Target detection method, domain controller and computer program product

Abstract

The invention provides a target detection method, a domain controller and a computer program product, and according to the target detection method provided by the invention, after point cloud data is collected by radar equipment carried by a vehicle for a surrounding environment, the point cloud data is analyzed and processed, and the target detection result is obtained. According to the embodiment of the invention, the method can determine the moving target around the vehicle and the road guardrail, and can determine that the moving target belongs to a mistakenly detected target if there is an overlapping region between the road guardrail and the moving target, thereby facilitating the improvement of the accuracy of a target detection scheme implemented based on radar equipment, and reducing the false recognition rate of the moving target.

Description

Technical Field

[0001] The present invention relates to the field of radar target detection technology, and in particular to a target detection method, a domain controller and a computer program product. Background Art

[0002] With the continuous development of science and technology, radar equipment has superior detection performance and performs well in environments such as night, rainy days, and foggy days. It can better identify surrounding static and dynamic targets, so people gradually begin to install radar equipment on vehicles to perceive the surrounding environment of the vehicle. However, when a vehicle equipped with radar equipment passes through a road guardrail, if the distance between the vehicle and the road guardrail remains unchanged, the radial velocity of the point cloud at the road guardrail measured by the radar equipment is usually almost zero, and because the radar equipment is usually unable to accurately measure the tangential velocity of the point cloud at this part of the road guardrail, once the noise factor causes the radial velocity of the point cloud at the road guardrail to meet the creation conditions of the moving target, it is easy to mistakenly create a moving target based on the measured point cloud data at the road guardrail, thereby affecting the accuracy of the target detection result. Summary of the invention

[0003] Based on this, the present invention provides a target detection method, a domain controller and a computer program product. By adopting the target detection method, it is possible to accurately determine whether the moving targets around the vehicle identified based on point cloud data are misdetected targets, which is beneficial to improving the accuracy of the target detection scheme implemented based on radar equipment and reducing the misrecognition rate of mobile targets.

[0004] In one aspect, the present invention provides a target detection method, the method comprising:

[0005] Obtain point cloud data collected by the radar equipment on the vehicle for the surrounding environment;

[0006] Performing target detection processing based on the point cloud data to obtain moving targets around the vehicle;

[0007] Based on the point cloud data, determining whether there is a road guardrail within a preset range around the vehicle to obtain a first determination result;

[0008] If the first judgment result indicates that there is a road guardrail within the preset range around the vehicle, judging whether there is an overlapping area between the road guardrail and the moving target according to the first position information of the road guardrail and the second position information of the moving target, and obtaining a second judgment result;

[0009] If the second judgment result indicates that there is an overlapping area between the road guardrail and the moving object, it is determined that the moving object is a misdetected object.

[0010] Furthermore, in some embodiments, the above method further comprises:

[0011] Based on the point cloud data, from various preset ranges around the vehicle, the preset range containing the largest number of stationary point clouds is determined to obtain a target preset range where the road guardrail exists; wherein the stationary point cloud is a point cloud whose absolute speed is less than or equal to a preset speed;

[0012] The average value of the distance values ​​between each of the stationary point clouds within the preset target range and the central axis of the vehicle is calculated to obtain the first position information reflecting the target distance value between the road guardrail and the central axis of the vehicle.

[0013] Furthermore, in some embodiments, the above method further comprises:

[0014] Acquire the area information of the mobile target obtained by performing target detection processing based on the point cloud data to obtain the second position information of the mobile target; or,

[0015] Acquiring information about the area where the mobile target is located, obtained by performing target detection processing based on the point cloud data;

[0016] Based on the area information of the moving target, the target distance range between the virtual front bumper and the virtual rear bumper of the moving target and the central axis of the vehicle is determined to obtain the second position information of the moving target.

[0017] Furthermore, in some implementations, before determining that the moving target is a misdetected target, the method further includes:

[0018] Determine whether the duration of the overlapping area between the road guardrail and the moving target reaches a first threshold, and obtain a third determination result;

[0019] Determining that the moving target is a misdetected target specifically includes:

[0020] If the third judgment result indicates that the duration of the overlapping area between the road guardrail and the moving target reaches a first threshold, the moving target is changed from a valid moving target state to a falsely detected target state.

[0021] Furthermore, in some embodiments, after changing the mobile target from the valid mobile target state to the misdetected target state, the method further includes:

[0022] Determine whether a first preset condition for changing the state of the moving target is satisfied, and obtain a fourth determination result; wherein the first preset condition includes: the moving target and the road guardrail have no overlapping area within a recent first preset time period, the relative speed between the moving target and the vehicle reaches a second threshold, and the absolute speed of the moving target reaches a third threshold;

[0023] If the fourth determination result indicates that the first preset condition is satisfied, the moving object is changed from the misdetected object state to the valid moving object state.

[0024] Furthermore, in some embodiments, if the first judgment result indicates that there is a road guardrail within a preset range around the vehicle, the method further includes:

[0025] Determine whether a second preset condition for changing the state of the road guardrail is met to obtain a sixth judgment result; wherein the second preset condition includes: the relative radial speed between the road guardrail and the vehicle is greater than a sixth threshold, the stationary point cloud contained in the preset range where the road guardrail is located has not changed within the latest second preset time length, there is an overlapping area between the road guardrail and the credible non-guardrail target, the vehicle is in a low-speed driving state, and the vehicle is in a turning driving state.

[0026] If the sixth judgment result indicates that the second preset condition for changing the state of the road guardrail is satisfied, the road guardrail is changed from a valid road guardrail state to an invalid road guardrail state;

[0027] If the sixth judgment result indicates that the second preset condition for changing the state of the road guardrail is not satisfied, determining that the road guardrail is in a valid road guardrail state;

[0028] The determining whether there is an overlapping area between the road guardrail and the moving target includes:

[0029] If the road guardrail is in a valid road guardrail state, it is determined whether there is an overlapping area between the road guardrail and the moving target.

[0030] Furthermore, in some embodiments, the above method further comprises:

[0031] Determine whether a third preset condition for changing the state of the road guardrail is satisfied, and obtain a seventh judgment result; wherein the third preset condition includes: the time interval between the current moment and the moment when the road guardrail is changed to the invalid road guardrail state reaches a third preset time length, the stationary point cloud included in the preset range where the road guardrail is located changes within the latest fourth preset time length, and the host vehicle is in a high-speed straight driving state;

[0032] If the seventh judgment result indicates that the third preset condition for changing the state of the road guardrail is met, the road guardrail is changed from the invalid road guardrail state to the valid road guardrail state.

[0033] On the other hand, the present invention also provides a blind spot alarm control method, comprising:

[0034] Obtain point cloud data collected by the radar equipment on the vehicle for the surrounding environment;

[0035] Performing target detection processing based on the point cloud data to obtain moving targets around the vehicle;

[0036] Based on the point cloud data, determining whether there is a road guardrail within a preset range around the vehicle to obtain a first determination result; the preset range includes a lateral visual blind spot of the vehicle;

[0037] If the first judgment result indicates that there is a road guardrail within the preset range around the vehicle, judging whether there is an overlapping area between the road guardrail and the moving target according to the first position information of the road guardrail and the second position information of the moving target, and obtaining a second judgment result;

[0038] If the second judgment result indicates that there is an overlapping area between the road guardrail and the moving target, control information for prohibiting the lateral blind spot monitoring function of the vehicle from issuing an early warning for the moving target is generated.

[0039] Further, in some embodiments, the preset range is: the distance from the central axis of the vehicle is greater than or equal to 0.5 times the width of the vehicle and less than or equal to the seventh threshold, and the distance from the center point of the vehicle is less than or equal to the eighth threshold.

[0040] On the other hand, the present invention also provides a domain controller, comprising: a processor and a memory; wherein the memory stores computer-readable instructions, and the computer-readable instructions are suitable for being loaded by the processor and executing the steps of the above-mentioned target detection method and blind spot alarm control method.

[0041] On the other hand, the present invention also provides a computer program product having at least one instruction stored thereon, which, when executed by a controller, implements the steps of the above-mentioned target detection method and blind spot alarm control method.

[0042] According to the target detection method provided by the present invention, after collecting point cloud data for the surrounding environment using the radar equipment carried by the vehicle, the point cloud data can be analyzed and processed to determine the mobile targets and road guardrails around the vehicle. If there is an overlapping area between the road guardrail and the mobile target, it can be determined that the mobile target is a misdetected target, which is beneficial to improving the accuracy of the target detection scheme implemented based on the radar equipment and reducing the misrecognition rate of mobile targets. In addition, the target detection method has low computational complexity and high operating efficiency, and is also beneficial to ensuring the real-time recognition of misdetected targets. In addition, the target detection method has low performance requirements for hardware resources such as processors and memories when implemented, and has good practicality.

[0043] It should be understood that the contents described in the summary of the invention are not intended to limit the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 A schematic diagram of a target detection scenario provided by an embodiment of the present invention;

[0045] Figure 2 A schematic diagram of a flow chart of a target detection method provided by an embodiment of the present invention;

[0046] Figure 3 A schematic diagram of a road guardrail within a preset range around a vehicle provided by an embodiment of the present invention;

[0047] Figure 4 A schematic diagram of a moving target existing around a vehicle provided by an embodiment of the present invention;

[0048] Figure 5 A flowchart of a blind spot alarm control method provided by an embodiment of the present invention;

[0049] Figure 6 A schematic diagram of the structure of a domain controller provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the specific embodiments of the present invention and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0051] In the description of one or more embodiments of the present invention, the term "including" and similar terms should be understood as open inclusion, that is, "including but not limited to". The term "based on" should be understood as "based at least in part on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The terms "first", "second", etc. may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0052] At present, when a vehicle equipped with radar equipment performs radar target detection while driving, it often has the problem of mistakenly identifying road guardrails as moving targets, thus affecting the accuracy of radar target detection results. Figure 1 The target detection scenario shown in FIG. 1 is used to explain the problem of false detection of moving targets.

[0053] like Figure 1 As shown, it is assumed that the vehicle 101 is traveling from left to right, and the rear side of the vehicle 101 is equipped with a radar device 102, and one side of the vehicle 101 has a road guardrail 103. When the user drives the vehicle 101 past the road guardrail 103, if the distance between the vehicle 101 and the road guardrail 103 remains relatively constant, the radial velocity of the point cloud measured by the radar device 102 for the designated position 107 at the road guardrail 103 (i.e., the position where the line connecting the radar device 102 is perpendicular to the central axis of the vehicle 101) is usually almost zero, and since the radar device 102 is usually unable to accurately measure the tangential velocity of the point cloud at the designated position 107 at the road guardrail 103, once the radial velocity of the point cloud at the designated position 107 at the road guardrail 103 meets the creation condition of the moving target due to noise factors, the designated position 107 at the road guardrail 103 may be mistakenly identified as the moving target 104.

[0054] Subsequently, as the vehicle 101 continues to travel, the radial velocity of the point cloud measured by the radar device 102 at the vehicle 101 for the target position 108 at the road guardrail 103 (i.e., the position where the line connecting the radar device 102 is perpendicular to the central axis of the vehicle 101) is usually almost zero, and the radar device 102 is usually unable to accurately measure the tangential velocity of the point cloud at the target position 108 at the road guardrail 103. At this time, since the radial velocity of the point cloud at the target position 108 is relatively close to the radial velocity of the moving target 104 created based on the point cloud at the designated position 107, it is easy to successfully associate the point cloud at the target position 108 with the moving target 104 during radar target detection, and then continue to mistakenly create the moving target 105 belonging to the same target object as the moving target 104 based on the point cloud at the target position 108 of the road guardrail 103.

[0055] Based on this, how to accurately determine whether the moving target obtained by radar target detection is a false detection target in order to ensure the accuracy of the target detection scheme implemented based on radar equipment has become a technical problem that needs to be solved urgently.

[0056] In order to solve the above technical problems, Figure 2 A flow chart of a target detection method provided by an embodiment of the present invention. From a program perspective, the execution subject of the process can be a program for radar target detection mounted on a vehicle control device (e.g., Microcontroller Unit, MCU; or, domain controller) or a vehicle. Alternatively, the execution subject of the process can also be a vehicle control device or a vehicle, or other devices that can communicate with a vehicle control device or a vehicle, without specific limitation.

[0057] The following is for Figure 2 The process shown is described in detail, and the target detection method may specifically include the following steps:

[0058] Step S202, obtaining point cloud data collected by the radar equipment mounted on the vehicle for the surrounding environment.

[0059] In the embodiment of the present invention, the vehicle may be equipped with a radar device to sense the surrounding environment. The radar device may be of various types, such as a millimeter wave radar, a laser radar, etc., which is not specifically limited.

[0060] Step S204: performing target detection processing based on the point cloud data to obtain moving targets around the vehicle.

[0061] In an embodiment of the present invention, by analyzing the point cloud data collected by the radar device for the surrounding environment and performing target detection processing, the target objects existing around the vehicle can be obtained. According to the relative movement between the target object and the ground, the target object may include: a moving target and a stationary target. Among them, the stationary target may refer to a target object that is stationary relative to the ground, for example, it may include but is not limited to: fixed road facilities, buildings, stationary traffic participants, etc. The moving target may refer to a target object that is in a moving state relative to the ground, for example, a traffic participant who is driving / walking, etc., and there is no specific limitation on this.

[0062] Step S206: Based on the point cloud data, determine whether there is a road guardrail within a preset range around the vehicle to obtain a first determination result.

[0063] In the embodiment of the present invention, the road guardrail may refer to a traffic safety facility set up on the road to divide different driving areas. The road guardrail may be of various types, such as metal fences, bridge railings, concrete columns, tunnel walls, etc., which are not specifically limited.

[0064] Since there may be a problem of mistakenly identifying a road guardrail as a mobile target when performing radar target detection, it is necessary to identify whether there is a road guardrail within the preset range around the vehicle during the radar target detection process. If there is a road guardrail within the preset range around the vehicle, it is necessary to further determine whether the mobile target obtained by radar target detection is a misdetected target to improve the accuracy of the final mobile target detection result. If there is no road guardrail within the preset range around the vehicle, the mobile target detected in step S204 can be directly used as a valid mobile target, which has good flexibility.

[0065] Step S208, if the first judgment result indicates that there is a road guardrail within a preset range around the vehicle, then based on the first position information of the road guardrail and the second position information of the moving target, it is determined whether there is an overlapping area between the road guardrail and the moving target to obtain a second judgment result.

[0066] In the embodiment of the present invention, the point cloud data collected by the radar device mounted on the vehicle for the surrounding environment may include: point cloud position data. Therefore, when the point cloud data is analyzed and processed to determine that there are road guardrails and mobile targets around the vehicle, the first position information of the road guardrail and the second position information of the mobile target can be analyzed and obtained. Therefore, it is possible to conveniently determine whether there is an overlapping area between the road guardrail and the mobile target based on the first position information of the road guardrail and the second position information of the mobile target, so as to decide whether the mobile target is a misdetected target. The computational complexity is low, which is conducive to saving computational resources and improving the detection efficiency of misdetected targets.

[0067] Step S210: If the second judgment result indicates that there is an overlapping area between the road guardrail and the moving target, it is determined that the moving target is a misdetected target.

[0068] In the embodiment of the present invention, when there is an overlapping area between the moving target and the road guardrail, it can usually be said that the road guardrail penetrates the moving target. Obviously, this is inconsistent with the actual situation. At this time, it can be considered that the moving target is a false detection target generated by the road guardrail and noise factors, that is, the road guardrail should be mistakenly regarded as the moving target. If there is no overlapping area between the moving target and the road guardrail, it can usually be said that the road guardrail does not penetrate the moving target. Obviously, this is consistent with the actual situation. Therefore, it can be considered that the moving target is not a false detection target, but a valid moving target.

[0069] Figure 2 The method in the invention can collect point cloud data for the surrounding environment by using the radar equipment carried by the vehicle, and then analyze and process the point cloud data to determine the mobile targets and road guardrails around the vehicle. If there is an overlapping area between the road guardrail and the mobile target, it can be determined that the mobile target is a misdetected target, which is beneficial to improving the accuracy of the target detection scheme based on the radar equipment and reducing the misrecognition rate of the mobile target. In addition, the target detection method does not involve calculations with a complexity of O(n) or above, so the calculation complexity is low and the operation efficiency is high, which is also beneficial to ensuring the real-time recognition of the misdetected targets and reducing the performance requirements of the scheme for hardware resources such as processors and memories, and has good practicality.

[0070] In a feasible implementation, Figure 2 The method may further include:

[0071] Based on the point cloud data, the preset range containing the largest number of stationary point clouds is determined from various preset ranges around the vehicle, and the target preset range where the road guardrail exists is obtained; wherein the stationary point cloud is a point cloud whose absolute speed is less than or equal to the preset speed.

[0072] The average value of the distance values ​​between each of the stationary point clouds within the preset target range and the central axis of the vehicle is calculated to obtain the first position information reflecting the target distance value between the road guardrail and the central axis of the vehicle.

[0073] In the embodiment of the present invention, since the road guardrail is a stationary target, the absolute speed of the point cloud constituting the road guardrail should generally be small. Furthermore, since the driving path of the vehicle is generally parallel to the road guardrail, the surrounding area of ​​the vehicle can be divided into several preset ranges parallel to the vehicle, and the preset range containing the largest number of stationary point clouds can be used as the target preset range where the road guardrail exists.

[0074] In the embodiment of the present invention, since the width of metal fences and bridge railings is usually small, and radar equipment is usually unable to penetrate tunnel walls and detect objects on the other side of the tunnel wall, when determining whether there is an overlapping area between the road guardrail and the moving target, the thickness of the road guardrail can be ignored, that is, the road guardrail can be regarded as a line segment roughly parallel to the central axis of the vehicle. Based on this, the target distance value between the road guardrail and the central axis of the vehicle can be determined as the first position information of the road guardrail.

[0075] In practical applications, the average value of the distance values ​​between each stationary point cloud within the target preset range and the central axis of the vehicle can be used as the target distance value between the road guardrail and the central axis of the vehicle; the median of the distance values ​​between each stationary point cloud within the target preset range and the central axis of the vehicle can be used as the target distance value; or, clustering processing can be performed on the position data of each stationary point cloud within the target preset range, and the distance value between the cluster center point and the central axis of the vehicle can be used as the target distance value. There is no specific limitation on this.

[0076] In addition, since the point cloud data collected by the radar device may be sparse, it is impossible to accurately determine the starting and ending positions of the road guardrail based on each static point cloud within the preset target range. Based on this, the starting and ending positions of the road guardrails within the preset target range can be pre-set according to actual needs and experience as the first position information of the road guardrail.

[0077] Among them, the distance between the starting position and the central axis of the vehicle can be the above-mentioned target distance value, and the vertical point of the starting position on the central axis of the vehicle can be a point located behind the vehicle and at a first distance from the center point of the vehicle; similarly, the distance between the ending position and the central axis of the vehicle can be the above-mentioned target distance value, and the vertical point of the ending position on the central axis of the vehicle can be a point located in front of the vehicle and at a second distance from the center point of the vehicle. In actual applications, the first distance and the second distance can be set according to actual needs. For example, the first distance can be several meters or tens of meters, and the second distance can be several meters, and there is no specific limitation on this.

[0078] For ease of understanding, Figure 3 A schematic diagram of a road guardrail within a preset range around a vehicle provided by an embodiment of the present invention, such as Figure 3 As shown, assuming that a radar device 301 is deployed at the left rear corner of the vehicle, and the central axis of the vehicle is a dotted line 302, the surrounding area of ​​the vehicle can be divided into several preset ranges parallel to the vehicle (for example, ranges 303-306). Assuming that the number of stationary point clouds contained in ranges 303-306 is 1, 5, 2, 0, and 1, respectively, since the number of stationary point clouds contained in range 304 is the largest, it can be considered that range 304 is the target preset range where a road guardrail exists.

[0079] If you will Figure 3 The average value of the distance between the stationary point cloud existing in the target preset range (i.e., range 304) and the central axis 302 of the vehicle is used as the target distance value between the road guardrail and the central axis of the vehicle; and the starting position and the ending position of the road guardrail existing in the preset target preset range are obtained, then the road guardrail existing around the vehicle can be determined as line segment 307, and at this time, the position information of line segment 307 can be the first position information of the road guardrail existing around the vehicle.

[0080] Optionally, in a feasible implementation mode, Figure 2 The method may further include:

[0081] Acquire the area information of the mobile target obtained by performing target detection processing based on the point cloud data to obtain the second position information of the mobile target. Or,

[0082] The area information of the moving target obtained by performing target detection processing based on the point cloud data is obtained.

[0083] Based on the area information of the moving target, the target distance range between the virtual front bumper and the virtual rear bumper of the moving target and the central axis of the vehicle is determined to obtain the second position information of the moving target.

[0084] In the embodiment of the present invention, the point cloud data collected by the radar device mounted on the vehicle for the surrounding environment may include: point cloud location data, so after the point cloud data is subjected to target detection processing, the area information of the mobile target around the vehicle can usually be obtained. At this time, the area information of the mobile target can be directly used as the second location information of the mobile target, which is convenient and quick.

[0085] Alternatively, since the moving target may generally include a vehicle object in a moving state, the front end and the rear end of the moving target may generally be a front bumper and a rear bumper. In combination with the above embodiments, it can be seen that the first position information of the road guardrail may be a target distance value between the road guardrail and the center axis of the vehicle. Therefore, in order to facilitate the measurement of whether there is overlap between the moving target and the road guardrail, the moving target may be simplified into a virtual front bumper and a virtual rear bumper perpendicular to the center axis of the vehicle. At this time, the closest distance and the farthest distance between the virtual front bumper and the virtual rear bumper of the moving target and the center axis of the vehicle may be used as the starting value and the ending value of the target distance range, so that the target distance range may be used as the second position information of the moving target.

[0086] In practical applications, for the short rear side of the area where the moving target is located, any endpoint of the short rear side can be used as a reference point, and the short rear side is projected in a direction perpendicular to the central axis of the vehicle through the reference point to obtain the virtual rear bumper of the moving target. Similarly, for the short front side of the area where the moving target is located, any endpoint of the short front side can be used as a reference point, and the short front side is projected in a direction perpendicular to the central axis of the vehicle through the reference point to obtain the virtual front bumper of the moving target. Of course, other methods can also be used to simplify the model of the moving target to obtain the position information of the virtual front bumper and virtual rear bumper of the moving target, and then determine the target distance range between the virtual front bumper and virtual rear bumper of the moving target and the central axis of the vehicle, and no specific limitation is made to this.

[0087] In practical applications, when judging whether there is an overlapping area between the road guardrail and the moving target, it is possible to judge whether the area where the moving target is located contains any position point of the road guardrail, or to directly judge whether the target distance value between the road guardrail and the central axis of the vehicle is within the target distance range between the virtual front bumper and the virtual rear bumper of the moving target and the central axis of the vehicle. The calculation complexity is low and the flexibility is good.

[0088] In addition, the third distance between the vertical point of the virtual front bumper of the moving target on the central axis of the vehicle and the center point of the vehicle, and the fourth distance between the vertical point of the virtual rear bumper of the moving target on the central axis of the vehicle and the center point of the vehicle can also be determined as the second position information of the moving target. At this time, it is also possible to further determine whether there is an overlapping area between the road guardrail and the moving target based on the third distance and fourth distance between the vertical point of the virtual front bumper and the virtual rear bumper of the moving target on the central axis of the vehicle and the center point of the vehicle, and the first distance and second distance between the vertical point of the starting position and the ending position of the road guardrail on the central axis of the vehicle and the center point of the vehicle, which is conducive to ensuring the recognition accuracy of the overlapping situation between the road guardrail and the moving target.

[0089] For ease of understanding, Figure 4 A schematic diagram of a mobile target existing around a vehicle provided by an embodiment of the present invention, such as Figure 4 As shown, there may be a road guardrail 402 around the vehicle 401, and the area where the mobile target exists around the vehicle 401 may be area 403. At this time, the position information of area 403 may be directly used as the second position information of the mobile target. Alternatively, the model of the mobile target may be simplified according to the position information of area 403 to obtain the target distance range between the virtual front bumper 404 and the virtual rear bumper 405 of the mobile target and the central axis of the vehicle, thereby serving as the second position information of the mobile target, which will not be described in detail.

[0090] In a feasible implementation manner, before determining that the moving target is a misdetected target, the following steps may also be included:

[0091] It is determined whether the duration of the overlapping area between the road guardrail and the moving target reaches a first threshold, and a third determination result is obtained.

[0092] Correspondingly, determining that the moving target is a misdetected target may specifically include:

[0093] If the third judgment result indicates that the duration of the overlapping area between the road guardrail and the moving target reaches a first threshold, the moving target is changed from a valid moving target state to a falsely detected target state.

[0094] In the embodiment of the present invention, since the point cloud data measured by the radar device may have deviations, or the first position information of the road guardrail determined based on the point cloud data and the second position information of the mobile target may also have deviations, the credibility of the result that the road guardrail and the mobile target are detected at a single moment and there is an overlapping area may not be high enough. Based on this, after determining that the duration of the overlapping area between the road guardrail and the mobile target reaches the first threshold, it can be determined that the mobile target is a false detection target, so that the mobile target can be changed from a valid mobile target state to a false detection target state. If the duration of the overlapping area between the road guardrail and the mobile target does not reach the first threshold, it can be prohibited to change the mobile target from a valid mobile target state to a false detection target state. This is conducive to ensuring the accuracy of the detection result for the mobile target. Among them, the valid mobile target state can indicate that the credibility of the mobile target is good; and the false detection target state can indicate that the mobile target is likely to be a mobile target generated incorrectly based on the road guardrail, which will not be described in detail.

[0095] It is understandable that radar equipment usually periodically collects point cloud data for the surrounding environment of the vehicle, so that the point cloud data collected in each cycle can be analyzed and processed to determine whether there are road guardrails and moving targets around the vehicle in each cycle; and through target tracking processing, it can be identified whether the road guardrails detected in each cycle are the same road guardrails, and whether the moving targets detected in each cycle are the same moving targets; thereby accurately identifying whether the duration of the overlapping area between the road guardrail and the moving target reaches the first threshold, which will not be elaborated. Among them, the first threshold can be set according to actual needs, for example, it can be hundreds of milliseconds or seconds, and there is no specific limitation on this.

[0096] In a feasible implementation manner, after changing the mobile target from a valid mobile target state to a misdetected target state, the method may further include:

[0097] Determine whether a first preset condition for changing the state of the moving target is met to obtain a fourth judgment result; wherein the first preset condition includes: the moving target and the road guardrail have no overlapping area within the most recent first preset time period, the relative speed between the moving target and the vehicle reaches a second threshold, and the absolute speed of the moving target reaches at least one of a third threshold.

[0098] If the fourth determination result indicates that the first preset condition is satisfied, the moving object is changed from the misdetected object state to the valid moving object state.

[0099] In an embodiment of the present invention, in order to reduce the adverse effects caused by possible errors in the operation of changing the mobile target from the valid mobile target state to the false detection target state, it is also possible to continue to determine whether the first preset condition for changing the mobile target from the false detection target state to the valid mobile target state is currently met; if so, the mobile target can be changed to the valid mobile target state to ensure the accuracy of the mobile target detection result.

[0100] The first preset condition can be set according to actual needs. For example, when the moving target and the road guardrail do not overlap for a long time, the moving target may not be a case where the road guardrail is mistakenly identified as a moving target; or, when the road guardrail is mistakenly identified as a moving target, the relative speed between the moving target and the vehicle and the absolute speed of the moving target are usually small. Therefore, when the relative speed between the moving target and the vehicle is large, or the absolute speed of the moving target is large, the moving target may not be a case where the road guardrail is mistakenly identified as a moving target. Based on this, the first preset condition can include: the moving target and the road guardrail have no overlapping area within the last first preset time length, the relative speed between the moving target and the vehicle reaches the second threshold, and the absolute speed of the moving target reaches at least one of the third threshold. The first preset time length can be hundreds of milliseconds or seconds, and the second threshold and the third threshold can be several meters or tens of meters / second, which are not specifically limited.

[0101] In a feasible implementation manner, before determining whether there is an overlapping area between the road guardrail and the moving target, the following steps may also be included:

[0102] The running speed information of the moving target obtained by performing target detection processing based on the point cloud data is obtained.

[0103] Based on the running speed information of the moving target, it is determined whether the relative speed between the moving target and the vehicle is less than a fourth threshold and the absolute speed of the moving target is greater than a fifth threshold, to obtain a fifth determination result.

[0104] Correspondingly, the determining whether there is an overlapping area between the road guardrail and the moving target may include:

[0105] If the fifth judgment result indicates that the relative speed between the moving target and the vehicle is less than the fourth threshold, and the absolute speed of the moving target is greater than the fifth threshold, it is determined whether there is an overlapping area between the road guardrail and the moving target.

[0106] In the embodiment of the present invention, when the relative speed between the moving target and the vehicle is large, the credibility of the moving target is usually high, and the possibility that the moving target is a moving target that is mistakenly identified for the road guardrail is low; therefore, it is not necessary to determine whether there is an overlapping area between the moving target and the road guardrail, and the moving target can be directly set as a valid moving target. Also, when the absolute speed of the moving target is too small, the moving target may belong to other stationary targets, and the possibility that the moving target is a moving target that is mistakenly identified for the road guardrail is also low; therefore, it is not necessary to determine whether there is an overlapping area between the moving target and the road guardrail, and the moving target can be directly set as an invalid moving target. Based on this, when it is determined that the relative speed between the moving target and the vehicle is less than the fourth threshold, and the absolute speed of the moving target is greater than the fifth threshold, it is determined whether there is an overlapping area between the road guardrail and the moving target, which is conducive to saving computing resources. Among them, the fourth threshold and the fifth threshold can be set according to actual needs, but the fourth threshold is usually greater than the fifth threshold, and there is no specific limitation on this.

[0107] In a feasible implementation manner, judging whether there is a road guardrail within a preset range around the vehicle based on the point cloud data to obtain a first judgment result may include:

[0108] Based on the point cloud data, determine whether there is a preset range around the vehicle in which the number of stationary point clouds reaches a preset number; wherein the stationary point cloud is a point cloud whose absolute speed is less than or equal to a preset speed.

[0109] If there is a preset range in which the number of stationary point clouds reaches a preset number, a first judgment result is generated indicating that there is a road guardrail within the preset range around the vehicle.

[0110] In the embodiment of the present invention, when the number of stationary point clouds contained in the preset range around the vehicle is too small, it is usually impossible to accurately identify whether there is a road guardrail within the preset range. Therefore, when it is determined that the number of stationary point clouds contained in the preset range around the vehicle reaches the preset number, a first judgment result indicating that there is a road guardrail within the preset range around the vehicle can be generated; otherwise, a first judgment result indicating that there is no road guardrail within the preset range around the vehicle can be generated, which is conducive to improving the accuracy of the recognition result for the road guardrail. The preset number can be set according to actual needs, and it can usually be a positive number greater than or equal to 1, and there is no specific limitation on this.

[0111] In a feasible implementation manner, if the first judgment result indicates that there is a road guardrail within a preset range around the vehicle, then Figure 2 The method may also include:

[0112] Determine whether a second preset condition for changing the state of the road guardrail is met to obtain a sixth judgment result; wherein the second preset condition includes: a relative radial speed between the road guardrail and the vehicle is greater than a sixth threshold, the stationary point cloud contained in the preset range where the road guardrail is located has not changed within the last second preset time length, there is an overlapping area between the road guardrail and a credible non-guardrail target, the vehicle is in a low-speed driving state, and the vehicle is in a turning driving state.

[0113] If the sixth judgment result indicates that the second preset condition for changing the state of the road guardrail is satisfied, the road guardrail is changed from a valid road guardrail state to an invalid road guardrail state.

[0114] If the sixth judgment result indicates that the second preset condition for changing the state of the road guardrail is not met, it is determined that the road guardrail is in a valid road guardrail state.

[0115] The determining whether there is an overlapping area between the road guardrail and the moving target includes:

[0116] If the road guardrail is in a valid road guardrail state, it is determined whether there is an overlapping area between the road guardrail and the moving target.

[0117] In the embodiment of the present invention, in order to reduce the adverse effects of the recognition errors of the road guardrail, it is also possible to continue to judge whether the second preset condition for changing the road guardrail from the valid road guardrail state to the invalid road guardrail state is currently met; if not, it can be determined that the road guardrail is in the valid road guardrail state, so as to allow the judgment of whether the moving target is a false detection target based on the road guardrail; if so, the currently detected road guardrail can be changed to the invalid road guardrail state, and the subsequent judgment of whether the moving target is a false detection target based on the road guardrail will not be continued to ensure the accuracy of the mobile target detection result. Among them, the invalid road guardrail state can indicate that the identified road guardrail is incorrect and has poor credibility; and the valid road guardrail state can indicate that the identified road guardrail has good credibility, which will not be elaborated.

[0118] In actual applications, since the distance between the vehicle and the road guardrail will change when the vehicle changes lanes, the point cloud detected by the radar equipment on the vehicle for the road guardrail will have a relative radial velocity with the vehicle. However, since it usually takes several seconds (for example, 3-10 seconds) for the vehicle to complete the vehicle change operation between adjacent lanes, and the width of a lane is generally several meters (for example, 3-4 meters), the relative radial velocity between the road guardrail and the vehicle is usually not too large. Based on this, the second preset condition that needs to be met to change the road guardrail to an invalid road guardrail state includes: the relative radial velocity between the road guardrail and the vehicle is greater than the sixth threshold; wherein the sixth threshold can be set according to actual needs, for example, it can be 1-3 meters / second, 2-4 meters / second, etc., and there is no specific limitation on this.

[0119] Alternatively, when the static point cloud contained in the preset range where the road guardrail is located has not changed for a long period of time, it can often indicate that the point cloud data detected by the radar device is unreliable, and therefore, the credibility of the road guardrail detected based on the point cloud data is also poor. Based on this, the second preset condition that needs to be met to change the road guardrail to the invalid road guardrail state includes: the static point cloud contained in the preset range where the road guardrail is located has not changed in the last second preset time length. Among them, the second preset time length can be set according to actual needs, for example, it can be from tens of millimeters to several seconds, and there is no specific limitation on this.

[0120] Alternatively, if a credible non-guardrail target that cannot be a road guardrail is determined through target detection processing, then since the credible non-guardrail target should not overlap with the road guardrail, the second preset condition that needs to be satisfied to change the road guardrail to the invalid road guardrail state includes: there is an overlapping area between the road guardrail and the credible non-guardrail target. The credible non-guardrail target may include other stationary targets or moving targets that are too fast to be road guardrails, and there is no specific limitation on this.

[0121] Alternatively, since the vehicle is easily mistakenly identifying the ground as a road guardrail because the point cloud detected by the radar device when the vehicle is driving at a low speed or turning, the vehicle may mistakenly identify the ground as a road guardrail by mistake. Based on this, the second preset condition required to change the road guardrail to an invalid road guardrail state includes: the vehicle is in a low-speed driving state or a turning driving state. In practical applications, the vehicle can be determined to be in a low-speed driving state when the absolute speed of the vehicle is less than a preset threshold. Also, the vehicle can be determined to be in a turning driving state when the parameters such as the steering wheel steering angle, yaw rate, and lateral speed of the vehicle are greater than a preset threshold, which will not be elaborated.

[0122] In a feasible implementation, Figure 2 The method may further include:

[0123] Determine whether a third preset condition for changing the state of the road guardrail is met to obtain a seventh judgment result; wherein the third preset condition includes: the time interval between the current moment and the moment when the road guardrail is changed to the invalid road guardrail state reaches a third preset duration, the stationary point cloud contained in the preset range where the road guardrail is located changes within the most recent fourth preset duration, and the vehicle is in a high-speed straight-ahead state.

[0124] If the seventh judgment result indicates that the third preset condition for changing the state of the road guardrail is met, the road guardrail is changed from the invalid road guardrail state to the valid road guardrail state.

[0125] In an embodiment of the present invention, the operation of changing the road guardrail to an invalid road guardrail state may also be an erroneous operation. Therefore, it is further possible to subsequently identify whether the third preset condition for changing the road guardrail to a valid road guardrail state is met; if so, the road guardrail can be changed from an invalid road guardrail state to a valid road guardrail state, thereby continuing to identify whether the moving target is a falsely detected target based on the road guardrail; if not, the road guardrail can be no longer tracked after the duration of the road guardrail being in the invalid road guardrail state reaches a preset threshold, which is beneficial for saving computing resources.

[0126] In actual application, the third preset condition can be set according to actual needs, and no specific limitation is made to this. However, since the time distance between the current moment and the moment when the road guardrail is changed to the invalid road guardrail state is large, the point cloud detected by the radar device for the road guardrail will usually change. Therefore, the third preset condition for changing the road guardrail to the valid road guardrail state may include: the time interval between the current moment and the moment when the road guardrail is changed to the invalid road guardrail state reaches the third preset time length, and the static point cloud contained in the preset range where the road guardrail is located has changed within the latest fourth preset time length. In addition, since the vehicle usually does not mistake the point cloud data collected by the radar device on the ground as the point cloud data collected for the road guardrail when it is in a high-speed straight state, the road guardrail identified based on the point cloud data usually has good credibility, so that the third preset condition can also include: the vehicle is in a high-speed straight state to ensure the accuracy and reliability of the state change of the road guardrail.

[0127] Figure 5A flowchart of a blind spot alarm control method provided in an embodiment of the present invention. From a program perspective, the execution subject of the process can be a program installed in a vehicle control device (e.g., Microcontroller Unit, MCU; or, domain controller) or a vehicle. Alternatively, the execution subject of the process can also be a vehicle control device or a vehicle, or other devices that can communicate with the vehicle control device or the vehicle, without specific limitation.

[0128] The following is for Figure 5 The process shown in the figure is described in detail, and the blind spot alarm control method can specifically include the following steps:

[0129] Step S502, obtaining point cloud data collected by the radar equipment mounted on the vehicle for the surrounding environment.

[0130] Step S504: performing target detection processing based on the point cloud data to obtain moving targets around the vehicle.

[0131] Step S506, based on the point cloud data, determine whether there is a road guardrail within a preset range around the vehicle to obtain a first judgment result; the preset range includes the lateral visual blind spot of the vehicle.

[0132] Step S508, if the first judgment result indicates that there is a road guardrail within a preset range around the vehicle, then based on the first position information of the road guardrail and the second position information of the moving target, it is determined whether there is an overlapping area between the road guardrail and the moving target to obtain a second judgment result.

[0133] Step S510: If the second judgment result indicates that there is an overlapping area between the road guardrail and the moving target, control information is generated for prohibiting the lateral blind spot monitoring function of the vehicle from issuing an early warning for the moving target.

[0134] In the embodiment of the present invention, due to Figure 5 The implementation principle of steps S502-S508 in Figure 2 The implementation principles of steps S202-S208 in the embodiment may be consistent, so they will not be described in detail.

[0135] In an embodiment of the present invention, the lateral blind spot detection function may refer to a function of monitoring the lateral visual blind spot of a vehicle through radar equipment around the vehicle body, and issuing an early warning to the driver when a vehicle is detected entering the lateral visual blind spot of the vehicle.

[0136] When the preset range around the vehicle that can be detected by the radar equipment installed in the vehicle includes the side vision blind spot of the vehicle (for example, Figure 1 When the vehicle detects a road guardrail in the lateral field of view of the vehicle, and there is an overlapping area between a certain moving target and the road guardrail, the moving target is usually a false detection target caused by the road guardrail and noise factors. Therefore, the control information for prohibiting the lateral blind spot monitoring function of the vehicle from issuing an early warning for the moving target can be generated, thereby reducing the false alarm rate of the lateral blind spot monitoring function, which is conducive to improving the user experience.

[0137] Optionally, due to the limited sensing range of the radar equipment, and the fact that the distance from the central axis of the vehicle is less than 0.5 times the width of the vehicle, there is usually no road guardrail. Based on this, the preset range around the vehicle can be: the distance from the central axis of the vehicle is greater than or equal to 0.5 times the width of the vehicle and less than or equal to the seventh threshold, and the distance from the center point of the vehicle is less than or equal to the eighth threshold. The seventh threshold and the eighth threshold can be set according to actual needs. For example, the seventh threshold can be 3-5 meters, 4-6 meters, etc., and the eighth threshold can be several meters or tens of meters, etc., without specific limitation.

[0138] In one embodiment, the present invention also provides Figure 6 The structural diagram of the domain controller is shown in Figure 1. Figure 6 At the hardware level, the domain controller may include a processor 61 and a memory 65, and may also include an internal bus 62, a network interface 63, a memory 64, and other hardware required for other services. The domain controller may be set in a vehicle, wherein the processor 61 may read the corresponding computer-readable instructions from the memory 65 into the memory and then run them to implement the above-mentioned target detection method and the blind spot alarm control method. The specific execution process may refer to the specific descriptions in the above-mentioned embodiments, and will not be repeated here.

[0139] In one embodiment, the present invention further provides a computer program product having at least one instruction stored thereon, and when the at least one instruction is executed by a controller, the steps of the above-mentioned target detection method and blind spot alarm control method are implemented. The specific execution process can be referred to the specific description in the above-mentioned embodiments, and will not be repeated here.

[0140] Finally, each embodiment of the present invention is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the embodiments such as the domain controller and the computer program product, since they are basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0141] The above description is only an embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.

Claims

1. A target detection method, comprising: Obtain point cloud data collected by the radar equipment on the vehicle for the surrounding environment; Performing target detection processing based on the point cloud data to obtain moving targets around the vehicle; Based on the point cloud data, determining whether there is a road guardrail within a preset range around the vehicle to obtain a first determination result; If the first judgment result indicates that there is a road guardrail within the preset range around the vehicle, judging whether there is an overlapping area between the road guardrail and the moving target according to the first position information of the road guardrail and the second position information of the moving target, and obtaining a second judgment result; If the second judgment result indicates that there is an overlapping area between the road guardrail and the moving object, it is determined that the moving object is a misdetected object.

2. The method according to claim 1, further comprising: Based on the point cloud data, from various preset ranges around the vehicle, the preset range containing the largest number of stationary point clouds is determined to obtain a target preset range where the road guardrail exists; wherein the stationary point cloud is a point cloud whose absolute speed is less than or equal to a preset speed; The average value of the distance values ​​between each of the stationary point clouds within the preset target range and the central axis of the vehicle is calculated to obtain the first position information reflecting the target distance value between the road guardrail and the central axis of the vehicle.

3. The method according to claim 1, further comprising: Acquire the area information of the mobile target obtained by performing target detection processing based on the point cloud data to obtain second position information of the mobile target; or, Acquiring information about the area where the mobile target is located, obtained by performing target detection processing based on the point cloud data; Based on the area information of the moving target, the target distance range between the virtual front bumper and the virtual rear bumper of the moving target and the central axis of the vehicle is determined to obtain the second position information of the moving target.

4. The method according to claim 1, before determining that the moving target is a misdetected target, further comprising: Determine whether the duration of the overlapping area between the road guardrail and the moving target reaches a first threshold, and obtain a third determination result; Determining that the moving target is a misdetected target specifically includes: If the third judgment result indicates that the duration of the overlapping area between the road guardrail and the moving target reaches a first threshold, the moving target is changed from a valid moving target state to a falsely detected target state.

5. The method according to claim 4, after changing the mobile target from a valid mobile target state to a misdetected target state, further comprising: Determine whether a first preset condition for changing the state of the moving target is satisfied, and obtain a fourth determination result; wherein the first preset condition includes: the moving target and the road guardrail have no overlapping area within a recent first preset time period, the relative speed between the moving target and the vehicle reaches a second threshold, and the absolute speed of the moving target reaches a third threshold; If the fourth determination result indicates that the first preset condition is satisfied, the moving object is changed from the misdetected object state to the valid moving object state.

6. The method according to claim 1, before determining whether there is an overlapping area between the road guardrail and the moving target, further comprising: Acquire the running speed information of the moving target obtained by performing target detection processing based on the point cloud data; Based on the running speed information of the moving target, determining whether the relative speed between the moving target and the vehicle is less than a fourth threshold, and the absolute speed of the moving target is greater than a fifth threshold, to obtain a fifth determination result; The determining whether there is an overlapping area between the road guardrail and the moving target includes: If the fifth judgment result indicates that the relative speed between the moving target and the vehicle is less than the fourth threshold, and the absolute speed of the moving target is greater than the fifth threshold, it is determined whether there is an overlapping area between the road guardrail and the moving target.

7. The method according to claim 1, wherein judging whether there is a road guardrail within a preset range around the vehicle based on the point cloud data to obtain a first judgment result comprises: Based on the point cloud data, determining whether there is a preset range in each preset range around the vehicle in which the number of stationary point clouds reaches a preset number; wherein the stationary point cloud is a point cloud whose absolute speed is less than or equal to a preset speed; If there is a preset range in which the number of stationary point clouds reaches a preset number, a first judgment result is generated indicating that there is a road guardrail within the preset range around the vehicle.

8. The method according to claim 7, if the first judgment result indicates that there is a road guardrail within a preset range around the vehicle, the method further comprises: Determine whether a second preset condition for changing the state of the road guardrail is met to obtain a sixth judgment result; wherein the second preset condition includes: the relative radial speed between the road guardrail and the vehicle is greater than a sixth threshold, the stationary point cloud contained in the preset range where the road guardrail is located has not changed within the latest second preset time length, there is an overlapping area between the road guardrail and the credible non-guardrail target, the vehicle is in a low-speed driving state, and the vehicle is in a turning driving state. If the sixth judgment result indicates that the second preset condition for changing the state of the road guardrail is satisfied, the road guardrail is changed from a valid road guardrail state to an invalid road guardrail state; If the sixth judgment result indicates that the second preset condition for changing the state of the road guardrail is not satisfied, determining that the road guardrail is in a valid road guardrail state; The determining whether there is an overlapping area between the road guardrail and the moving target includes: If the road guardrail is in a valid road guardrail state, it is determined whether there is an overlapping area between the road guardrail and the moving target.

9. The method according to claim 8, further comprising: Determine whether a third preset condition for changing the state of the road guardrail is satisfied, and obtain a seventh judgment result; wherein the third preset condition includes: the time interval between the current moment and the moment when the road guardrail is changed to the invalid road guardrail state reaches a third preset time length, the stationary point cloud included in the preset range where the road guardrail is located changes within the latest fourth preset time length, and the host vehicle is in a high-speed straight driving state; If the seventh judgment result indicates that the third preset condition for changing the state of the road guardrail is met, the road guardrail is changed from the invalid road guardrail state to the valid road guardrail state.

10. A blind spot alarm control method, comprising: Obtain point cloud data collected by the radar equipment on the vehicle for the surrounding environment; Performing target detection processing based on the point cloud data to obtain moving targets around the vehicle; Based on the point cloud data, determining whether there is a road guardrail within a preset range around the vehicle to obtain a first determination result; the preset range includes a lateral visual blind spot of the vehicle; If the first judgment result indicates that there is a road guardrail within the preset range around the vehicle, judging whether there is an overlapping area between the road guardrail and the moving target according to the first position information of the road guardrail and the second position information of the moving target, and obtaining a second judgment result; If the second judgment result indicates that there is an overlapping area between the road guardrail and the moving target, control information for prohibiting the lateral blind spot monitoring function of the vehicle from issuing an early warning for the moving target is generated.

11. According to the method of claim 12, the preset range is: the distance between the central axis of the vehicle is greater than or equal to 0.5 times the width of the vehicle and less than or equal to the seventh threshold, and the distance between the center point of the vehicle is less than or equal to the eighth threshold.

12. A domain controller, comprising: A processor and a memory; wherein the memory stores computer-readable instructions, and the computer-readable instructions are suitable for being loaded by the processor and executing the steps of the method as claimed in any one of claims 1 to 11.

13. A computer program product having at least one instruction stored thereon, wherein when the at least one instruction is executed by a controller, the steps of the method according to any one of claims 1 to 11 are implemented.

Images