Method, electronic device and computer program product for detecting mirrored object

Abstract

The embodiment of the invention relates to a method for detecting a mirror image object, electronic equipment and a computer program product. The method includes acquiring object points in a sensing space by transmitting a beam from a sensor on an own vehicle, the object points including static object points and dynamic object points. The method further includes, for each of the transmitted beams, selecting candidate mirrored object points of the dynamic object points based on a comparison of a minimum radial distance of the static object points and a radial distance of the dynamic object points. The method further includes determining whether the object is a mirrored object based on candidate mirrored object points associated with the object and corresponding mirrored thresholds. In this way, the mirror image object in the driving scene can be accurately and stably identified, and therefore filtering of the mirror image object is promoted to reduce the risk of misjudgment.

Description

Technical Field

[0001] Embodiments of the present disclosure relate generally to the field of driving, and in particular to methods, electronic devices, and computer program products for detecting mirrored objects. Background Art

[0002] One of the keys to achieving an accurate understanding of complex driving environments is powerful sensor sensing capabilities, which is a key link in achieving reasonable planning and effective control, and is also an important foundation for improving driving safety and reducing accident risks.

[0003] Depending on the sensing needs, a vehicle can be equipped with a variety of sensors. With the continuous evolution and upgrading of technology, radar sensors such as millimeter wave radars have shown improved performance in terms of sensing distance, speed, horizontal or vertical field of view, etc. Summary of the invention

[0004] Embodiments of the present disclosure provide a method, an electronic device, and a computer program product for detecting a mirrored object.

[0005] According to a first aspect of the present disclosure, a method for detecting a mirror object is provided. The method includes acquiring object points in a sensing space by emitting a beam from a sensor on a self-vehicle, wherein the object points include static object points and dynamic object points. The method also includes selecting a candidate mirror object point among the dynamic object points based on a comparison of a minimum radial distance of a static object point and a radial distance of a dynamic object point for each of the emitted beams. The method also includes determining whether the object is a mirror object based on the candidate mirror object point associated with the object and a corresponding mirror threshold.

[0006] According to a second aspect of the present disclosure, an electronic device is provided. The electronic device includes at least one processor and a memory coupled to the at least one processor. The memory includes instructions stored therein, which, when executed by the at least one processor, cause the electronic device to perform the steps of the method in the first aspect of the present disclosure.

[0007] According to a third aspect of the present disclosure, a vehicle is provided, comprising the electronic device according to the second aspect of the present disclosure.

[0008] According to a fourth aspect of the present disclosure, a computer program product is provided, which is tangibly stored on a computer-readable medium and includes computer-executable instructions, which, when executed by a processor of a computer, cause the computer to perform the steps of the method in the first aspect of the present disclosure.

[0009] According to a fifth aspect of the present disclosure, a machine-readable storage medium is provided, on which instructions are stored, which, when executed by a processor, enable a machine to execute the steps of the method in the first aspect of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] By describing the exemplary embodiments of the present disclosure in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present disclosure will become more clearly understood. In the exemplary embodiments of the present disclosure, the same or similar reference numerals generally represent the same or similar parts, components, etc.

[0011] Figure 1 A schematic diagram illustrating an exemplary environment in which the method and / or apparatus according to an embodiment of the present disclosure may be implemented;

[0012] Figure 2 A flow chart of a method for detecting a mirrored object according to an embodiment of the present disclosure is illustrated;

[0013] Figure 3 A diagram illustrating a position sensing process according to an embodiment of the present disclosure is illustrated;

[0014] Figure 4 A diagram illustrating a mirror computing process according to an embodiment of the present disclosure;

[0015] Figure 5 A diagram illustrating an example of a non-mirror detection area according to an embodiment of the present disclosure;

[0016] Figure 6 A diagram illustrating a mirrored object detection process according to an embodiment of the present disclosure;

[0017] Figure 7 A diagram illustrating a mirrored object verification process for lateral movement according to an embodiment of the present disclosure is illustrated;

[0018] Figure 8 A diagram illustrating a mirrored object verification process for longitudinal movement according to an embodiment of the present disclosure is illustrated;

[0019] Fig. 9 A diagram illustrating an example of an interpolation curve of longitudinal distance and speed according to an embodiment of the present disclosure;

[0020] Fig.10 A schematic block diagram of an example device suitable for implementing embodiments of the present disclosure is shown.

[0021] In the various drawings, the same or corresponding reference numerals represent the same or corresponding parts. DETAILED DESCRIPTION

[0022] 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 set forth herein. On the contrary, these embodiments are provided to provide 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.

[0023] In the description of the embodiments of the present disclosure, the term "including" and its variations 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, unless explicitly indicated to be different.

[0024] As mentioned above, in order to enable the vehicle to accurately understand the complex driving environment (for example, to achieve accurate navigation and safe driving of the vehicle, etc.), the vehicle can be equipped with various sensors to give it sensor sensing capabilities. With the continuous evolution and upgrading of technology, radar sensors such as millimeter wave radars have shown improved performance in terms of sensing distance, speed, horizontal or vertical field of view, etc.

[0025] Users want reliable object information in traffic and driving scenarios. However, even under normal circumstances, due to some technical limitations or bottlenecks (for example, in hardware and perception), mirror ghosts (or, referred to as mirror objects) may appear, which is a huge challenge for radar sensors. Mirror objects are the sensor's erroneous understanding of the scene, which may lead to degradation of planning, control, etc., and even affect safety.

[0026] The mirrored object is not a real object that actually exists in the scene, but an erroneous information similar to a "ghost" or "double image" generated by the sensor perception deviation. For example, there is only one car on the actual road, but due to the unavoidable constraints and restraints of the sensor, it may sense that there is also a "mirror" vehicle opposite to this car. Such a false perception will inevitably interfere with the driver or the driving system (such as an automatic driving (AD) system, an advanced driver assistance system (ADAS)). For example, this false perception is displayed on the dashboard or the assisted driving screen, which will distract the driver from the real road conditions and increase the tension and anxiety during driving. In addition, when judging the road conditions, the driving system may make wrong decisions. By way of example, in order to try to avoid this non-existent "mirror" vehicle, the vehicle is unnecessarily emergency braked, or an unreasonable driving route is planned for the vehicle.

[0027] More specifically, the properties of these ghost positions are almost the same as the real positions. These ghost positions will be clustered together to form ghost objects (i.e., so-called mirror objects). These mirror objects may appear in the lane of the self-driving car, which is likely to cause the false triggering time of the autonomous driving functionality. In order to solve such defects, the challenge lies in the recognition of mirror objects. However, since these mirror objects have reasonable properties and will not be filtered out by the abnormal property filtering process.

[0028] To this end, an embodiment of the present disclosure provides a scheme for detecting mirror objects. The scheme includes acquiring object points in a sensing space by emitting beams from sensors on the vehicle, wherein the object points include static object points and dynamic object points. The scheme also includes selecting candidate mirror object points among dynamic object points based on a comparison of the minimum radial distance of static object points and the radial distance of dynamic object points for each of the emitted beams. The scheme also includes determining whether the object is a mirror object based on the candidate mirror object points associated with the object and the corresponding mirror threshold. In this way, mirror objects in driving scenes can be accurately and stably identified, thereby facilitating filtering thereof to reduce the risk of misjudgment.

[0029] Reference below Figures 1 to 10 It should be understood that these exemplary embodiments are provided only to enable those skilled in the art to better understand and implement the embodiments of the present disclosure, and are not intended to limit the scope of the present disclosure in any way.

[0030] Figure 1 1 is a schematic diagram of an exemplary environment 100 in which the method and / or device according to an embodiment of the present disclosure may be implemented. Figure 1 As shown in FIG. 1 , the environment 100 may include a self-vehicle 110, a target object 120, a mirror object 130, and a guardrail 140, wherein the self-vehicle 110 may be equipped with a sensor such as a radar. It should be understood that for the purpose of easy understanding and convenience of illustration, Figure 1 Only limited entities and examples are shown, for example, guardrail 140 can be an example of a roadblock that has the ability to reflect beams from a sensor, and other similar objects with such reflective properties can also replace guardrail 140 or be an addition to guardrail 140, such as a small pool on the tunnel floor, or other traffic participants.

[0031] exist Figure 1 In the example of FIG. 1 , the vehicle 110 is traveling on a lane, and a sensor such as a radar on the vehicle 110 emits a beam in the driving direction to sense the target object 120. Taking the millimeter wave radar as an example, there are four possible propagation paths. Figure 1As shown in , 1) the sensor's beam goes from the ego vehicle 110 to the target object 120 (e.g., at a direction angle 102) and back to the ego vehicle 110; 2) the beam goes from the ego vehicle 110 to the guardrail 140 (e.g., at a reflection angle 104) and then reflects to the target object 120, then the beam reflects back to the guardrail 140 and back to the ego vehicle 110; 3) there may be a beam from the ego vehicle 110 to the guardrail 140 (e.g., at a reflection angle 104) and then reflects to the target object 120, then the beam reflects back to the ego vehicle 110; and 4) there may be a beam from the ego vehicle 110 to the target object 120 (e.g., at a direction angle 102), then the beam reflects to the guardrail 140 and back to the ego vehicle 110. Paths 3) and 4) are also referred to as multipath.

[0032] In the case of paths 3) and 4), due to the possible presence of large power, some ghost positions (e.g., ghost points) with wrong positions and angles will be generated. The properties of these ghost positions are almost the same as the real positions, such as radar cross-sectional area (RCS), azimuth, etc. Of course, these ghost positions will be clustered together to form ghost objects, the so-called mirror objects. These mirror objects may appear in the lane of the vehicle, which is likely to cause false triggering time of, for example, adaptive cruise control (ACC), automatic emergency braking (AEB), etc. In the following, we will combine Figure 2 A mirror object detection method according to an embodiment of the present disclosure is described for accurately identifying a mirror object in a driving scene.

[0033] Figure 2 The flowchart of the method for detecting mirrored objects according to an embodiment of the present disclosure is illustrated. At block 210, an object point in a sensing space is acquired by emitting a beam from a sensor on the ego vehicle, such object points including static object points and dynamic object points. The ego vehicle (e.g., Figure 1 A sensor such as a radar on the self-vehicle 110 in the vehicle 110 transmits a beam in the driving direction to obtain object information in the sensing space. In the sensing space, there are static object points associated with static objects, and dynamic object points associated with dynamic objects. By way of example, a cluster of associated static object points may indicate the static object, and a cluster of associated dynamic object points may indicate the dynamic object, wherein the static object may be a stationary object in the driving scene (e.g., a roadblock, a road sign, etc.), and as Figure 1 The dynamic object of the target object 120 may be a dynamic object in the driving scene (eg, a moving vehicle, a pedestrian, etc.).

[0034] At block 220, for each of the transmitted beams, a candidate mirror object point among the dynamic object points is selected based on a comparison of the minimum radial distance of the static object point and the radial distance of the dynamic object point. On each beam, the static object point with the smallest radial distance from the ego vehicle among the static object points is identified (i.e., the minimum radial distance of the static object point is identified), and is compared with the radial distance of the dynamic object point from the ego vehicle to select or identify a candidate or potential mirror object point among the dynamic object points.

[0035] On the propagation path of the corresponding radar beam, if it reaches a static object, it should be reflected back. However, if other objects can be sensed behind the static object (that is, in the sensing result, there are other objects behind the static object), this situation is obviously abnormal, and it is reasonable to suspect that the additionally sensed object behind the static object is a mirror object that may be misdetected due to reflection. In other words, the radial distance of the real object is smaller than that of the mirror object. At the level of position and point, in some embodiments, for each beam in the transmitted beam, a dynamic object point whose radial distance is greater than the minimum radial distance of the static object point can be identified as a candidate mirror object point.

[0036] At box 230, based on the candidate mirror object points associated with the object and the corresponding mirror threshold, it is determined whether the object is a mirror object. A cluster or set including multiple object points indicates a corresponding object. Among the multiple object points associated with the object, there may be a candidate mirror object point identified at 220. Based on the number of candidate mirror object points in the multiple object points associated with an object or the ratio of the candidate mirror object points to all object points, and the corresponding mirror threshold, it is determined whether the object is a mirror object. This process can be for a specific object or can be object-by-object until all dynamic objects in the driving scene are determined. In the following, the mirror object determination according to an embodiment of the present disclosure will be further described in detail.

[0037] By means of the method 200 for detecting mirror objects according to an embodiment of the present disclosure, mirror objects in a driving scene can be effectively identified in a highly accurate and robust manner, thereby filtering mirror objects to reduce the risk of misjudgment and improve driving reliability and safety.

[0038] Next, we will combine Figure 3 A position sensing process according to an embodiment of the present disclosure is described for accurately identifying candidate or potential mirror object points at the object point level. Figure 3A diagram of a position sensing process 300 according to an embodiment of the present disclosure is illustrated. Through the position sensing process 300, a radar sensor such as a millimeter wave radar can be used to establish a sensing space for a driving environment, and object points associated with various objects (including static objects and dynamic objects) are identified in the space. The position sensing process 300 is also referred to as a freespace process. At 302, the position sensing process 300 is started. At 304, beams are created, such as coarse beams and thin beams within a field of view (FOV).

[0039] According to an embodiment of the present disclosure, the beam emitted from the sensor on the vehicle may include a coarse beam (also referred to as the first beam in this article), such a coarse beam is arranged to be spaced apart at a first angle within the sensing angle range of the sensor. The beam emitted from the sensor on the vehicle may also include a fine beam (also referred to as the second beam in this article), such a fine beam is arranged to be spaced apart at a second angle within the sensing angle range, and the second angle is smaller than the first angle. In a non-limiting example, the sensing angle range may be from -60 degrees to 60 degrees, the first angle may be 5 degrees, and the second angle may be 2 degrees. Distinguishing between coarse beams and fine beams and spacing them at different angles is to enhance the credibility of free space. Because in different divisions of the beam, there may be some dynamic points after the static points.

[0040] At 306, determine whether all dynamic object points have been traversed. If "yes", proceed to 318 to end the position sensing process 300; if "no", proceed to 308 to initialize the free space flag of the object point to be judged and identified to "false". At 310, inappropriate or unsuitable object points can be filtered. After the object points in the sensing space are acquired, one or more filtering processes in the filtering process according to the embodiment of the present disclosure can be used to filter the object points. For example, after the object points in the sensing space are acquired, object points that exceed the sensing angle range of the sensor can be filtered, that is, object points that exceed the FOV can be filtered, such as positions from -60 degrees to 60 degrees.

[0041] In addition, object points whose radial distance is less than a distance threshold can be filtered. For example, positions whose radial distance is less than 5 meters can be filtered, because there are always some static ghost positions within 5 meters. Object points whose radar cross section (RCS) is less than a first threshold or whose alpha (α) quality value is less than a second threshold can also be filtered. For example, positions whose RCS is lower than -15 or whose alpha quality value is lower than 0.2 can be filtered, because these positions can usually be regarded as ghosts.

[0042] At 312, the minimum radial distance of the static object point for each beam is recorded. At 314, for each beam, a free space flag is set by comparing the radial distance of the dynamic object point with the corresponding minimum radial distance. For example, for each beam, if the radial distance of the dynamic object point is greater than the corresponding minimum radial distance (indicating that the dynamic position is behind the static position), it is identified as a candidate mirror object point and the free space flag of the dynamic object point is set to "true".

[0043] At 316, the above process is continued for the next dynamic object point until all dynamic object points of each beam are iterated. After the traversal of all dynamic object points of each beam is completed, the process proceeds to 318 to end the position sensing process 300. It should be understood that the above steps are only exemplary and not limiting, and one or more steps may be added or reduced, and their order may be exchanged.

[0044] According to an embodiment of the present disclosure, dynamic object points associated with an object may be determined. A ratio of candidate mirror object points among the associated dynamic object points is determined for the object, and based on a comparison between the determined ratio and a mirror threshold corresponding to the ratio, it is determined whether the object is a mirror object. Figure 4 A mirror calculation process according to an embodiment of the present disclosure is described to determine the probability or possibility that an object is a mirror object at the object level.

[0045] Figure 4 A diagram of a mirror calculation process 400 according to an embodiment of the present disclosure is illustrated. After all candidate mirror object points in the sensing space are identified (i.e., the radial distances are greater than the minimum radial distance of the static object points on the corresponding beams), the mirror calculation process is started at 402. At 404, for the object to be determined currently (also referred to as the current object), the dynamic object points associated with the object are determined.

[0046] At 406, determine whether all dynamic objects have been traversed. If yes, proceed to 416 to end the mirror calculation process 400; if no, proceed to 408 to retrieve the associated dynamic object points of the current object. At 410, the associated dynamic object points of the current object (the number of which is n) are retrieved. a ), the points that have been identified as candidate mirror object points in the position sensing process 300 are counted, that is, n ft At 412, the proportion of candidate mirror object points among the associated dynamic object points of the current object (ie, n ft / n a ), the proportion can be used to indicate the mirror probability that the object is a mirror object.

[0047] At 414, the above process may continue for the next object until all dynamic objects in the sensing space have been iterated. After the traversal of the dynamic objects in the space is completed, the image calculation process 400 is terminated by proceeding to 416. It should be understood that the above steps are only exemplary and not restrictive, and one or more steps may be added or reduced, and their order may be exchanged.

[0048] According to an embodiment of the present disclosure, the mirror probability p of the object to be determined is a mirror object. mirror The number of candidate mirror object points (i.e., free space locations) identified among the associated dynamic object points of the object can be calculated by ft The total number of dynamic object points associated with this a It is determined by the quotient of , as shown in the following formula (1).

[0049]

[0050] In the context of equation (1), the mirror used for positional reflection can be simplified to be described by a static position, while the position on the beam that is located after the static position can be understood as a candidate or potential mirror object position.

[0051] Figure 5 A diagram of an example 500 of a non-mirror detection area according to an embodiment of the present disclosure is illustrated. According to an embodiment of the present disclosure, an area where detection of a mirror object is not performed (also referred to as a non-mirror detection area in this document) can be identified, the area including a plurality of sub-areas, the plurality of sub-areas being continuous in the longitudinal direction, and the plurality of sub-areas having narrower sub-areas close to the vehicle in the transverse direction and wider sub-areas far from the vehicle in the transverse direction.

[0052] Sensors such as radar have improved their performance with technology upgrades, such as in azimuth separability. However, in some cases, such as oncoming parallel objects close to each other in the lane of the vehicle, the radar sensor may misidentify these objects as mirror objects to some extent. Therefore, it is necessary to create non-mirror detection areas. Figure 5 As shown in the example, assuming that each sub-area in the area is continuous in the longitudinal direction and has the same size, the sub-area closest to the vehicle can be 3.6 meters long in the transverse direction, which is relatively narrow. As the distance from the vehicle increases, the size of these sub-areas in the transverse direction gradually increases, becoming wider and wider. This is because the possibility of the above-mentioned misidentification by the radar sensor on the vehicle increases as the distance increases from near to far.

[0053] According to an embodiment of the present disclosure, if the count of an object being determined as not being a mirrored object is greater than a predetermined count threshold, the object can be determined as a non-mirror object. Objects located within the above-defined area (i.e., the non-mirror detection area), or the count of the non-mirror counter is continuously greater than a threshold (such as 15, 25, etc.), can be regarded as reliable or non-mirror objects. For example, if an object is determined to be not a mirrored object in a detection cycle, the count of the non-mirror counter is increased by 1. If the count is greater than the predetermined count threshold after several detection cycles, the object is regarded as a non-mirror object.

[0054] As described above, after determining the proportion of candidate mirror object points among the dynamic object points associated with the object (i.e., the mirror probability of the object being a mirror object), it is possible to determine whether the object is a mirror object based on a comparison between the determined proportion and a mirror threshold corresponding to the proportion. How to determine the mirror probability threshold will be discussed below.

[0055] According to an embodiment of the present disclosure, the mirror threshold may include a lateral mirror threshold and a non-lateral mirror threshold to distinguish and consider the moving direction of the dynamic object. The lateral mirror threshold and the non-lateral mirror threshold associated with the proportion of the candidate mirror object point may be determined. Table 1 below gives a non-limiting example.

[0056] Table 1

[0057]

[0058] As shown above, the table takes into account the object's motion state or direction, and also the number of candidate mirror object points. For objects that move laterally, the number of reflection positions is obviously greater than for objects that do not move laterally, so the threshold for the mirror probability will be higher. On the other hand, if an object is associated with more mirror positions (e.g., previously identified candidate mirror object points), a higher threshold is required. Obviously, the conditions are strict.

[0059] More specifically, the first column of Table 1 exemplarily shows the number of candidate mirror object points among the associated dynamic object points for a certain object, and the column may also be the ratio of candidate mirror object points among the associated dynamic object points. For different numbers or ratios of candidate mirror object points, customized thresholds based on the above considerations are provided, the second column of Table 1 exemplarily shows the lateral mirror threshold, and the third column of Table 1 exemplarily shows the non-lateral mirror threshold, where "\" indicates that this situation is not applicable.

[0060] According to an embodiment of the present disclosure, based on the moving direction of the object, the proportion of candidate mirror object points (i.e., the mirror probability of the object being a mirror object) is compared with a corresponding one of the associated transverse mirror threshold and the non-transverse mirror threshold. If the moving direction of the object is transverse movement, the mirror probability of the object is compared with the transverse mirror threshold, and if the moving direction of the object is non-transverse movement, the mirror probability of the object is compared with the non-transverse mirror threshold. In the case where the mirror probability of the object is less than a corresponding one of the associated transverse mirror threshold and the non-transverse mirror threshold, the object can be determined to be a mirror object.

[0061] Figure 6 A diagram of a mirror object detection process 600 according to an embodiment of the present disclosure is illustrated. Please note that the "&&" in the figure indicates that the "AND operation" is used to represent "and", and the "||" indicates that the "OR operation" is used to represent "or". At 602, if the object is determined to be a mirror object in the previous detection cycle, and at 604, its mirror probability is greater than the corresponding mirror threshold in the current cycle, then at 610, the object can continue to be determined as a mirror object in the current cycle.

[0062] Alternatively or additionally, according to an embodiment of the present disclosure, if an object is determined to be a mirror object in a previous detection cycle at 602, and the object is not measured in the current cycle at 606 and the number of candidate mirror object points associated with the object is 0 at 608, then the object may continue to be determined to be a mirror object in the current cycle at 610.

[0063] Figure 7 The diagram of the mirror object verification process 700 for lateral movement according to an embodiment of the present disclosure is illustrated. The mirror object verification process 700 is provided to clarify the similarity of the motion state of an object that moves laterally or is laterally away from the vehicle lane. Please note that the "&&" in the figure indicates that the "and operation" is used to represent "and", and the "||" indicates that the "or operation" is used to represent "or". In response to the moving direction of the mirror object being lateral movement, at 702, it is determined whether the lateral distance difference between the mirror object and the associated target object is less than the lateral distance difference threshold. At 704, it is determined whether the lateral speed difference between the mirror object and the associated target object is less than the lateral speed difference threshold. At 706, it is determined whether the longitudinal speed sum of the mirror object and the associated target object (i.e., the sum of the speeds of the two objects in the longitudinal direction) is less than the longitudinal speed sum threshold.

[0064] In addition, at 714, it is determined whether the longitudinal distance difference between the mirrored object and the associated target object is greater than the longitudinal distance and a threshold. For objects moving laterally, the difference in the lateral position and velocity between the objects and the sum of the longitudinal position and velocity are calculated and compared with the corresponding thresholds. The purpose of checking the longitudinal position difference is to ensure that two objects that are close to each other in the longitudinal direction and moving laterally are not identified as mirrored objects.

[0065] If the answers at 702, 704, 706 and 714 are all “yes”, that is, when the lateral distance difference is less than the lateral distance difference threshold, the lateral speed difference is less than the lateral speed difference threshold, the longitudinal speed sum is less than the longitudinal speed sum threshold, and the longitudinal distance difference is greater than the longitudinal distance sum threshold, the mirror object is determined to have passed the mirror object verification, and it is further verified at 716 that it is indeed a mirror object.

[0066] In response to the moving direction of the mirror object being lateral movement, at 708, it is determined whether the lateral position variance of the mirror object is greater than a variance threshold. At 710, it is determined whether the lateral velocity variance of the mirror object is greater than the variance threshold. At 712, it is determined whether the mirror object was determined to have passed the mirror object verification in the previous cycle. For some mirror objects, they may have a large lateral position variance in some cases, which causes the properties of the object to be asymmetric. Therefore, if the object was determined to be a mirror object in the previous cycle, and the variances of its lateral velocity and position are both large in the current cycle, it is identified as a mirror object.

[0067] In addition, at 714, it is determined whether the longitudinal distance difference between the mirror object and the associated target object is greater than the longitudinal distance and the threshold. If 708, 710, 712, and 714 are all "yes", that is, when the lateral position variance is greater than the variance threshold and the lateral speed variance is greater than the variance threshold and the mirror object is determined to have passed the mirror object verification in the previous cycle and the longitudinal distance difference is greater than the longitudinal distance and the threshold, the mirror object is determined to have passed the mirror object verification, and it is further verified at 716 that it is indeed a mirror object.

[0068] Figure 8A diagram of a mirror object verification process 800 for longitudinal movement according to an embodiment of the present disclosure is illustrated. Please note that the "&&" in the figure indicates that the "and operation" is used to represent "and". In response to the moving direction of the mirror object being longitudinal movement, at 802, it is determined whether the longitudinal distance difference between the mirror object and the associated target object is less than the longitudinal distance difference threshold. At 804, it is determined whether the longitudinal speed difference between the mirror object and the associated target object is less than the longitudinal speed difference threshold. At 806, it is determined whether the lateral distance difference between the mirror object and the associated target object is greater than the lateral distance sum threshold. At 808, it is determined whether the lateral speed sum of the mirror object and the associated target object is less than the lateral speed sum threshold.

[0069] For a longitudinal moving object, the difference between the longitudinal position and the velocity, and the sum of the lateral position and the velocity are calculated and compared with the corresponding thresholds. If 802, 804, 806 and 808 are all "yes", that is, when the longitudinal distance difference is less than the longitudinal distance difference threshold and the longitudinal velocity difference is less than the longitudinal velocity difference threshold and the lateral distance difference is greater than the lateral distance sum threshold and the lateral velocity sum is less than the lateral velocity sum threshold, the mirror object is determined to pass the mirror object verification, and it is further verified at 810 that it is indeed a mirror object.

[0070] Fig. 9 9 is a diagram illustrating an example of an interpolated curve of longitudinal distance and speed according to an embodiment of the present disclosure. Fig. 9 As shown in Fig. 9 The left side of shows the distance interpolation curve, Fig. 9 The right side of shows the speed interpolation curve, where the horizontal axis is the object distance and speed, and the vertical axis is the distance threshold and speed threshold. The distance and speed thresholds are calculated by interpolation. According to a large amount of mirror data, the speed and position differences between mirror pairs increase with increasing distance. Therefore, the interpolation method is used to meet such radar characteristics. The lateral situation is similar and will not be repeated here.

[0071] Fig.10A schematic block diagram of an example device 1000 suitable for implementing an embodiment of the present disclosure is shown. The controller in the above text can be implemented using device 1000. As shown in the figure, device 1000 includes a processor 1001, which can be loaded into a computer program instruction in a random access memory (RAM) 1003 according to a computer program instruction stored in a read-only memory (ROM) 1002, to perform various appropriate actions and processes. In RAM 1003, various programs and data required for the operation of device 1000 can also be stored. Processor 1001, ROM 1002 and RAM 1003 are connected to each other via bus 1004. Input / output (I / O) interface 1005 is also connected to bus 1004.

[0072] The various processes and processing described above, such as method 200 and other processes, may be performed by processor 1001. For example, in some embodiments, method 200 and other processes may be implemented as a computer software program, which is tangibly contained in a machine-readable medium. In some embodiments, part or all of the computer program may be loaded and / or installed on device 1000 via ROM 1002. When the computer program is loaded into RAM 1003 and executed by processor 1001, one or more actions of method 200 and other processes described above may be performed. According to an embodiment of the present disclosure, a vehicle is provided, which may include device 1000 as described above for performing various aspects of the present disclosure.

[0073] The present disclosure may be a method, an apparatus, an electronic device, a vehicle, a computer-readable storage medium, and / or a computer program product. The electronic device may be a domain controller or a millimeter wave radar. The computer program product may include a computer-readable storage medium having computer-readable program instructions for executing various aspects of the present disclosure.

[0074] Computer readable storage medium can be a tangible device that can hold and store instructions used by an instruction execution device. Computer readable storage medium can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the above. More specific examples (non-exhaustive list) of computer readable storage medium include: portable computer disk, hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disk read-only memory (CD-ROM), digital versatile disk (DVD), memory stick, floppy disk, mechanical encoding device, such as a punch card or a convex structure in a groove on which instructions are stored, and any suitable combination of the above. The computer readable storage medium used here is not interpreted as a transient signal itself, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagated by a waveguide or other transmission medium (e.g., a light pulse by an optical fiber cable), or an electrical signal transmitted by a wire.

[0075] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0076] The computer program instructions for performing the operation of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. Computer-readable program instructions may be executed completely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be personalized by utilizing the state information of the computer-readable program instructions, and the electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present disclosure.

[0077] Various aspects of the present disclosure are described herein with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer-readable program instructions.

[0078] These computer-readable program instructions can be provided to a processing unit of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processing unit of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0079] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operating steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0080] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of the module, program segment or instruction includes one or more executable instructions for realizing the specified logical function. In some alternative implementations, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or action, or can be implemented with a combination of special hardware and computer instructions.

[0081] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The choice of terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A method (200) for detecting a mirrored object, comprising: Acquire (210) object points in a sensing space by emitting beams from sensors on the vehicle, the object points including static object points and dynamic object points; For each of the beams, selecting (220) a candidate mirror object point among the dynamic object points based on a comparison of a minimum radial distance of the static object point and a radial distance of the dynamic object point; and Based on the candidate mirror object points associated with the object and the corresponding mirror threshold, it is determined (230) whether the object is a mirror object.

2. The method (200) of claim 1, wherein the beam comprises: first beams arranged to be spaced apart at a first angle within a sensing angle range of the sensor; and The second beams are arranged to be spaced apart at a second angle within the sensing angle range, the second angle being smaller than the first angle.

3. The method (200) of claim 1, wherein: After acquiring the object point in the sensing space, the method further includes at least one of the following: filtering object points beyond the sensing angle range of the sensor; Filter object points whose radial distance is less than a distance threshold; or Object points whose radar cross-sectional area RCS is smaller than a first threshold or whose α quality value is smaller than a second threshold are filtered out.

4. The method (200) of claim 1, wherein identifying the candidate mirror object points among the dynamic object points comprises: For each of the beams, a dynamic object point having a radial distance greater than the minimum radial distance is identified as the candidate mirror object point.

5. The method (200) of claim 1, wherein determining whether the object is the mirrored object comprises: determining a dynamic object point associated with the object; Determining, for the object, a proportion of the associated candidate mirror object points among the associated dynamic object points; as well as Based on a comparison between the proportion and a mirror threshold corresponding to the proportion, it is determined whether the object is a mirror object.

6. The method (200) according to claim 5, wherein the mirror threshold comprises a lateral mirror threshold and a non-lateral mirror threshold, and determining whether the object is the mirror object further comprises: Determine a transverse mirror threshold and a non-transverse mirror threshold associated with the proportion; Based on the moving direction of the object, by comparing the proportion with a corresponding one of the associated lateral mirror threshold and non-lateral mirror threshold; as well as In a case where the proportion is less than the corresponding one of the associated transverse mirror threshold and non-transverse mirror threshold, the object is determined to be the mirror object.

7. The method (200) of claim 6, further comprising: If the object was determined as a mirror object in the previous cycle but is not measured in the current cycle, the object continues to be determined as the mirror object.

8. The method (200) of claim 6, further comprising: In response to the moving direction of the mirrored object being lateral movement, determining whether a lateral distance difference between the mirrored object and the associated target object is less than a lateral distance difference threshold, determining whether a lateral velocity difference between the mirrored object and the target object is less than a lateral velocity difference threshold, Determine whether the sum of the longitudinal velocities of the mirrored object and the target object is less than a longitudinal velocity sum threshold, and Determine whether a longitudinal distance difference between the mirrored object and the target object is greater than a longitudinal distance and a threshold; as well as When the lateral distance difference is less than the lateral distance difference threshold, the lateral speed difference is less than the lateral speed difference threshold, the longitudinal speed sum is less than the longitudinal speed sum threshold, and the longitudinal distance difference is greater than the longitudinal distance sum threshold, the mirror object is determined to pass the mirror object verification.

9. The method (200) of claim 6, further comprising: In response to the moving direction of the mirrored object being lateral movement, determining whether the lateral position variance of the mirrored object is greater than a variance threshold, determining whether the lateral velocity variance of the mirror object is greater than the variance threshold, determining whether the mirror object was determined to have passed mirror object verification in a previous cycle, and determining whether a longitudinal distance difference between the mirrored object and an associated target object is greater than the longitudinal distance and a threshold; as well as When the lateral position variance is greater than the variance threshold and the lateral speed variance is greater than the variance threshold and the mirror object is determined to have passed the mirror object verification in the previous cycle and the longitudinal distance difference is greater than the longitudinal distance and the threshold, the mirror object is determined to have passed the mirror object verification.

10. The method (200) of claim 6, further comprising: In response to the moving direction of the mirrored object being longitudinal movement, determining whether a longitudinal distance difference between the mirrored object and the associated target object is less than a longitudinal distance difference threshold, determining whether a longitudinal velocity difference between the mirrored object and the target object is less than a longitudinal velocity difference threshold, determining whether a lateral distance difference between the mirrored object and the target object is greater than a lateral distance and a threshold, and Determining whether a sum of lateral velocities of the mirrored object and the target object is less than a lateral velocity sum threshold; and When the longitudinal distance difference is less than the longitudinal distance difference threshold and the longitudinal speed difference is less than the longitudinal speed difference threshold and the lateral distance difference is greater than the lateral distance and threshold and the lateral distance difference is greater than the lateral distance and threshold, the mirror object is determined to pass the mirror object verification.

11. The method (200) of claim 1, further comprising: An area where detection of a mirrored object is not performed is identified, the area including a plurality of sub-areas, the plurality of sub-areas being continuous in a longitudinal direction, and a sub-area close to the vehicle in a lateral direction being narrower and a sub-area far from the vehicle in a lateral direction being wider.

12. The method (200) of claim 1, further comprising: If the count of an object being determined as not being a mirrored object is greater than a predetermined count threshold, the object is determined as a non-mirrored object.

13. The method of claim 1, wherein: The sensor is a millimeter wave radar.

14. An electronic device (700), comprising: at least one processor (701); as well as A memory (703) coupled to the at least one processor (701) and having instructions stored thereon, which, when executed by the at least one processor (701), cause the electronic device (700) to perform a method according to any one of claims 1-13.

15. The electronic device (700) according to claim 14, wherein the electronic device (700) is a domain controller or a millimeter wave radar.

16. A computer program product stored on a computer readable medium (1102) and comprising computer executable instructions which, when executed by a processor (1101) of a computer, cause the computer to perform the method according to any one of claims 1 to 13.

Images