Target screening method, device, equipment, medium and vehicle

Abstract

This disclosure relates to a target screening method, apparatus, device, medium, and vehicle. The target screening method includes: acquiring target attribute information; wherein the attribute information includes location information and at least one of the following: size information, motion information, presence probability, obstacle probability, overlap rate, scattering area, and collision probability; based on the location information, selecting targets located in a preset area as first candidate targets; and based on the attribute information, selecting candidate targets that meet preset screening conditions from the first candidate targets as valid targets. Thus, embodiments of this disclosure can screen targets based on their attribute information, thereby selecting valid targets. This allows the attribute information of valid targets to be uploaded to the Automatic Emergency Response (AEB) system instead of uploading the attribute information of all targets to the AEB, facilitating rapid decision-making by the AEB based on the attribute information of valid targets and shortening the AEB response time.

Description

Technical Field

[0001] The present disclosure relates to the field of vehicle technology, and in particular to a target screening method, device, equipment, medium and vehicle. Background Art

[0002] Autonomous Emergency Braking (AEB) is an active safety system that activates the vehicle's brakes when a potential collision is detected. It works automatically without the driver actually touching the brake pedal. It can also increase braking force if the driver is applying the brakes but not enough to prevent a collision.

[0003] Vehicles are usually equipped with detection equipment (such as millimeter wave radar) to detect targets around the vehicle and upload the information of the detected targets to AEB, so that AEB can output effective decisions based on the received target information to avoid collisions. At present, detection equipment usually uploads the information of all detected targets. However, among the information of many uploaded targets, there is a lot of information that we don’t need, which is not conducive to AEB making timely and effective decisions, resulting in a long AEB response time. Summary of the invention

[0004] In order to solve the above technical problems, the present disclosure provides a target screening method, device, equipment, medium and vehicle.

[0005] In a first aspect, the present disclosure provides a target screening method, comprising:

[0006] Acquire attribute information of the target; wherein the attribute information includes position information and at least one of the following: size information, motion information, existence probability, obstacle probability, overlap rate, scattering area, and collision probability;

[0007] Based on the location information, a target located in a preset area is selected from the targets as a first candidate target;

[0008] Based on the attribute information, candidate targets that meet preset screening conditions are screened out from the first candidate targets as valid targets.

[0009] In a second aspect, the present disclosure provides a target screening device, comprising:

[0010] A first acquisition module acquires attribute information of a target; wherein the attribute information includes position information and at least one of the following: size information, motion information, existence probability, obstacle probability, overlap rate, scattering area, and collision probability;

[0011] A first screening module, based on the location information, screens out targets located in a preset area from the targets as first candidate targets;

[0012] The second screening module screens out candidate targets that meet preset screening conditions from the first candidate targets based on the attribute information as valid targets.

[0013] In a third aspect, the present disclosure provides a target screening device, comprising:

[0014] processor;

[0015] A memory for storing executable instructions;

[0016] The processor is used to read executable instructions from the memory and execute the executable instructions to implement the target screening method described in the first aspect.

[0017] In a fourth aspect, the present disclosure provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor implements the target screening method described in the first aspect.

[0018] In a fifth aspect, the present disclosure provides a vehicle, comprising at least one of the following:

[0019] The target screening device described in the second aspect above;

[0020] The target screening device described in the third aspect above;

[0021] The computer-readable storage medium described in the fourth aspect above.

[0022] Compared with the prior art, the technical solution provided by the embodiments of the present disclosure has the following advantages:

[0023] The target screening method, device, equipment, medium and vehicle of the embodiments of the present disclosure can obtain attribute information of the target; wherein the attribute information includes position information and at least one of the following: size information, motion information, existence probability, obstacle probability, overlap rate, scattering area, and collision probability; based on the position information, the target located in a preset area is screened out from the targets as the first candidate target; based on the attribute information, the candidate targets that meet the preset screening conditions are screened out from the first candidate targets as valid targets. In this way, the embodiments of the present disclosure can screen the targets based on the attribute information of the targets, thereby screening out valid targets, so that the attribute information of the valid targets can be uploaded to the AEB later instead of uploading the attribute information of all targets to the AEB, which is beneficial for the AEB to make quick decisions based on the attribute information of the valid targets, thereby shortening the response time of the AEB. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale.

[0025] Figure 1 A schematic diagram of a target screening method provided by an embodiment of the present disclosure is shown;

[0026] Figure 2 A schematic diagram of another target screening method provided by an embodiment of the present disclosure is shown;

[0027] Figure 3 A schematic diagram of a process of another target screening method provided by an embodiment of the present disclosure is shown;

[0028] Figure 4 A schematic diagram of a process flow of another target screening method provided by an embodiment of the present disclosure is shown;

[0029] Figure 5 A schematic diagram of a target screening process provided by an embodiment of the present disclosure is shown;

[0030] Figure 6 A schematic diagram of the structure of a target screening device provided by an embodiment of the present disclosure is shown;

[0031] Figure 7 A schematic structural diagram of a target screening device provided in an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0032] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0033] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.

[0034] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0035] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0036] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0037] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0038] In order to solve the above problems, the embodiments of the present disclosure provide a target screening method, device, equipment, medium and vehicle. Below, the target screening method provided by the embodiments of the present disclosure is first described.

[0039] Figure 1 A schematic flow chart of a target screening method provided in an embodiment of the present disclosure is shown.

[0040] In some embodiments of the present disclosure, Figure 1 The target screening method shown can be executed by a target screening device, and the target screening device can be applied to a detection device capable of detecting targets around a vehicle, such as a millimeter wave radar, without specific limitation.

[0041] like Figure 1 As shown, the target screening method may include the following steps.

[0042] S110: Obtain attribute information of the target.

[0043] The attribute information includes location information and at least one of the following: size information, motion information, existence probability, obstacle probability, overlap rate, scattering area, and collision probability.

[0044] In the embodiment of the present disclosure, the target screening device may obtain attribute information of targets around the target vehicle.

[0045] Specifically, the target vehicle may be any vehicle, for example, the target vehicle may be the vehicle where the target screening device is located, or may be any vehicle other than the vehicle where the target screening device is located, which is not limited here.

[0046] Specifically, the target may be any target around the target vehicle, such as vehicles, bicycles, pedestrians, manhole covers, road boundaries, etc. around the vehicle, but is not limited thereto.

[0047] Specifically, the position information may include at least one of the following: the X-direction distance difference between the target and the target vehicle (i.e., the longitudinal distance difference), the standard deviation of the X-direction distance difference, the Y-direction distance difference between the target and the vehicle (i.e., the lateral distance difference), the standard deviation of the Y-direction distance difference, and the Z-direction distance difference between the target and the ground (i.e., the vertical distance difference), but is not limited to these.

[0048] Among them, the X direction, Y direction and Z direction are perpendicular to each other. The X direction can be the driving direction of the target vehicle, the Y direction can be a direction perpendicular to the driving direction and parallel to the road surface where the target vehicle is located (i.e., the left and right direction of the target vehicle), and the Z direction can be perpendicular to the road surface where the target vehicle is located.

[0049] Specifically, the size information may include at least one of the following: length, width, and height of the target.

[0050] Specifically, the motion information may include at least one of the following: the target's X-axis acceleration, the standard deviation of the X-axis acceleration, the target's Y-axis acceleration, the standard deviation of the Y-axis acceleration, and the motion probability, where the motion probability is used to characterize the probability that the target is in a moving state relative to the ground, but is not limited to this.

[0051] Specifically, the existence probability is used to characterize the probability that a target is a real target (a real target as opposed to a virtual target).

[0052] Specifically, the obstacle probability is used to characterize the probability that a target obstructs the driving of a target vehicle.

[0053] Specifically, the overlap rate is used to characterize the overlap degree between the target and the target vehicle in the Y direction. For example, when the target is located directly in front of the target vehicle in the driving direction, the overlap rate is 100%.

[0054] Specifically, the scattering area is used to characterize the intensity of the detection wave reflected back by the target after the target detection device sends the detection wave to the target.

[0055] Specifically, the collision probability is used to characterize the probability of a collision between a target and a target vehicle.

[0056] In some embodiments, S110 may include: receiving attribute information of the target sent by the detection module.

[0057] Specifically, the target vehicle is provided with a detection device, and the detection device includes a detection module capable of detecting targets around the target vehicle.

[0058] S120. Based on the location information, select a target located in a preset area from the targets as a first candidate target.

[0059] In the disclosed embodiment, the target screening device may screen out targets located in a preset area as first candidate targets, that is, filter out targets located outside the preset area.

[0060] Specifically, the preset area can be an area delineated based on the position of the target vehicle. The position of the target vehicle mentioned here is the position of the target vehicle when the detection module detects the targets around the target vehicle and obtains the attribute information of the targets. For example, when the detection module detects the targets around the target vehicle and obtains the attribute information of the targets, the position of the target vehicle is position A, then the preset area used for screening the target is the area delineated based on position A.

[0061] Specifically, the targets screened out after screening the targets described in S110 are all candidate targets. In order to clearly indicate the targets screened out after different screening strategies, relational terms such as "first" and "second" are used to distinguish the targets screened out after different screening strategies.

[0062] In some embodiments, S120 may include: based on the position information, selecting a target whose X-direction distance difference is less than a preset X-direction distance difference threshold from the targets as a first candidate target.

[0063] In some other embodiments, S120 may include: based on the position information, screening out, from the targets, targets whose Y-direction distance difference satisfies a preset Y-direction distance difference threshold as first candidate targets.

[0064] In some other embodiments, S120 may include: based on the position information, selecting from the targets a target whose X-direction distance difference is less than a preset X-direction distance difference threshold and whose Y-direction distance difference is less than a preset Y-direction distance difference threshold as a first candidate target.

[0065] Specifically, those skilled in the art may set specific values ​​of the preset X-direction distance difference threshold and the preset Y-direction distance difference threshold according to actual conditions, which are not limited here. For example, the preset X-direction distance difference threshold may be 100 meters, 120 meters, 150 meters, etc., and the preset Y-direction distance difference threshold may be 2 meters, 3 meters, etc., but are not limited thereto.

[0066] It can be understood that when the X-direction distance difference of the target is greater than or equal to the preset X-direction distance difference threshold, it indicates that the target is far away from the target vehicle in the driving direction, and the possibility of collision between the target vehicle and this type of target is extremely small. Filtering out this type of target can improve the decision efficiency of AEB. The same is true when the Y-direction distance difference of the target is greater than or equal to the preset Y-direction distance difference threshold, which will not be repeated here.

[0067] It can also be understood that by setting up to filter out targets located in a preset area from the targets as the first candidate targets, a large number of targets with very low possibility of colliding with the target vehicle can be filtered out first, which can not only significantly reduce the base number of targets faced in subsequent screening, but also almost never miss targets with a high possibility of colliding with the target vehicle due to excessive filtering. This is conducive to improving the target screening speed and accuracy.

[0068] S130. Based on the attribute information, select candidate targets that meet preset screening conditions from the first candidate targets as valid targets.

[0069] In the disclosed embodiment, the target screening device may continue to screen the first candidate target, and screen out the candidate targets that meet the preset screening conditions as valid targets.

[0070] Specifically, the valid target may be a target that plays an important role in AEB decision-making. Subsequent target screening may upload attribute information of the valid target to AEB.

[0071] Specifically, the preset screening condition may be a screening condition matched and set based on at least one item in the attribute information. For example, the preset screening condition may include at least one of the following: existence probability is greater than a preset existence probability threshold, obstacle probability is greater than a preset obstacle probability threshold, overlap rate is greater than a preset overlap rate probability threshold, does not belong to a preset type, longitudinal standard deviation is greater than a preset longitudinal standard deviation threshold, height is greater than a preset height threshold, scattering area is greater than a preset scattering area threshold, position information is the same as the verification position information, opposite to the target vehicle speed direction, lateral standard deviation is greater than a preset lateral standard deviation threshold, collision probability is greater than a preset collision probability threshold, but is not limited thereto.

[0072] The target screening method of the embodiment of the present disclosure can obtain attribute information of the target; wherein the attribute information includes at least one of the following: position information, size information, motion information, existence probability, obstacle probability, overlap rate, scattering area, and collision probability; based on the position information, the target located in a preset area is screened out from the targets as the first candidate target; based on the attribute information, the candidate targets that meet the preset screening conditions are screened out from the first candidate targets as valid targets. In this way, the embodiment of the present disclosure can screen the targets based on the attribute information of the targets, thereby screening out valid targets, so that the attribute information of the valid targets can be subsequently uploaded to the AEB instead of uploading the attribute information of all targets to the AEB, which is beneficial for the AEB to make quick decisions based on the attribute information of the valid targets, thereby shortening the response time of the AEB.

[0073] In another embodiment of the present disclosure, S130 may include:

[0074] S1310. Filter out, from the first candidate targets, the first candidate targets that satisfy at least one of the following conditions: existence probability greater than a preset existence probability threshold, obstacle probability greater than a preset obstacle probability threshold, and overlap rate greater than a preset overlap rate probability threshold, as the second candidate targets.

[0075] Specifically, the specific values ​​of the preset existence probability threshold, the preset obstacle probability threshold, and the preset overlap rate probability threshold can be set by those skilled in the art according to actual conditions and are not limited here. For example, the preset existence probability threshold may be 90%, the preset obstacle probability threshold may be 60%, and the preset overlap rate probability threshold may be 25%, but are not limited thereto.

[0076] It should be noted that, when a first candidate target that satisfies at least two of the conditions that the existence probability is greater than a preset existence probability threshold, the obstacle probability is greater than a preset obstacle probability threshold, and the overlap rate is greater than a preset overlap rate probability threshold is screened out from the first candidate target as the second candidate target, the order of screening based on the items is not limited, and there can be a variety of combinations. For example, when a first candidate target that satisfies the conditions that the existence probability is greater than a preset existence probability threshold, the obstacle probability is greater than a preset obstacle probability threshold, and the overlap rate is greater than a preset overlap rate probability threshold is screened out from the first candidate target as the second candidate target, there are 6 combinations based on the order of screening these three items. One is, first, from the first candidate target, screen out the first candidate target that satisfies the conditions that the existence probability is greater than a preset existence probability threshold, the obstacle probability is greater than a preset obstacle probability threshold, and the overlap rate is greater than a preset overlap rate probability threshold. Suppose there are candidate targets with a probability threshold for existence, and then, continue to screen out candidate targets that satisfy the obstacle probability greater than the preset obstacle probability threshold from the screened candidate targets, and finally, continue to screen out candidate targets that satisfy the overlap rate greater than the preset overlap rate probability threshold from the screened candidate targets as the second candidate targets; the other is, first, from the first candidate target, screen out candidate targets that satisfy the obstacle probability greater than the preset obstacle probability threshold, and then continue to screen out candidate targets that satisfy the existence probability greater than the preset existence probability threshold from the screened candidate targets, and finally, continue to screen out candidate targets that satisfy the overlap rate greater than the preset overlap rate probability threshold from the screened candidate targets as the second candidate targets, and so on, and no further details are given.

[0077] It can be understood that when the existence probability of the first candidate target is less than or equal to the preset existence probability threshold, it indicates that the first candidate target is likely to be a virtual target, that is, it is not a real target, and the possibility of a collision between the target vehicle and this type of target is very small. Filtering out this type of target can further improve the decision-making efficiency of AEB.

[0078] It can be understood that when the obstacle probability of the first candidate target is less than or equal to the preset obstacle probability threshold, it indicates that the first candidate target is less likely to hinder the driving of the target vehicle, and the possibility of the target vehicle colliding with this type of target is very small. Filtering out this type of target can further improve the decision-making efficiency of AEB.

[0079] It can be understood that when the overlap rate probability of the first candidate target is less than or equal to the preset overlap rate probability threshold, it indicates that the first candidate target deviates from the driving direction of the target vehicle to a large extent, and the possibility of a collision between the target vehicle and this type of target is very small. Filtering out this type of target can further improve the decision-making efficiency of AEB.

[0080] It can also be understood that by filtering out candidate targets whose existence probability is less than or equal to the preset existence probability threshold, the candidate targets with a small probability of colliding with the target vehicle are filtered out, and targets with a high probability of colliding with the target vehicle are almost never missed due to excessive filtering, thus avoiding over-screening. The same is true for filtering out candidate targets with an obstacle probability less than or equal to the preset obstacle probability threshold and filtering out candidate targets with an overlap rate probability less than or equal to the preset overlap rate probability threshold, which will not be elaborated here.

[0081] S1320: Filter out candidate targets whose scattering areas meet a scattering area threshold greater than a preset scattering area threshold from the second candidate targets as third candidate targets.

[0082] Specifically, the specific value of the preset scattering area threshold can be set by those skilled in the art according to actual conditions and is not limited here. For example, when more attention is paid to avoiding collision between the target vehicle and surrounding vehicles, the preset scattering area threshold can be determined based on the scattering areas of multiple vehicles. Since the material of the vehicle is usually metal, and the reflectivity of metal is relatively high, the preset scattering area threshold can be set higher. For example, the preset scattering area threshold can be 10DB, but is not limited to this.

[0083] It can be understood that by screening out candidate targets whose scattering areas are greater than a preset scattering area threshold from the second candidate targets, the cardinality of targets faced in subsequent screening based on collision probability can be reduced, further improving the screening speed.

[0084] S1330: Filter out valid targets from the third candidate targets based on the collision probability.

[0085] In some embodiments, S1330 may include: selecting, from the third candidate targets, candidate targets whose collision probability is greater than or equal to a preset collision probability threshold as valid targets.

[0086] Specifically, the specific value of the preset collision probability threshold can be set by those skilled in the art according to actual conditions and is not limited here.

[0087] Figure 2 The flowchart of another target screening method provided by the embodiment of the present disclosure is shown. The embodiment of the present disclosure is optimized on the basis of the above embodiment, and the embodiment of the present disclosure can be combined with various optional solutions in one or more of the above embodiments.

[0088] like Figure 2 As shown, the target screening may include the following steps.

[0089] S210: Obtain attribute information of the target.

[0090] Specifically, S210 is similar to S110 and will not be described in detail here.

[0091] S220: Based on the location information, select a target located in a preset area from the targets as a first candidate target.

[0092] Specifically, S220 is similar to S120 and will not be described in detail here.

[0093] S230. Filter out, from the first candidate targets, the first candidate targets that satisfy at least one of the following conditions: existence probability greater than a preset existence probability threshold, obstacle probability greater than a preset obstacle probability threshold, and overlap rate greater than a preset overlap rate probability threshold, as second candidate targets.

[0094] Specifically, S230 is similar to S1310 and will not be described in detail here.

[0095] S240: Filter out candidate targets of a preset type from the second candidate targets based on at least one of the position information, the size information, the motion information, and the scattering area, and use the remaining candidate targets as fourth candidate targets.

[0096] Specifically, the target types included in the preset types can be set by those skilled in the art according to actual conditions and are not limited here.

[0097] Optionally, the preset types include: at least one of a mirror target, a manhole cover, a pedestrian, a bicycle, and a road boundary.

[0098] Specifically, a mirror target refers to a mirror image of a real target around the target vehicle. For example, when the target vehicle is traveling in a tunnel, after the detection module of the detection device sends electromagnetic waves to the target, part of the electromagnetic waves are reflected by the target to the tunnel wall and then reflected back again and received by the detection module, then the detection module can detect the target and the mirror image corresponding to the target (i.e., the mirror image target), and since the moment when the detection module receives the electromagnetic wave corresponding to the mirror image target (i.e., the detection moment of the mirror image target) is later than the moment when the electromagnetic wave reflected back by the target is received (i.e., the detection moment of the real target), therefore, under normal circumstances, it is possible to determine which target is the real target and which target is the mirror image target based on the moment when the detection module detects the target and its corresponding mirror image target.

[0099] Since the target and its mirror image have the following characteristics: the same size and symmetry, the candidate target that is the mirror image target can be filtered out from the second candidate targets based on the position information and size information. Optionally, when the size information of the two second candidate targets is the same and the Y-direction distance difference between the target and the vehicle is the same, the second candidate target with a later detection time is determined to be the mirror image target; and the mirror image target is filtered out from the second candidate targets.

[0100] Specifically, since the manhole cover has the following characteristics: since the material is usually metal, the reflectivity is high, and since it is located on the ground, the height is almost 0, therefore, the candidate targets belonging to the manhole cover can be filtered out from the second candidate targets based on the size information and the scattering area. Optionally, from the second candidate targets, the candidate targets that meet the conditions that the scattering area is greater than the first preset scattering area threshold and the height is less than the first preset height threshold are filtered out. Of course, the manhole cover also has the characteristic of being in a stationary state, therefore, the candidate targets belonging to the manhole cover can also be filtered out from the second candidate targets based on the size information, the scattering area, and the motion information. Optionally, from the second candidate targets, the candidate targets that meet the conditions that the scattering area is greater than the first preset scattering area threshold, the height is less than the first preset height threshold, and the motion probability is less than the preset motion probability threshold are filtered out, so that the accuracy of the manhole cover filtering can be improved. The specific values ​​of the first preset scattering area threshold, the first preset height threshold, and the preset motion probability threshold can be set by those skilled in the art according to actual conditions, and are not limited here.

[0101] Specifically, since pedestrians have the following characteristics: slender, candidate targets belonging to pedestrians can be filtered out from the second candidate targets based on size information. Optionally, candidate targets that satisfy a ratio of length to height less than a first preset ratio and a ratio of width to height less than a second preset ratio are filtered out from the second candidate targets. The specific values ​​of the first preset ratio and the second preset ratio can be set by those skilled in the art according to actual conditions and are not limited here.

[0102] Specifically, since the bicycle has the following characteristics: flat, the candidate object belonging to the bicycle can be filtered out from the second candidate objects based on the size information. Optionally, the candidate objects whose width to height ratio is less than the third preset ratio are filtered out from the second candidate objects. The specific value of the third preset ratio can be set by those skilled in the art according to actual conditions and is not limited here.

[0103] Specifically, the road boundary refers to objects located on the edges of both sides of the road, such as bushes and guardrails on both sides of the road, but is not limited thereto.

[0104] Since the road boundary has the following characteristics: the part of the road boundary detected by the detection module of the detection device in the current cycle is continuous with the part of the road boundary measured in the previous cycle, and the size is the same. Therefore, the candidate targets belonging to the road boundary can be filtered out from the second candidate targets based on the position information and size information. Optionally, when the size information of the second candidate target is the same as the size information of a certain target in the previous cycle, and the X-direction distance difference, Y-direction distance difference, and Z-direction distance difference of the second candidate target are continuous with the X-direction distance difference, Y-direction distance difference, and Z-direction distance difference of the target in the previous cycle, the second candidate target is determined to be a road boundary; the road boundary is filtered out from the second candidate targets. Of course, the road boundary also has the characteristic of being in a static state, therefore, the candidate targets belonging to the road boundary can also be filtered out from the second candidate targets based on the position information, size information, and motion information. Optionally, when the size information of the second candidate target is the same as the size information of a target in the previous cycle, and the X-direction distance difference, Y-direction distance difference, and Z-direction distance difference of the second candidate target are continuous with the X-direction distance difference, Y-direction distance difference, and Z-direction distance difference of the target in the previous cycle, and the motion probability is less than a preset motion probability threshold, the second candidate target is determined to be a road boundary; and the road boundary is filtered out from the second candidate targets, thereby improving the accuracy of road boundary filtering.

[0105] It is understandable that in traffic accidents, most of them are vehicle-to-vehicle collisions. Therefore, the vehicles around the target vehicle should be the targets that need the most attention. Targets of preset types are less likely to collide with the vehicle (such as pedestrians, bicycles, road boundaries, etc.) or will not collide with the vehicle (such as mirror targets and manhole covers, etc.), so they can be filtered out to further improve the decision-making efficiency of AEB.

[0106] S250 , selecting a candidate target whose scattering area satisfies a scattering area threshold greater than a preset scattering area threshold from the fourth candidate targets as a third candidate target.

[0107] Specifically, S250 is similar to S1320 and will not be described in detail here.

[0108] S260: Filter out valid targets from the third candidate targets based on the collision probability.

[0109] Specifically, S260 is similar to S1330 and will not be described in detail here.

[0110] The target screening method provided by the embodiment of the present disclosure can further filter out targets with relatively small possibility of colliding with the target vehicle by setting a filter out candidate targets of a preset type from the second candidate targets based on at least one of position information, size information, motion information and scattering area, thereby further reducing the cardinality of targets faced in subsequent screening, and almost never missing targets with high possibility of colliding with the target vehicle due to excessive filtering, which is conducive to further improving the target screening speed and accuracy, and further improving the AEB decision-making efficiency.

[0111] Figure 3 The flowchart of another target screening method provided by the embodiment of the present disclosure is shown. The embodiment of the present disclosure is optimized on the basis of the above embodiment, and the embodiment of the present disclosure can be combined with various optional solutions in one or more of the above embodiments.

[0112] like Figure 3 As shown, the target screening may include the following steps.

[0113] S310: Obtain attribute information of the target.

[0114] Specifically, S310 is similar to S110 and will not be described in detail here.

[0115] S320: Based on the location information, select a target located in a preset area from the targets as a first candidate target.

[0116] Specifically, S320 is similar to S120 and will not be described in detail here.

[0117] S330. Filter out, from the first candidate targets, the first candidate targets that satisfy at least one of the following conditions: existence probability greater than a preset existence probability threshold, obstacle probability greater than a preset obstacle probability threshold, and overlap rate greater than a preset overlap rate probability threshold, as the second candidate targets.

[0118] Specifically, S330 is similar to S1310 and will not be described in detail here.

[0119] S340. Filter out, from the second candidate targets, candidate targets that satisfy at least one of a longitudinal standard deviation greater than a preset longitudinal standard deviation threshold and a height greater than a preset height threshold as fifth candidate targets.

[0120] Optionally, before S340, it may also include: from the second candidate targets, based on at least one of the position information, size information, motion information and scattering area, filtering out candidate targets belonging to a preset type, and taking the remaining candidate targets as fourth candidate targets; at this time, S340 may include: from the fourth candidate targets, screening out candidate targets that meet at least one of the longitudinal standard deviation being greater than a preset longitudinal standard deviation threshold and the height being less than a preset height threshold, as fifth candidate targets.

[0121] Specifically, the location information includes a longitudinal standard deviation threshold. After the detection module of the detection device detects a target in the current cycle and obtains the attribute information of the target, the X-direction distance difference can be compared with the X-direction distance difference of the target detected in the historical cycle to calculate the X-direction standard deviation, that is, the longitudinal standard deviation.

[0122] Specifically, the size information includes height.

[0123] Specifically, the specific values ​​of the preset longitudinal standard deviation threshold and the preset height threshold can be set by those skilled in the art according to actual conditions and are not limited here.

[0124] Specifically, when a candidate target whose longitudinal standard deviation is greater than a preset longitudinal standard deviation threshold and whose height is greater than a preset height threshold is selected from the second candidate target as the fifth candidate target, the order of selecting based on these two items is not limited, and there can be two combinations.

[0125] It can be understood that when the longitudinal standard deviation is greater than the preset longitudinal standard deviation threshold, it indicates that the X-distance difference of the target detected by the detection module is abnormal, the position information of the target is inaccurate, and it is useless for making effective AEB decisions. Therefore, the target can be filtered out to improve the decision-making efficiency of subsequent AEB.

[0126] It can also be understood that when the height is greater than the preset height threshold, it indicates that the target is an object in the sky and the probability of collision with the target vehicle is extremely small. Therefore, the target can be filtered out to improve the decision-making efficiency of subsequent AEB.

[0127] S350: Filter out a candidate target whose scattering area satisfies a scattering area threshold greater than a preset scattering area threshold from the fifth candidate targets as a third candidate target.

[0128] Specifically, S350 is similar to S1330 and will not be described in detail here.

[0129] S360: Filter out valid targets from the third candidate targets based on the collision probability.

[0130] Specifically, S360 is similar to S130 and will not be described in detail here.

[0131] The target screening method provided by the embodiment of the present disclosure can further filter out targets with a relatively small possibility of colliding with the target vehicle by setting up a setting to screen out candidate targets that meet at least one of the longitudinal standard deviation being greater than a preset longitudinal standard deviation threshold and the height being greater than a preset height threshold from the second candidate targets, thereby further reducing the cardinality of targets faced in subsequent screening, and almost never missing targets with a high possibility of colliding with the target vehicle due to excessive filtering. This is conducive to further improving the target screening speed and accuracy, and further improving the AEB decision-making efficiency.

[0132] Figure 4 The flowchart of another target screening method provided by the embodiment of the present disclosure is shown. The embodiment of the present disclosure is optimized on the basis of the above embodiment, and the embodiment of the present disclosure can be combined with various optional solutions in one or more of the above embodiments.

[0133] like Figure 4 As shown, the target screening may include the following steps.

[0134] S410: Obtain attribute information of the target.

[0135] Specifically, S410 is similar to S110 and will not be described in detail here.

[0136] S420: Based on the location information, select a target located in a preset area from the targets as a first candidate target.

[0137] Specifically, S420 is similar to S420 and will not be described in detail here.

[0138] S430. Filter out, from the first candidate targets, the first candidate targets that satisfy at least one of the following conditions: existence probability greater than a preset existence probability threshold, obstacle probability greater than a preset obstacle probability threshold, and overlap rate greater than a preset overlap rate probability threshold, as the second candidate targets.

[0139] Specifically, S430 is similar to S1310 and will not be described in detail here.

[0140] S440: Filter out candidate targets whose scattering areas meet a scattering area threshold greater than a preset scattering area threshold from the second candidate targets as third candidate targets.

[0141] S450. Filter out, from the third candidate targets, a candidate target that satisfies at least one of the following conditions: the position information is the same as the verification position information, the speed direction is opposite to the target vehicle, and the lateral standard deviation is greater than a preset lateral standard deviation threshold, as the sixth candidate target.

[0142] Optionally, before S440, it may also include: from the second candidate targets, based on at least one of the position information, size information, motion information and scattering area, filtering out candidate targets of a preset type, and taking the remaining candidate targets as fourth candidate targets; at this time, S440 may include: from the fourth candidate targets, screening out candidate targets that satisfy a scattering area greater than a preset scattering area threshold as third candidate targets.

[0143] Optionally, before S440, it may also include: from the second candidate targets, based on at least one of the position information, size information, motion information and scattering area, filtering out candidate targets belonging to a preset type, and taking the remaining candidate targets as fourth candidate targets; from the fourth candidate targets, screening out candidate targets that satisfy at least one of a longitudinal standard deviation greater than a preset longitudinal standard deviation threshold and a height less than a preset height threshold as fifth candidate targets; at this time, S440 may include: from the fifth candidate target, screening out candidate targets that satisfy a scattering area greater than a preset scattering area threshold as third candidate targets.

[0144] Specifically, the verification position information is position information of the target acquired by the image acquisition device.

[0145] Specifically, an image acquisition device, such as a camera, may also be provided on the target vehicle to detect the position information of targets around the target vehicle.

[0146] It can be understood that, for the same target, when the position information detected by the detection module (i.e., the verification position information) is the same as the position information detected by the image acquisition device, the position information of the target can be considered to be reliable, otherwise the position information of the target may be unreliable. Therefore, the target can be filtered out to improve the decision-making efficiency of subsequent AEB.

[0147] Specifically, the motion information includes speed (vector speed), and a target in a direction opposite to the speed of the target vehicle can be determined based on the speed.

[0148] Specifically, the position information includes a lateral standard deviation threshold. After the detection module of the detection device detects a target in the current cycle and obtains the attribute information of the target, the Y-direction standard deviation, i.e., the lateral standard deviation, can be calculated based on the Y-direction distance difference and the Y-direction distance difference of the target detected in the historical cycle.

[0149] Specifically, the specific value of the preset horizontal standard deviation threshold can be set by those skilled in the art according to actual conditions and is not limited here.

[0150] Specifically, when a candidate target that satisfies at least two of the following conditions, namely, position information being the same as the verification position information, direction being opposite to the target vehicle's speed, and a lateral standard deviation being greater than a preset lateral standard deviation threshold, is screened out from the third candidate target as the sixth candidate target, the order of screening based on each item is not limited, and multiple combinations are possible.

[0151] It is understandable that, usually, the target vehicle and the vehicle traveling in the opposite direction (such as a vehicle) are traveling in different lanes. Therefore, the probability of a collision between the target vehicle and the vehicle traveling in the opposite direction is relatively small. Therefore, the target can be filtered out to improve the decision-making efficiency of subsequent AEB.

[0152] It can be understood that when the lateral standard deviation is greater than the preset lateral standard deviation threshold, it indicates that the Y-distance difference of the target detected by the detection module is abnormal, the position information of the target is inaccurate, and it is useless for making effective AEB decisions. Therefore, the target can be filtered out to improve the decision-making efficiency of subsequent AEB.

[0153] S460: Filter out valid targets from the sixth candidate targets based on the collision probability.

[0154] Specifically, S460 is similar to S130 and will not be described in detail here.

[0155] The target screening method provided by the embodiment of the present disclosure can further filter out targets with a relatively small possibility of colliding with the target vehicle, further reduce the cardinality of targets faced in subsequent screening, and almost never miss targets with a high possibility of colliding with the target vehicle due to excessive filtering, which is conducive to further improving the target screening speed and accuracy, and further improving the AEB decision-making efficiency.

[0156] In another embodiment of the present disclosure, the location information includes motion probability; wherein, before selecting valid targets from the third candidate targets based on the collision probability, the method further includes: determining the motion state of each third candidate target based on the motion probability; selecting a candidate target whose collision probability is greater than a first preset collision probability threshold from the third candidate targets whose motion state is a stationary state as a seventh candidate target; selecting a candidate target whose collision probability is greater than a second preset collision probability threshold from the third candidate targets whose motion state is a moving state as an eighth candidate target; wherein, selecting valid targets from the third candidate targets based on the collision probability includes: selecting a candidate target with the largest collision probability from the seventh candidate target, and / or selecting a candidate target with the largest collision probability from the eighth candidate target as a valid target.

[0157] Specifically, the third candidate target can be a candidate target screened out by the second candidate target after S240 and S250, or a candidate target screened out by S340 and S350, or a candidate target screened out by S440 and S450, which is not limited here and is not limited to this.

[0158] Specifically, when the motion probability of the third candidate target is less than the preset motion probability threshold, its motion state can be determined to be a stationary state; when the motion probability of the third candidate target is greater than or equal to the preset motion probability threshold, its motion state can be determined to be a moving state.

[0159] Specifically, the specific values ​​of the first preset collision probability threshold and the second preset collision probability threshold can be set by those skilled in the art according to actual conditions. The first preset collision probability threshold and the second preset collision probability threshold can be the same or different, and are not limited here. For example, when the focus is on the target whose motion state is in a moving state, the second preset collision probability threshold can be set to be less than the first preset collision probability threshold; when the focus is on the target whose motion state is in a stationary state, the second preset collision probability threshold can be set to be greater than the first preset collision probability threshold; when the focus is on the target whose motion state is in a stationary state and in a moving state, the second preset collision probability threshold can be set to be equal to the first preset collision probability threshold, and this is not limited.

[0160] Optionally, selecting a candidate target whose collision probability is greater than a first preset collision probability threshold from the third candidate target whose motion state is a stationary state as the seventh candidate target includes: selecting a candidate target whose X-direction distance difference is less than a first preset X-direction distance difference threshold and whose collision probability is greater than the first preset collision probability threshold from the third candidate target as the seventh candidate target; and / or;

[0161] The method further comprises: selecting, from the third candidate target whose motion state is a moving state, a candidate target whose collision probability is greater than a second preset collision probability threshold as an eighth candidate target, including: selecting, from the third candidate target, a candidate target whose X-direction distance difference is less than a first preset X-direction distance threshold and whose collision probability is greater than a second preset collision probability threshold as the eighth candidate target.

[0162] Specifically, the first preset X-direction distance difference threshold is smaller than the preset X-direction distance difference threshold, and the specific value thereof can be set by those skilled in the art according to actual conditions and is not limited here.

[0163] In some embodiments, a candidate target with the highest collision probability is screened out from the seventh candidate target as a valid target.

[0164] In other embodiments, the candidate target with the highest collision probability is screened out from the eighth candidate targets as a valid target.

[0165] In some further embodiments, the candidate target with the highest collision probability is screened out from the seventh candidate target as a valid target, and the candidate target with the highest collision probability is screened out from the eighth candidate target as a valid target.

[0166] In some further embodiments, the candidate target with the highest collision probability is screened out from the seventh candidate target and the eighth candidate target as a valid target.

[0167] It can be understood that by setting the target with the highest collision probability from the seventh candidate target and / or the eighth candidate target based on the collision probability as the effective target, the target with the highest probability of collision with the target vehicle can be finally selected, which is conducive to the subsequent AEB to make effective decisions accordingly. In addition, by determining the motion state of the third candidate target, the first preset collision probability threshold and the second preset collision probability threshold can be flexibly set according to the degree of attention paid to the targets in the stationary state and the moving state, so as to select more effective targets of concern.

[0168] Below, the target screening method provided by the embodiment of the present disclosure will be described in detail based on a specific example.

[0169] Figure 5 A flowchart of a target screening process provided by an embodiment of the present disclosure is shown.

[0170] like Figure 5 As shown, the target screening process may specifically include the following steps.

[0171] S510: Obtain attribute information of the target.

[0172] The attribute information includes location information, size information, motion information, existence probability, obstacle probability, overlap rate, scattering area, and collision probability.

[0173] S520: Based on the location information, select a target located in a preset area from the targets as a first candidate target.

[0174] S530. Filter out, from the first candidate targets, the first candidate targets that satisfy the conditions that the existence probability is greater than a preset existence probability threshold, the obstacle probability is greater than a preset obstacle probability threshold, and the overlap rate is greater than a preset overlap rate probability threshold as the second candidate targets.

[0175] S540: Filter out candidate targets of a preset type from the second candidate targets based on at least one of the position information, the size information, the motion information, and the scattering area, and use the remaining candidate targets as fourth candidate targets.

[0176] S550: Determine the motion state of each fourth candidate target based on the motion probability.

[0177] S560, selecting from the fourth candidate targets a candidate target whose longitudinal standard deviation is greater than a preset longitudinal standard deviation threshold and whose height is greater than a preset height threshold as a fifth candidate target;

[0178] S570: Filter out a candidate target whose scattering area satisfies a scattering area threshold greater than a preset scattering area threshold from the fifth candidate targets as a third candidate target.

[0179] S580. Filter out, from the third candidate targets, a candidate target that satisfies the conditions that its position information is the same as the verification position information, its speed direction is opposite to that of the target vehicle, and its lateral standard deviation is greater than a preset lateral standard deviation threshold as the sixth candidate target.

[0180] Specifically, the verification position information is position information of the target acquired by the image acquisition device.

[0181] S590: Filter out valid targets from the sixth candidate targets based on the collision probability.

[0182] The target screening method provided by the embodiment of the present disclosure screens the targets layer by layer to ultimately screen out a smaller number of valid targets, thereby improving the decision-making efficiency of subsequent AEB, and in the process of screening the targets layer by layer, the targets located in the preset area are screened out as the first layer of screening, which can filter out a large number of targets with extremely low probability of collision with the target vehicle, reduce the base number of targets faced by subsequent screening, and improve the screening efficiency, and in the process of screening layer by layer, the targets with the lower probability of collision with the target vehicle and the easier to accurately screen out are preferentially filtered out, thereby avoiding missing targets with a high probability of collision with the target vehicle due to over-screening.

[0183] Figure 6 A schematic structural diagram of a target screening device provided in an embodiment of the present disclosure is shown.

[0184] In some embodiments of the present disclosure, Figure 6 The target screening device 600 shown can be applied to a detection device capable of detecting targets around a vehicle, such as a millimeter wave radar, without specific limitation.

[0185] like Figure 6As shown, the target screening device 600 may include: a first acquisition module 610, which acquires attribute information of the target; wherein the attribute information includes position information and at least one of the following: position information, size information, motion information, existence probability, obstacle probability, overlap rate, scattering area, and collision probability; a first screening module 620, which, based on the position information, screens out targets located in a preset area from the targets as first candidate targets; and a second screening module 630, which, based on the attribute information, screens out candidate targets that meet preset screening conditions from the first candidate targets as valid targets.

[0186] The target screening device of the embodiment of the present disclosure is capable of acquiring attribute information of the target; wherein the attribute information includes at least one of the following: position information, size information, motion information, existence probability, obstacle probability, overlap rate, scattering area, and collision probability; based on the position information, the target located in a preset area is screened out from the targets as the first candidate target; based on the attribute information, the candidate targets that meet the preset screening conditions are screened out from the first candidate targets as valid targets. In this way, the embodiment of the present disclosure can screen the targets based on the attribute information of the targets, thereby screening out valid targets, so that the attribute information of the valid targets can be subsequently uploaded to the AEB instead of uploading the attribute information of all targets to the AEB, which is beneficial for the AEB to make quick decisions based on the attribute information of the valid targets, thereby shortening the response time of the AEB.

[0187] In another embodiment of the present disclosure, the second screening module 630 may include:

[0188] A first screening submodule is used to screen out, from the first candidate targets, a first candidate target that satisfies at least one of the following conditions: an existence probability greater than a preset existence probability threshold, an obstacle probability greater than a preset obstacle probability threshold, and an overlap rate greater than a preset overlap rate probability threshold, as a second candidate target;

[0189] A second screening submodule is used to screen out a candidate target whose scattering area is greater than a preset scattering area threshold from the second candidate targets as a third candidate target;

[0190] The third screening submodule is used to screen out valid targets from the third candidate targets based on the collision probability.

[0191] In yet another embodiment of the present disclosure, the second screening module 630 may further include:

[0192] a fourth screening submodule, configured to filter out candidate targets of a preset type from the second candidate targets based on at least one of position information, size information, motion information and scattering area, and use the remaining candidate targets as fourth candidate targets before screening out candidate targets whose scattering areas are greater than a preset scattering area threshold from the second candidate targets as the third candidate targets;

[0193] The second screening submodule is specifically used to screen out a candidate target whose scattering area satisfies a scattering area greater than a preset scattering area threshold from the fourth candidate targets as the third candidate target.

[0194] In yet another embodiment of the present disclosure, the preset type includes at least one of a mirror target, a manhole cover, a pedestrian, a bicycle, and a road boundary.

[0195] In another embodiment of the present disclosure, the position information includes a longitudinal standard deviation threshold, and the size information includes a height; the second screening module 630 may further include:

[0196] a fifth screening submodule, configured to screen out candidate targets satisfying at least one of a longitudinal standard deviation greater than a preset longitudinal standard deviation threshold and a height greater than a preset height threshold from the second candidate targets as fifth candidate targets before screening out candidate targets satisfying a scattering area greater than a preset scattering area threshold from the second candidate targets as the third candidate targets;

[0197] The second screening submodule is specifically used to screen out a candidate target whose scattering area satisfies a scattering area greater than a preset scattering area threshold from the fifth candidate targets as the third candidate target.

[0198] In another embodiment of the present disclosure, the position information includes a horizontal standard deviation threshold, and the motion information includes a speed; the second screening module 630 may further include:

[0199] a sixth screening submodule, for screening out, from the second candidate targets, candidate targets satisfying a scattering area greater than a preset scattering area threshold as the third candidate targets, and then screening out, from the third candidate targets, candidate targets satisfying at least one of the following conditions: the position information is the same as the verification position information, the speed direction is opposite to the target vehicle speed, and the lateral standard deviation is greater than a preset lateral standard deviation threshold, as the sixth candidate targets; the verification position information is the position information of the target acquired by the image acquisition device;

[0200] The third screening submodule is specifically used to screen out valid targets from the sixth candidate targets based on the collision probability.

[0201] In another embodiment of the present disclosure, the position information includes motion probability; the second screening module 630 may further include:

[0202] A first determination submodule is used to determine the motion state of each third candidate target based on the motion probability before screening out the valid target from the third candidate targets based on the collision probability;

[0203] a seventh screening submodule, configured to screen out, from the third candidate targets whose motion state is a stationary state, a candidate target whose collision probability is greater than a first preset collision probability threshold value as a seventh candidate target;

[0204] an eighth screening submodule, configured to screen out, from the third candidate targets whose motion state is a moving state, candidate targets whose collision probability is greater than a second preset collision probability threshold value as eighth candidate targets;

[0205] The third screening submodule is specifically used to screen out a candidate target with the highest collision probability from the seventh candidate target, and / or screen out a candidate target with the highest collision probability from the eighth candidate target as a valid target.

[0206] It should be noted that Figure 6 The target screening device 600 shown can perform Figures 1 to 5 The various steps in the method embodiment shown in the figure are implemented Figures 1 to 5 The various processes and effects in the method embodiment shown are not described in detail here.

[0207] Figure 7 A schematic structural diagram of a target screening device provided in an embodiment of the present disclosure is shown.

[0208] like Figure 7 As shown, the target screening device may include a controller 701 and a memory 702 storing computer program instructions.

[0209] Specifically, the controller 701 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0210] The memory 702 may include a large capacity memory for information or instructions. By way of example and not limitation, the memory 702 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a tape, or a universal serial bus (USB) drive or a combination of two or more of these. Where appropriate, the memory 702 may include a removable or non-removable (or fixed) medium. Where appropriate, the memory 702 may be inside or outside the integrated gateway device. In a specific embodiment, the memory 702 is a non-volatile solid-state memory. In a specific embodiment, the memory 702 includes a read-only memory (ROM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (Electrically Erasable Programmable ROM, EEPROM), an electrically rewritable ROM (EAROM) or a flash memory, or a combination of two or more of these.

[0211] The controller 701 reads and executes the computer program instructions stored in the memory 702 to perform the steps of the target screening method provided in the embodiment of the present disclosure.

[0212] In one example, the target screening device may further include a transceiver 703 and a bus 704. Figure 7 As shown, the controller 701, the memory 702 and the transceiver 703 are connected via a bus 704 and communicate with each other.

[0213] The bus 704 includes hardware, software, or both. For example, but not limitation, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industrial Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a Memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses or a combination of two or more of these. Where appropriate, the bus 704 may include one or more buses. Although embodiments of the present application describe and illustrate a particular bus, the present application contemplates any suitable bus or interconnect.

[0214] The embodiments of the present disclosure also provide a computer-readable storage medium, which may store a computer program. When the computer program is executed by a processor, the processor implements the target screening method provided by the embodiments of the present disclosure.

[0215] The present disclosure also provides a vehicle, including at least one of the following:

[0216] The target screening device provided by the embodiment of the present disclosure;

[0217] The target screening device provided by the embodiment of the present disclosure;

[0218] The computer-readable storage medium provided by the embodiments of the present disclosure.

[0219] The above-mentioned storage medium may, for example, include a memory 702 of computer program instructions, and the above-mentioned instructions may be executed by the processor 701 of the target screening device to complete the target screening method provided by the embodiment of the present disclosure. Optionally, the storage medium may be a non-temporary computer-readable storage medium, for example, the non-temporary computer-readable storage medium may be a ROM, a random access memory (Random Access Memory, RAM), a compact disc read-only memory (Compact DiscROM, CD-ROM), a magnetic tape, a floppy disk, and an optical data storage device, etc.

[0220] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the term "comprising" is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0221] The above description is only a specific embodiment of the present disclosure, so that those skilled in the art can understand or implement the present disclosure. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present disclosure. Therefore, the present disclosure will not be limited to the embodiments described herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A target screening method, characterized in that: include: Acquire attribute information of the target; wherein the attribute information includes position information and at least one of the following: size information, motion information, existence probability, obstacle probability, overlap rate, scattering area, and collision probability; Based on the location information, select a target located in a preset area from the targets as a first candidate target; Based on the attribute information, select candidate targets that meet preset screening conditions from the first candidate targets as valid targets; The step of screening out candidate targets that meet a preset screening condition from the first candidate targets based on the attribute information as valid targets includes: From the first candidate targets, a first candidate target that satisfies at least one of the following conditions: an existence probability greater than a preset existence probability threshold, an obstacle probability greater than a preset obstacle probability threshold, and an overlap rate greater than a preset overlap rate probability threshold is selected as a second candidate target; Selecting, from the second candidate targets, a candidate target whose scattering area is greater than a preset scattering area threshold as a third candidate target; Based on the collision probability, the valid target is screened out from the third candidate targets.

2. The method according to claim 1, characterized in that Before selecting a candidate target having a scattering area greater than a preset scattering area threshold from the second candidate targets as a third candidate target, the method further includes: From the second candidate targets, based on at least one of the position information, the size information, the motion information, and the scattering area, filter out candidate targets of a preset type, and use the remaining candidate targets as fourth candidate targets; The step of selecting a candidate target whose scattering area is greater than a preset scattering area threshold from the second candidate targets as a third candidate target includes: From the fourth candidate targets, a candidate target satisfying a scattering area greater than a preset scattering area threshold is screened out as the third candidate target.

3. The method according to claim 2, wherein the preset type comprises: Mirror at least one of a target, a manhole cover, a pedestrian, a bicycle, and a road boundary.

4. The method according to claim 1, characterized in that: The position information includes a longitudinal standard deviation threshold, and the size information includes a height; Before selecting a candidate target having a scattering area greater than a preset scattering area threshold from the second candidate targets as a third candidate target, the method further includes: From the second candidate targets, select a candidate target that satisfies at least one of a longitudinal standard deviation greater than a preset longitudinal standard deviation threshold and a height greater than a preset height threshold as a fifth candidate target; The step of selecting a candidate target whose scattering area is greater than a preset scattering area threshold from the second candidate targets as a third candidate target includes: From the fifth candidate targets, a candidate target satisfying the requirement that the scattering area is greater than a preset scattering area threshold is selected as the third candidate target.

5. The method according to claim 1, characterized in that The position information includes a lateral standard deviation threshold, and the motion information includes a speed; Wherein, after selecting a candidate target whose scattering area is greater than a preset scattering area threshold from the second candidate targets as a third candidate target, the method further includes: From the third candidate targets, a candidate target that satisfies at least one of the following conditions: the position information is the same as the verification position information, the speed direction is opposite to the target vehicle, and the lateral standard deviation is greater than a preset lateral standard deviation threshold is selected as a sixth candidate target; the verification position information is the position information of the target acquired by the image acquisition device; The step of selecting the valid target from the third candidate targets based on the collision probability includes: Based on the collision probability, the valid target is screened out from the sixth candidate targets.

6. The method according to any one of claims 1 to 5, characterized in that: The position information includes motion probability; Wherein, before selecting the valid target from the third candidate targets based on the collision probability, the method further includes: Based on the motion probability, determining the motion state of each of the third candidate targets; From the third candidate targets whose motion state is a stationary state, select candidate targets whose collision probability is greater than a first preset collision probability threshold as seventh candidate targets; From the third candidate targets whose motion state is a moving state, select candidate targets whose collision probability is greater than a second preset collision probability threshold as eighth candidate targets; The step of selecting the valid target from the third candidate targets based on the collision probability includes: A candidate target with the highest collision probability is selected from the seventh candidate target, and / or a candidate target with the highest collision probability is selected from the eighth candidate target as the valid target.

7. A target screening device, characterized in that: include: A first acquisition module, acquiring attribute information of a target; wherein the attribute information includes position information and at least one of the following: size information, motion information, existence probability, obstacle probability, overlap rate, scattering area, and collision probability; A first screening module, based on the location information, screening out targets located in a preset area from the targets as first candidate targets; A second screening module, based on the attribute information, screens out candidate targets that meet preset screening conditions from the first candidate targets as valid targets; The second screening module is specifically used to screen out, from the first candidate targets, the first candidate targets that satisfy at least one of the following conditions: the existence probability is greater than a preset existence probability threshold, the obstacle probability is greater than a preset obstacle probability threshold, and the overlap rate is greater than a preset overlap rate probability threshold, as the second candidate targets; Selecting, from the second candidate targets, a candidate target whose scattering area is greater than a preset scattering area threshold as a third candidate target; Based on the collision probability, the valid target is screened out from the third candidate targets.

8. A target screening device, characterized in that: include: processor; A memory for storing executable instructions; Wherein, the processor is used to read the executable instructions from the memory and execute the executable instructions to implement the target screening method described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the processor implements the target screening method according to any one of claims 1 to 6.

10. A vehicle, characterized in that: Include at least one of the following: The target screening device as described in claim 7; The target screening device as described in claim 8 above; The computer readable storage medium of claim 9.

Images