation device based on target tracking does not constitute a limitation on the heading angle calculation device based on target tracking, and the various modules can be implemented in whole or in part by software, hardware, and a combination thereof. The above modules can be embedded in or independent of the controller in the computer device in the form of hardware, or can be stored in the memory in the computer device in the form of software, so that the controller can call and execute the operations corresponding to the above modules. In other embodiments, the heading angle calculation device based on target tracking may include more or fewer modules than those shown in the figure.

[0221] See also Fig. 9 In another aspect of the embodiment of the present application, a vehicle device 10 is provided, including a memory 3011 and a processor 3012. The memory 3011 stores a computer program. When the computer program is executed by the processor, the processor 3012 executes the steps of the heading angle calculation method based on target tracking provided in any of the above embodiments of the present application. The vehicle device 10 is a device installed on any type of mobile body, such as a vehicle, an electric vehicle, a hybrid electric vehicle, a motorcycle, a bicycle, a personal mobile device, an airplane, a drone, a ship, or a robot, etc.

[0222] The processor 3012 is the control center, which uses various interfaces and lines to connect various parts of the entire device, and executes various functions of the device and processes data by running or executing software programs and / or modules stored in the memory 3011, and calling data stored in the memory 3011. Optionally, the processor 3012 may include one or more processing cores; preferably, the processor 3012 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user pages and application programs, etc., and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processor 3012.

[0223] The memory 3011 can be used to store software programs and modules. The processor 3012 executes various functional applications and data processing by running the software programs and modules stored in the memory 3011. The memory 3011 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data created according to the use of the device, etc. In addition, the memory 3011 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices. Accordingly, the memory 3011 may also include a memory controller to provide the processor 3012 with access to the memory 3011.

[0224] In some embodiments, see Fig.10 In another aspect of the embodiment of the present application, a radar device 12 is provided, including a storage unit 4011 and a processing unit 4012. The storage unit 4011 stores a computer program. When the computer program is executed by the processing unit, the processing unit 4012 executes the steps of the heading angle calculation method based on target tracking provided in any of the above embodiments of the present application. The radar device 12 is a 4D millimeter wave radar, a 3D millimeter wave radar, or other types of three-dimensional or four-dimensional radar devices.

[0225] The processing unit 4012 is the control center, which uses various interfaces and lines to connect various parts of the entire device, and executes various functions of the device and processes data by running or executing software programs and / or modules stored in the storage unit 4011, and calling data stored in the storage unit 4011. Optionally, the processing unit 4012 may include one or more processing cores; preferably, the processing unit 4012 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user pages and application programs, etc., and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processing unit 4012.

[0226] The storage unit 4011 can be used to store software programs and modules. The processing unit 4012 executes various functional applications and data processing by running the software programs and modules stored in the storage unit 4011. The storage unit 4011 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the device, etc. In addition, the storage unit 4011 can include a high-speed random access memory, and can also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices. Accordingly, the storage unit 4011 can also include a memory controller to provide the processing unit 4012 with access to the storage unit 4011.

[0227] On the other hand, an embodiment of the present application further provides a storage medium storing a computer program. When the computer program is executed by a processor, the processor executes the steps of the heading angle calculation method based on target tracking provided in any of the above embodiments of the present application.

[0228] Those skilled in the art can understand that all or part of the processes in the methods provided in the above embodiments can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in the present application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0229] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. The protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A heading angle calculation method based on target tracking, characterized in that: include: Get the reference heading angle of the target's track in the current frame; Acquire the associated point cloud data of the track in the current frame, and calculate the point cloud enclosing heading angle based on the associated point cloud data; Calculate the speed and heading angle of the track in the current frame; Calculating a fitting heading angle based on the associated point cloud data; Determine the target heading angle of the track based on the reference heading angle, the point cloud enclosing heading angle, the speed heading angle and the fitting heading angle; Based on the target heading angle, the heading angle of the track in the current frame is obtained.

2. The method for calculating a heading angle based on target tracking according to claim 1, characterized in that: Before obtaining the reference heading angle of the track of the target in the current frame, the method further includes: Acquire tracking data of the target in the current frame, the tracking data comprising at least one of the following: a tracking attribute indicating whether the target is a moving target, and a ground speed of the track; Based on the tracking data, determining whether to update a reference heading angle corresponding to the target; When the tracking attribute of the target indicates that the target is a moving target and / or the ground speed of the track is greater than a preset reference speed value, it is determined to update the reference heading angle corresponding to the target.

3. The method for calculating a heading angle based on target tracking according to claim 1, characterized in that: The step of obtaining the reference heading angle of the track of the target in the current frame includes: When a reference heading angle calculation condition is met, obtaining a current heading angle of the track in the current frame and a heading angle of the track in a previous preset frame of the current frame, and calculating a first reference heading angle based on the current heading angle and the heading angle of the track in the previous preset frame; Acquire the ground speed of the track in the current frame, and calculate a second reference heading angle according to the ground speed of the track; A reference heading angle difference between the first reference heading angle and the second reference heading angle is calculated, and based on the reference heading angle difference, the reference heading angle is determined from the first reference heading angle and the second reference heading angle.

4. The method for calculating a heading angle based on target tracking according to claim 3, characterized in that: The determining the reference heading angle from the first reference heading angle and the second reference heading angle based on the reference heading angle difference comprises: Wherein headingRef represents the reference heading angle, headingRef1 represents the first reference heading angle, headingRef2 represents the second reference heading angle, xVel represents the component of the ground velocity of the track in the X-axis direction, yVel represents the component of the ground velocity of the track in the Y-axis direction, wherein the Y-axis direction is perpendicular to the X-axis direction and is located in the same plane, the X-axis direction is along the center line of the vehicle equipment and points to the forward direction of the vehicle equipment, headingRef diff Represents the reference heading angle difference, and angleThre1 represents the preset reference angle difference threshold.

5. The method for calculating heading angle based on target tracking according to claim 1, characterized in that: The step of calculating the point cloud enclosing heading angle based on the associated point cloud data comprises: When the point cloud enclosing heading angle calculation conditions are met, the current feature angle is obtained within the preset angle range and with the preset angle step size; Rotating the position coordinates of the associated point cloud in the associated point cloud data according to the current feature angle to obtain the position coordinate limit of the rotated point cloud bounding box; Calculating size data of the rotated point cloud bounding box based on the position coordinate limit of the rotated point cloud bounding box; Calculating the fitting loss error corresponding to the current feature angle based on the position coordinates of the associated point cloud and the size data; When the fitting loss error corresponding to the current feature angle is less than the current fitting loss error, the fitting loss error corresponding to the current feature angle is updated to the current fitting loss error, and the current feature angle is continuously acquired in a loop until the minimum fitting loss error is acquired, and the point cloud enclosing heading angle is determined based on the feature angle corresponding to the minimum fitting loss error.

6. The method for calculating heading angle based on target tracking according to claim 5, characterized in that: The determining the point cloud enclosing heading angle based on the characteristic angle corresponding to the minimum fitting loss error comprises: Under the condition that the first enclosing heading angle calculation condition is met, determining the point cloud enclosing heading angle to be the characteristic angle corresponding to the minimum fitting loss error minus 90°, wherein the first enclosing heading angle calculation condition includes that the length of the rotated point cloud enclosing frame calculated based on the characteristic angle corresponding to the minimum fitting loss error is less than the width; Under the condition that the second enclosing heading angle calculation condition is met, determining the point cloud enclosing heading angle as the characteristic angle corresponding to the minimum fitting loss error plus 180°, wherein the second enclosing heading angle calculation condition includes: when the angle formed by the characteristic angle corresponding to the minimum fitting loss error and the radial velocity of the associated point cloud is greater than 90°; When the first enclosing heading angle calculation condition or the second enclosing heading angle calculation condition is not met, the point cloud enclosing heading angle is determined to be a characteristic angle corresponding to the minimum fitting loss error.

7. The method for calculating heading angle based on target tracking according to claim 1, characterized in that: The calculating the speed and heading angle of the track in the current frame comprises: When the speed and heading angle calculation conditions are met, obtaining the ground speed of the track in the previous preset frame of the current frame, and obtaining the ground speed of the track in the current frame; Calculating a weighted ground speed of the track in the current frame based on the ground speed of the track in the previous preset frame and the ground speed of the track in the current frame; The speed and heading angle of the track in the current frame are calculated based on the weighted ground speed of the track in the current frame.

8. The method for calculating heading angle based on target tracking according to claim 7, characterized in that: The step of calculating the speed and heading angle of the track in the current frame based on the weighted ground speed of the track in the current frame includes: Based on the weighted ground speed of the track in the current frame, the speed and heading angle of the track in the current frame is calculated using the speed and heading angle calculation formula. The speed and heading angle calculation formula of the track is as follows: Wherein speedHeading represents the speed heading angle of the track, yVelFilter′ represents the weighted ground speed component of the track in the Y-axis direction, xVelFilter′ represents the weighted ground speed component of the track in the X-axis direction, xVelFilter represents the ground speed component of the track in the previous preset frame in the X-axis direction, yVelFilter represents the ground speed component of the track in the previous preset frame in the Y-axis direction, xVelAbs represents the ground speed component of the track in the current frame in the X-axis direction, yVelAbs represents the ground speed component of the track in the current frame in the Y-axis direction, α and β represent coefficients, the Y-axis direction is perpendicular to the X-axis direction and is located in the same plane, and the X-axis direction is along the center line of the vehicle equipment and points to the forward direction of the vehicle equipment.

9. The method for calculating heading angle based on target tracking according to claim 1, characterized in that: The calculating the fitting heading angle based on the associated point cloud data comprises: When the fitting heading angle calculation condition is met, the azimuths of all associated point clouds are acquired based on the associated point cloud data, and a first matrix is ​​generated based on the azimuths of all associated point clouds; Based on the associated point cloud data, obtaining radial velocities of all associated point clouds relative to the ground, and generating a second matrix based on the radial velocities of all associated point clouds relative to the ground; Calculating a fitting speed of the track based on the first matrix and the second matrix; When it is determined that the fitting speed of the track meets a preset fitting speed condition, the fitting heading angle is calculated based on the fitting speed of the track.

10. The method for calculating heading angle based on target tracking according to claim 9, characterized in that: The first matrix is ​​represented by A, and the first matrix is ​​as follows: in Indicates the azimuth of the nth associated point cloud; The second matrix is ​​denoted by Y, and the second matrix is ​​as follows: Where VradVelAbs n represents the radial velocity of the nth associated point cloud over the ground; The fitting speed of the track is represented by V, and the fitting speed of the track is as follows: Wherein Vx represents the component of the fitting speed of the track in the X-axis direction, Vy represents the component of the fitting speed of the track in the Y-axis direction, and the X-axis direction is along the center line of the vehicle equipment and points to the forward direction of the vehicle equipment.

11. The method for calculating heading angle based on target tracking according to claim 1, characterized in that: The determining the target heading angle of the track based on the reference heading angle, the point cloud enclosed heading angle, the speed heading angle and the fitted heading angle comprises: Based on the reference heading angle, the point cloud enclosing heading angle is compensated to obtain an extended enclosing heading angle, the speed heading angle is compensated to obtain an extended speed heading angle, and the fitting heading angle is compensated to obtain an extended fitting heading angle; Calculating an absolute difference between the extended enclosing heading angle and the reference heading angle to obtain a first absolute difference, calculating an absolute difference between the extended speed heading angle and the reference heading angle to obtain a second absolute difference, and calculating an absolute difference between the extended fitting heading angle and the reference heading angle to obtain a third absolute difference; In a case where the state indicator indicating that the point cloud encloses the heading angle is valid, determining the target heading angle by using a first screening strategy based on the first absolute difference, the second absolute difference, and the third absolute difference; When the status indicator of the point cloud enclosed heading angle indicates invalid, the target heading angle is determined using a second screening strategy based on the first absolute difference, the second absolute difference and the third absolute difference.

12. The method for calculating heading angle based on target tracking according to claim 11, characterized in that: The determining of the target heading angle by using a first screening strategy based on the first absolute difference, the second absolute difference and the third absolute difference when the state indicator of the point cloud enclosing heading angle indicates that the target heading angle is valid includes: When the state indicator of the point cloud enclosing heading angle indicates that it is valid, it includes the following: When a first preset condition is met, if the first absolute difference is less than the second absolute difference, the extended enveloping heading angle is determined as the target heading angle; if the first absolute difference is greater than or equal to the second absolute difference, the extended speed heading angle is determined as the target heading angle, wherein the first preset condition is that the state identification indication of the speed heading angle is valid; Under the condition that the second preset condition is met, if the first absolute difference is less than the third absolute difference, the extended enclosed heading angle is determined as the target heading angle, and if the first absolute difference is greater than or equal to the third absolute difference, the extended fitted heading angle is determined as the target heading angle, wherein the second preset condition is that the state identification indication of the speed heading angle is invalid, and the state identification indication of the fitted heading angle is valid; Under the third preset condition, if the first absolute difference is greater than the preset angle difference, the reference heading angle is determined as the target heading angle; if the first absolute difference is less than or equal to the preset angle difference, the extended enveloping heading angle is determined as the target heading angle, and the third preset condition is that the track is in a confirmed state and the track is in a non-stationary state; When the first preset condition, the second preset condition or the third preset condition is not satisfied, the extended enclosing heading angle is determined as the target heading angle.

13. The method for calculating heading angle based on target tracking according to claim 11, characterized in that: When the state mark indicating that the point cloud enclosed heading angle is invalid, determining the target heading angle by using a second screening strategy based on the first absolute difference, the second absolute difference and the third absolute difference, comprises: When the state indicator of the point cloud enclosing heading angle indicates invalid, it includes the following: When the first execution condition is met, if the third absolute difference is less than the second absolute difference, the extended fitting heading angle is determined as the target heading angle; if the third absolute difference is greater than or equal to the second absolute difference, the extended speed heading angle is determined as the target heading angle, wherein the first execution condition is that the state identification indication of the fitting heading angle is valid and the state identification indication of the speed heading angle is valid; When a second execution condition is met, determining the extended fitting heading angle as the target heading angle, wherein the second execution condition is that the state identification of the fitting heading angle indicates that it is valid, and the state identification of the speed heading angle indicates that it is invalid; When a third execution condition is met, the extended speed heading angle is determined as the target heading angle; wherein the third execution condition is that the state identification indication of the fitting heading angle is invalid and the state identification indication of the speed heading angle is valid; When the fourth execution condition is met, the heading angle of the track in the previous preset frame is obtained, and the target heading angle is predicted based on the heading angle of the track in the previous preset frame, and the fourth execution condition is that the status mark indication of the fitting heading angle is invalid and the status mark indication of the speed heading angle is invalid.

14. The method for calculating heading angle based on target tracking according to claim 1, characterized in that: The determining of the target heading angle of the track based on the reference heading angle, the point cloud enclosed heading angle, the speed heading angle, and the fitted heading angle comprises at least one of the following: In the case where the moving speed of the target is in a fluctuating state, if the state identification indication of the speed and heading angle is valid, determining the target heading angle based on the speed and heading angle; In the case where the target is in a U-turn motion state, if the state identification indication of the fitting heading angle is valid, the target heading angle is determined based on the fitting heading angle.

15. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method for calculating a heading angle based on target tracking as claimed in any one of claims 1 to 14 is implemented.

16. A radar device, characterized in that: It comprises a storage unit and a processing unit, wherein the storage unit stores a computer program, and when the computer program is executed by the processing unit, the processing unit executes the heading angle calculation method based on target tracking as claimed in any one of claims 1 to 14.

17. A vehicle device, characterized in that: The radar device comprises the radar device as claimed in claim 16, or comprises a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the heading angle calculation method based on target tracking as claimed in any one of claims 1 to 14.

Images