Method and device for determining ghosting object, equipment, storage medium and program product

Abstract

The embodiment of the invention relates to a ghosting object determination method and device, equipment, a storage medium and a program product. The method comprises the following steps: determining a first point cloud corresponding to a real object and a second point cloud corresponding to a suspected ghosting object from point clouds detected by a laser radar; determining laser detection information corresponding to each point in the first point cloud and the second point cloud, wherein the laser detection information is used for representing the emission direction of target laser light of the detected corresponding point; and determining whether the suspected ghosting object is a ghosting object or not according to the laser detection information corresponding to each point in the first point cloud and the second point cloud. By adopting the method, the accuracy of determining the ghosting object can be improved.

Description

Technical Field

[0001] The disclosed embodiments relate to the field of autonomous driving technology, and in particular, to a method, device, equipment, storage medium, and program product for determining a ghost object. Background Art

[0002] On-board LiDAR is an indispensable sensor device in autonomous driving technology. In order to improve the detection accuracy of LiDAR, higher power, greater gain, and more sensitive LiDAR are usually selected. This will cause the LiDAR to easily produce ghosting due to highly reflective objects (usually highly reflective traffic signs) when detecting obstacles, affecting autonomous driving decisions. Therefore, in the distance detection of LiDAR, it is necessary to accurately determine the ghost objects corresponding to the highly reflective objects to improve the accuracy of distance detection.

[0003] In traditional technology, the position of a highly reflective object is determined based on its reflectivity and reflection distance, and the transmit power of the lidar at that position is adjusted to reduce high-reflective noise and thus determine the ghost object.

[0004] However, there is a problem in the conventional technology that the accuracy of determining ghost objects is low. Summary of the invention

[0005] The embodiments of the present disclosure provide a method, apparatus, device, storage medium and program product for determining a ghost object, which can be used to improve the accuracy of the determined ghost objects.

[0006] In a first aspect, an embodiment of the present disclosure provides a method for determining a ghost object, the method comprising:

[0007] Determine a first point cloud corresponding to a real object and a second point cloud corresponding to a suspected ghost object from the point cloud detected by the laser radar;

[0008] Determine laser detection information corresponding to each point in the first point cloud and the second point cloud, wherein the laser detection information is used to characterize the emission direction of the target laser light that detects the corresponding point;

[0009] Determine whether the suspected ghost object is a ghost object according to the laser detection information corresponding to each point in the first point cloud and the second point cloud.

[0010] In a second aspect, an embodiment of the present disclosure provides a device for determining a ghost object, the device comprising:

[0011] A first determination module is used to determine a first point cloud corresponding to a real object and a second point cloud corresponding to a suspected ghost object from a point cloud detected by the laser radar;

[0012] A second determination module is used to determine laser detection information corresponding to each point in the first point cloud and the second point cloud, wherein the laser detection information is used to characterize the emission direction of the target laser light that detects the corresponding point;

[0013] The third determination module is used to determine whether the suspected ghost object is a ghost object based on the laser detection information corresponding to each point in the first point cloud and the second point cloud.

[0014] In a third aspect, an embodiment of the present disclosure provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in the first aspect when executing the computer program.

[0015] In a fourth aspect, an embodiment of the present disclosure provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method described in the first aspect above.

[0016] In a fifth aspect, an embodiment of the present disclosure provides a computer program product, including a computer program, which implements the method described in the first aspect when executed by a processor.

[0017] The ghost object determination method, apparatus, device, storage medium and program product provided in the embodiments of the present disclosure determine a first point cloud corresponding to a real object and a second point cloud corresponding to a suspected ghost object from the point cloud detected by a laser radar. Since the first point cloud and the second point cloud can accurately represent the real object and the suspected ghost object, the laser detection information corresponding to each point in the first point cloud and the second point cloud for characterizing the emission direction of the target laser light detected at the corresponding point can be accurately determined, so that whether the suspected ghost object is a ghost object can be accurately determined based on the laser detection information corresponding to each point in the first point cloud and the second point cloud, thereby improving the accuracy of the determined ghost objects. In addition, since the process of determining whether the suspected ghost object is a ghost object based on the laser detection information corresponding to each point in the first point cloud and the second point cloud is relatively simple, the efficiency of determining ghost objects can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A diagram of an application environment of a method for determining a ghost object in one embodiment;

[0019] Figure 2 is a flow chart of a method for determining a ghost object in one embodiment;

[0020] Figure 3 is a flow chart of a method for determining a ghost object in another embodiment;

[0021] Figure 4is a flow chart of a method for determining a ghost object in another embodiment;

[0022] Figure 5 is a flow chart of a method for determining a ghost object in another embodiment;

[0023] Figure 6 is a structural block diagram of a device for determining a ghost object in one embodiment;

[0024] Figure 7 is a structural block diagram of a device for determining a ghost object in one embodiment;

[0025] Figure 8 FIG. 4 is a diagram showing the internal structure of an electronic device in one embodiment. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure more clear, the embodiments of the present disclosure are further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the embodiments of the present disclosure and are not used to limit the embodiments of the present disclosure.

[0027] First of all, before specifically introducing the technical solutions of the embodiments of the present disclosure, the technical background or technical evolution context on which the embodiments of the present disclosure are based is introduced. Generally, in the field of autonomous driving, the current technical background is: when using a laser radar to detect an object, the object will reflect the laser light onto an object with high reflectivity, thereby forming a ghost object, which affects the decision-making of autonomous driving. Based on this background, the applicant has discovered through long-term model simulation research and development and the collection, demonstration and verification of experimental data that the point cloud of the ghost object is located on the extension line of the laser light emission line where the point cloud of the real object is located, resulting in a partial overlap between the point cloud of the ghost object and the point cloud of the real object. How to accurately determine the point cloud of the ghost object and filter out the point cloud corresponding to the ghost object has become a difficult problem that needs to be solved urgently. In addition, it should be noted that the applicant has put in a lot of creative work from the point cloud of the ghost object being on the extension line of the laser light emission line where the point cloud of the real object is located and the technical solutions introduced in the following embodiments.

[0028] The technical solutions involved in the embodiments of the present disclosure are introduced below in combination with the scenarios in which the embodiments of the present disclosure are applied. The terms "first", "second", etc. involved in this application are only used to distinguish similar objects and do not represent a specific ordering of the objects. The "multiple" involved in this application means greater than or equal to two.

[0029] The on-board laser radar not only needs to be able to detect at a long distance to reserve more reaction time for the driver, but also needs to have better accuracy in close-range detection. Therefore, in order to improve the detection accuracy of the laser radar, a laser radar with higher power, greater gain and higher sensitivity is usually selected. This will cause the laser radar to reflect the laser light onto other objects when detecting real objects due to diffuse reflection of real objects, such as high-reflective traffic signs. The reflectivity of the surface of high-reflective traffic signs is high. Even if the received light is very weak, it can produce strong reflected light, resulting in the laser radar receiving the light reflected by the high-reflective traffic signs and forming point cloud data based on the light reflected by the high-reflective traffic signs. According to the point cloud data, a false image corresponding to the high-reflective traffic sign is formed, which is called a ghost object. In the field of autonomous driving, it is necessary to judge the distance between the vehicle and the obstacle based on the information detected by the laser radar in order to make relevant autonomous driving decisions. However, ghost objects will interfere with the judgment of real objects. For example, if the ghost object overlaps with the real object, the distance between the real object and the vehicle cannot be accurately determined, thereby affecting the autonomous driving decision. Therefore, it is necessary to accurately identify ghost objects and filter them out to improve the accuracy of autonomous driving decisions.

[0030] In traditional technology, in order to solve the ghost problem in laser radar detection, one method is to determine the ghost noise based on the reflected light data received by the laser radar through a deep learning method. However, this method requires collecting a large amount of data to train the model, which will cause the problem of high difficulty in model training; another method is to determine the position of the highly reflective object based on the reflectivity and reflection distance of the highly reflective object, thereby reducing the emission power of the laser light in this direction to reduce the high-reflective noise. However, this method requires adjusting the firmware of the laser radar, and the operation is not flexible enough. Based on this, the present application provides a method that can improve the accuracy of the determined ghost objects.

[0031] The method for determining a ghost object provided by the embodiment of the present disclosure can be applied to Figure 1 In the application environment shown, the laser radar 102 emits laser light to the real object 104. After the real object receives the laser light, diffuse reflection occurs. One diffuse reflection light is reflected to the laser radar 102. The laser radar 102 obtains the point cloud position corresponding to the real object according to the received reflected light, i.e., 104. Another diffuse reflection light is reflected to the high-reflective object 106. After the high-reflective object 106 receives the diffuse reflection light, it emits reflected light to the laser radar 102. The laser radar 102 forms the point cloud position corresponding to the high-reflective object according to the received reflected light, i.e., 108.

[0032] In one embodiment, Figure 2 As shown, a method for determining a ghost object is provided, and the method is applied to Figure 1The laser radar in the example is used to illustrate, including the following steps:

[0033] S201, determining a first point cloud corresponding to a real object and a second point cloud corresponding to a suspected ghost object from the point cloud detected by the laser radar.

[0034] Among them, the laser radar is a radar system that emits laser light to detect the position, speed and other characteristic quantities of the target. Its working principle is to emit laser light to the target, and then compare the received reflected signal reflected from the target with the transmitted signal, so as to obtain relevant information of the target, such as the target distance, direction, height, speed, attitude, and even shape parameters, so as to detect, track and identify various targets. The laser radar is composed of a laser transmitter, a laser receiver, a processor, etc. The laser transmitter converts electrical pulses into light pulses and emits them. The laser receiver then restores the light pulses reflected from the target into electrical pulses and sends them to the processor for processing. The processor can form a point cloud corresponding to the target based on the received reflected light. In this embodiment, the point cloud corresponding to the real object detected by the laser radar can be determined as the first point cloud, and the point cloud corresponding to the suspected ghost object detected by the laser radar can be determined as the second point cloud.

[0035] It should be noted that the laser radar can obtain the position information, size information, and shape information corresponding to the object from the reflected light, thereby forming a point cloud corresponding to the object. If the laser radar receives multiple different reflected lights, multiple different point clouds can be formed based on each reflected light. If a real object reflects the laser light onto a highly reflective object, a ghost phenomenon will occur, forming a ghost object corresponding to the highly reflective object; if the real object does not reflect the laser light onto the highly reflective object, the point cloud formed by the laser radar may be a point cloud corresponding to other objects. Therefore, it is necessary to determine the first point cloud corresponding to the real object and the second point cloud corresponding to the suspected ghost object from the point cloud detected by the laser radar.

[0036] In this embodiment, the laser radar can emit laser light to a real object, and the real object forms reflected light and reflects to the laser radar. The laser radar can form a point cloud based on the received reflected light. The point cloud formed by the reflected light with the shortest receiving time can be determined as the first point cloud corresponding to the real object, and the point cloud formed by the reflected light corresponding to the real object with the same direction and a receiving time greater than that of the reflected light corresponding to the real object can be determined as the second point cloud of the suspected ghost object.

[0037] S202, determining laser detection information corresponding to each point in the first point cloud and the second point cloud, where the laser detection information is used to characterize the emission direction of the target laser light that detects the corresponding point.

[0038] It should be noted that the multiple laser emitters in the laser radar can emit laser light in different directions, and the corresponding emission angles of each laser emitter are also different. In this embodiment, the laser light emitted by the laser radar to the real object can be determined as the target laser light, and the relevant information of the target laser light can be determined as the laser detection information, wherein the laser detection information can characterize the emission direction of the target laser light of the corresponding point, and optionally, the laser detection information can also characterize the emission power of the target laser light, and can also characterize the distance between the real object and the laser radar.

[0039] In this embodiment, the laser radar can determine the emission angle of the laser light corresponding to each point in the first point cloud according to the reflection angle of the reflected light corresponding to each point in the first point cloud, and determine the distance of each point from the laser radar according to the emission time and reception time of the reflected light corresponding to each point in the first point cloud, thereby determining the laser detection information corresponding to each point in the first point cloud according to the emission angle and distance, and determine the emission angle of the laser light corresponding to each point in the second point cloud according to the reflection angle of the reflected light corresponding to each point, and determine the distance of each point from the laser radar according to the emission time and reception time of the reflected light corresponding to each point in the second point cloud, thereby determining the laser detection information corresponding to each point in the first point cloud according to the emission angle and distance.

[0040] S203, determining whether the suspected ghost object is a ghost object according to the laser detection information corresponding to each point in the first point cloud and the second point cloud.

[0041] The suspected ghost object refers to an object that has an overlapping portion with a real object. The suspected ghost object may be a ghost object or another real object that is blocked by a real object. It is understandable that if the suspected ghost object is a ghost object, the laser detection information of the second point cloud contains the same information as the laser detection information of the first point cloud.

[0042] In this embodiment, the laser detection information of the first point cloud can be compared with the laser detection information of the second point cloud, and whether the suspected ghost object is a ghost object can be determined based on the comparison result. For example, if the information representing the emission direction of the target laser light in the laser detection information of the second point cloud is the same as the laser detection information of the first point cloud, then the suspected ghost object can be determined to be a ghost object.

[0043] In the above-mentioned method for determining ghost objects, a first point cloud corresponding to a real object and a second point cloud corresponding to a suspected ghost object are determined from the point cloud obtained by laser radar detection. Since the first point cloud and the second point cloud can accurately represent the real object and the suspected ghost object, the laser detection information corresponding to each point in the first point cloud and the second point cloud for characterizing the emission direction of the target laser light detected at the corresponding point can be accurately determined. Therefore, based on the laser detection information corresponding to each point in the first point cloud and the second point cloud, it is possible to accurately determine whether the suspected ghost object is a ghost object, thereby improving the accuracy of the determined ghost object. In addition, since the process of determining whether the suspected ghost object is a ghost object based on the laser detection information corresponding to each point in the first point cloud and the second point cloud is relatively simple, the efficiency of determining ghost objects can be improved.

[0044] In the above scenario of determining the laser detection information corresponding to each point in the first point cloud and the second point cloud, the emitter identifier of the laser emitter to which the target laser light belongs and the horizontal position angle of the target laser light can be used as the laser detection information. Figure 3 As shown, the above S202 includes:

[0045] S301, determining the emitter identifier of the laser emitter to which the target laser light corresponding to each point in the first point cloud and the second point cloud belongs.

[0046] It should be noted that the laser radar may include multiple laser emitters, for example, the laser radar may include 64 laser emitters, each laser emitter corresponds to an emission angle, which is different from each other, and each laser emitter corresponds to a unique number, that is, a ring value. In this embodiment, the ring value can be used as the emitter identifier of the laser emitter to which the target laser light belongs. It can be understood that the laser radar can emit laser light to a real object through multiple laser emitters with different emission directions, and each laser light forms a corresponding emission point on the real object, and the laser light corresponding to each emission point is determined as the target laser light.

[0047] In this embodiment, the laser emitter to which the target laser light belongs can be determined according to the emission angle of the target laser light corresponding to the received first point cloud, thereby determining the laser emitter to which the target laser light belongs according to the emitter identification of the laser emitter to which the target laser light corresponds to each point in the first point cloud in a pre-stored emitter identification database; and the laser emitter to which the target laser light belongs can be determined according to the emission angle of the target laser light corresponding to the received second point cloud, thereby determining the laser emitter to which the target laser light belongs according to the emitter identification of the laser emitter to which the target laser light corresponds to each point in the second point cloud in a pre-stored emitter identification database.

[0048] S302, determining the horizontal position angle of the target laser light corresponding to each point in the first point cloud and the second point cloud.

[0049] The horizontal position angle refers to the angle of the horizontal component of the laser light emitted by the laser radar. In this embodiment, the horizontal position angle of the target laser light corresponding to each point in the first point cloud can be determined according to the distance between the laser radar and the real object, and the horizontal position angle of the target laser light corresponding to each point in the second point cloud can be determined according to the distance between the laser radar and the suspected ghost object.

[0050] As an optional implementation, the horizontal position angle of the target laser light corresponding to each point can be determined based on the position coordinates of each point in the first point cloud and the second point cloud in the spatial coordinate system and the position coordinates of the laser radar in the spatial coordinate system.

[0051] Taking one of the points in the point cloud as an example, the horizontal position angle of the target laser light is explained. Assuming that the position coordinates of the laser radar in the spatial coordinate system are (x, y, z), the position coordinates of point A in the first point cloud in the spatial coordinate system are (x1, y1, z1), and the position coordinates of point B in the second point cloud in the spatial coordinate system can be (x2, y2, z2). Then, the horizontal position angle of the target laser light corresponding to the first point cloud can be determined to be h(A)=atan2(y1-y, x1-x), so that the horizontal position angle of the target laser light corresponding to each point in the first point cloud can be obtained according to the calculation formula of the above-mentioned horizontal position angle h(A), and the horizontal position angle of the target laser light corresponding to the second point cloud can be determined to be h(B)=atan2(y2-y, x2-x), so that the horizontal position angle of the target laser light corresponding to each point in the second point cloud can be obtained according to the calculation formula of the above-mentioned horizontal position angle h(B).

[0052] S303: Use the transmitter identifier and the horizontal position angle corresponding to each point in the first point cloud and the second point cloud as the laser detection information corresponding to each point.

[0053] In this embodiment, the emitter identification and horizontal position angle of the target laser light corresponding to each point in the first point cloud can be used as the laser detection information corresponding to each point in the first point cloud, and the emitter identification and horizontal position angle of the target laser light corresponding to each point in the second point cloud can be used as the laser detection information corresponding to each point in the second point cloud.

[0054] Exemplarily, the transmitter identification and the horizontal position angle may be used as two-dimensional coordinates, namely (ring, h), and the laser detection information of each point in the point cloud may be identified by the two-dimensional coordinates (ring, h).

[0055] In this embodiment, by determining the emitter identification of the laser emitter to which the target laser light corresponding to each point in the first point cloud and the second point cloud belongs, and determining the horizontal position angle of the target laser light corresponding to each point in the first point cloud and the second point cloud, the emitter identification and the horizontal position angle corresponding to each point in the first point cloud and the second point cloud can be used as the laser detection information corresponding to each point. Since the emitter identification can uniquely represent each laser emitter, and the horizontal position angle can correspond to the emission angle of the laser light emitted by each laser emitter, the determined laser detection information can accurately represent each point in the first point cloud and the second point cloud, thereby improving the accuracy of determining whether a suspected ghost object is a ghost object based on the laser detection information of each point.

[0056] In the above scenario where the suspected ghost object is determined to be a ghost object based on the laser detection information corresponding to each point in the first point cloud and the second point cloud, whether the suspected ghost object is a ghost object can be determined based on whether there is a ghost point in the second point cloud. Figure 4 As shown, the above S203 includes:

[0057] S401, determining whether there are ghost points in the second point cloud according to laser detection information corresponding to each point in the first point cloud and the second point cloud.

[0058] In this embodiment, the laser detection information corresponding to each point in the first point cloud can be compared with the laser detection information corresponding to each point in the second point cloud, and whether there are ghost points in the second point cloud can be determined based on the comparison results. It should be noted that if there are ghost points in the second point cloud, since the ghost points are formed when the real object reflects the laser light to the high-reflective object, and the high-reflective object reflects it to the laser radar, the ghost points and the real points in the first point cloud are formed by the laser light emitted by the same laser transmitter, that is, the ring values ​​of the two are the same. According to the principle of light reflection, the location of the ghost point must be on the extension line of the reflection line of the real point in the first point cloud, that is, the horizontal position angles of the two are also the same. Therefore, it can be determined whether there are ghost points in the second point cloud based on the laser detection information corresponding to each point in the first point cloud and the second point cloud.

[0059] As an optional implementation, for each point in the second point cloud, it is possible to determine whether there is a target point in the first point cloud corresponding to the point in the second point cloud based on the laser detection information corresponding to the point in the second point cloud. The laser detection information corresponding to the target point is consistent with the laser detection information corresponding to the point. If there is a target point, the point is determined to be a ghost point.

[0060] The target point refers to a point in the first point cloud whose detection information is completely consistent with that of a point in the second point cloud. Taking a point C in the second point cloud as an example, in this embodiment, the laser detection information (ring1, h1) corresponding to point C in the second point cloud can be compared with the laser detection information (ring2, h2) of point D in the first point cloud, and whether there is a target point is determined based on the comparison result. If there is a target point, point C is determined as a ghost point.

[0061] Optionally, if ring2 is the same as ring1, and h2 is the same as h1, then it can be determined that the laser detection information corresponding to point D is consistent with the laser detection information corresponding to point C, that is, there is a target point in the first point cloud, and point C can be determined to be a ghost point; if ring2 is different from ring1, h2 is the same as h1, or ring2 is the same as ring1, h2 is different from h1, or ring2 is different from ring1, and h2 is different from h1, then it can be determined that the laser detection information corresponding to point D is inconsistent with the laser detection information corresponding to point C, that is, point D is not the target point, that is, point C is not a ghost point. Therefore, the laser detection information corresponding to point C can be compared with the laser detection information corresponding to other points in the first point cloud in turn, until it is determined that there is a target point in the first point cloud, and point C can be determined to be a ghost point.

[0062] S402 : When there are ghost points in the second point cloud, determine whether the suspected ghost object is a ghost object according to the ghost points in the second point cloud.

[0063] It is understandable that since the angles of the laser light emitted by each laser emitter of the lidar are different, one of the laser lights may be reflected on a highly reflective object to form a ghost point. Therefore, if there is one or a few ghost points in the suspected ghost object, it cannot be determined that the suspected ghost object is a ghost object. It is necessary to further determine the number and proportion of ghost points in the suspected ghost object, so as to determine whether the suspected ghost object is a ghost object based on the proportion of ghost points.

[0064] As an optional implementation, if the proportion of ghost points in the second point cloud is greater than a preset ratio threshold, the suspected ghost object is determined to be a ghost object.

[0065] The preset ratio threshold refers to the threshold of the ratio of the number of ghost points to all points in the second point cloud of the suspected ghost object. In this embodiment, the ratio of the number of ghost points in the second point cloud to the number of all points in the second point cloud can be compared with the preset ratio threshold. Optionally, if the comparison result is that the ratio is greater than the preset ratio threshold, the suspected ghost object can be determined to be a ghost object; if the comparison result is that the ratio is less than or equal to the preset ratio threshold, it can be determined that the suspected ghost object is not a ghost object.

[0066] In this embodiment, by determining whether there are ghost points in the second point cloud based on the laser detection information corresponding to each point in the first point cloud and the second point cloud, it is possible to determine whether a suspected ghost object is a ghost object based on the ghost points in the second point cloud when there are ghost points in the second point cloud. Since the laser detection information corresponding to each point is determined based on the laser emitter identifier and the horizontal position angle, the accuracy of the determined ghost points can be improved, thereby improving the accuracy of determining whether a suspected ghost object is a ghost object.

[0067] Before determining the first point cloud corresponding to the real object and the second point cloud corresponding to the suspected ghost object from the point cloud detected by the laser radar, it is necessary to first determine the real object and the suspected ghost object based on the point cloud detected by the laser radar. Figure 5 As shown, the above method also includes:

[0068] S501, determining the object closest to the laser radar based on the point cloud detected by the laser radar, and taking the object closest to the laser radar as the real object.

[0069] It should be noted that, according to the principle of light reflection, the distance L between the real object and the laser radar is equal to half of the product of the time difference between the emission time t1 of the laser light and the reception time t2 of the reflected light and the speed of light c, that is, L = [(t1-t2)*c] / 2. Therefore, among the multiple three-dimensional objects generated by the laser radar based on the multiple point clouds obtained by detection, the reflected light corresponding to the real object can reach the laser radar the fastest. Therefore, the point cloud corresponding to the real object is the object closest to the laser radar.

[0070] In this embodiment, the laser radar can obtain a corresponding point cloud based on the received reflected light, and determine the closest object as a real object based on the distance between the point cloud and the laser radar.

[0071] S502, based on the point cloud obtained by laser radar detection, determine the objects located on the extension line of the line connecting the laser radar and the real object, and regard the objects located on the extension line as suspected ghost objects.

[0072] It should be noted that when a highly reflective object receives the reflected light emitted by a real object based on the laser light, it will emit light to the laser radar. Therefore, the time it takes for the reflected light from the highly reflective object to reach the laser radar is greater than the time it takes for the reflected light from the real object to reach the laser radar. The laser radar then obtains the corresponding point cloud based on the received reflected light, and the distance between the suspected ghost object and the laser radar is also greater than the distance between the real object and the laser radar, and the suspected ghost object will be located on the extension line of the line connecting the laser radar and the real object.

[0073] In this embodiment, the laser radar can determine, based on the point cloud obtained by detection, an object located on the extension line of the laser radar and the real object and whose distance from the laser radar is greater than the distance between the real object and the laser radar as a suspected ghost object. Optionally, the suspected ghost object may include one, or the suspected ghost object may include multiple.

[0074] In this embodiment, the object closest to the laser radar is determined based on the point cloud obtained by laser radar detection, and the object closest to the laser radar is taken as the real object. Based on the point cloud obtained by laser radar detection, the object located on the extension line of the line connecting the laser radar and the real object is determined, and the object located on the extension line is taken as a suspected ghost object. The real object and the suspected ghost objects can be accurately determined, thereby improving the accuracy of determining ghost objects using real objects and suspected ghost objects.

[0075] In the case where the suspected ghost object is determined to be a ghost object, in order to further improve the accuracy of the autonomous driving decision, the ghost object can be filtered out. In one embodiment, the method further includes: in the case where the suspected ghost object is determined to be a ghost object, filtering out the ghost object using a preset filtering algorithm.

[0076] The preset filtering algorithm may be a point cloud segmentation method, or the preset filtering algorithm may be a gradient filtering algorithm, which is not limited in this embodiment. It should be noted that the ghost object constructed by the laser radar according to the received second point cloud is a three-dimensional model.

[0077] Optionally, in this embodiment, ghost objects can be directly filtered out using a three-dimensional image filtering algorithm to filter out ghost objects, or three-dimensional ghost objects can be projected onto a two-dimensional plane to generate a corresponding depth map, and the depth map can be filtered to filter out ghost objects.

[0078] In this embodiment, when it is determined that a suspected ghost object is a ghost object, a preset filtering algorithm is used to filter out the ghost object, which can avoid the influence of the ghost object on the autonomous driving decision, thereby improving the accuracy of the autonomous driving decision.

[0079] To facilitate understanding by those skilled in the art, the following is a detailed introduction to the method for determining a ghost object provided by the present application. The method may include:

[0080] S1, based on the point cloud obtained by the laser radar detection, determine the object closest to the laser radar, and take the object closest to the laser radar as the real object.

[0081] S2, based on the point cloud obtained by laser radar detection, determine the objects located on the extension line of the line connecting the laser radar and the real object, and regard the objects located on the extension line as suspected ghost objects.

[0082] S3, determining a first point cloud corresponding to the real object and a second point cloud corresponding to the suspected ghost object from the point cloud detected by the laser radar.

[0083] S4, determining the emitter identifier of the laser emitter to which the target laser light corresponding to each point in the first point cloud and the second point cloud belongs.

[0084] S5, determining the horizontal position angle of the target laser light corresponding to each point according to the position coordinates of each point in the first point cloud and the second point cloud in the spatial coordinate system and the position coordinates of the laser radar in the spatial coordinate system.

[0085] S6, taking the transmitter identification and horizontal position angle corresponding to each point in the first point cloud and the second point cloud as the laser detection information corresponding to each point.

[0086] S7, for each point in the second point cloud, determine whether there is a target point in the first point cloud corresponding to the point in the second point cloud based on the laser detection information corresponding to the point in the second point cloud; if there is a target point, determine that the point is a ghost point, wherein the laser detection information corresponding to the target point is consistent with the laser detection information corresponding to the point in the second point cloud.

[0087] S8, when there are ghost points in the second point cloud, if the proportion of the ghost points in the second point cloud is greater than a preset ratio threshold, determining that the suspected ghost object is a ghost object.

[0088] S9, using a preset filtering algorithm to filter out ghost objects.

[0089] It should be noted that for the descriptions in the above S1-S9, reference may be made to the relevant descriptions in the above embodiments, and the effects are similar, so this embodiment will not be repeated here.

[0090] It should be understood that although Figure 2-5 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 2-5 At least part of the steps may include multiple steps or multiple stages. These steps or stages are not necessarily performed at the same time, but can be performed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed in turn or alternately with other steps or at least part of the steps or stages in other steps.

[0091] In one embodiment, Figure 6 As shown, a device for determining a ghost object is provided, comprising: a first determining module 11, a second determining module 12 and a third determining module 13, wherein:

[0092] A first determination module 11 is used to determine a first point cloud corresponding to a real object and a second point cloud corresponding to a suspected ghost object from a point cloud detected by a laser radar;

[0093] A second determination module 12 is used to determine laser detection information corresponding to each point in the first point cloud and the second point cloud, where the laser detection information is used to characterize the emission direction of the target laser light detected at the corresponding point;

[0094] The third determination module 13 is used to determine whether the suspected ghost object is a ghost object according to the laser detection information corresponding to each point in the first point cloud and the second point cloud.

[0095] The device for determining ghost objects provided in this embodiment can execute the above method embodiments, and its implementation principles and technical effects are similar, which will not be repeated here.

[0096] In one embodiment, Figure 7 As shown, the second determination module 12 includes: a first determination unit 121, a second determination unit 122 and a third determination unit 123, wherein:

[0097] A first determining unit 121 is used to determine the emitter identifier of the laser emitter to which the target laser light corresponding to each point in the first point cloud and the second point cloud belongs;

[0098] A second determining unit 122, used to determine the horizontal position angle of the target laser light corresponding to each point in the first point cloud and the second point cloud;

[0099] The third determining unit 123 is used to use the transmitter identification and the horizontal position angle corresponding to each point in the first point cloud and the second point cloud as the laser detection information corresponding to each point.

[0100] The device for determining ghost objects provided in this embodiment can execute the above method embodiments, and its implementation principles and technical effects are similar, which will not be repeated here.

[0101] In one embodiment, please continue to refer to Figure 7 The above-mentioned second determination unit 122 is specifically used to determine the horizontal position angle of the target laser light corresponding to each point according to the position coordinates of each point in the first point cloud and the second point cloud in the spatial coordinate system and the position coordinates of the laser radar in the spatial coordinate system.

[0102] The device for determining ghost objects provided in this embodiment can execute the above method embodiments, and its implementation principles and technical effects are similar, which will not be repeated here.

[0103] In one embodiment, please continue to refer to Figure 7 The third determination module 13 includes: a fourth determination unit 131 and a fifth determination unit 132, wherein:

[0104] A fourth determining unit 131, configured to determine whether there is a ghost point in the second point cloud according to laser detection information corresponding to each point in the first point cloud and the second point cloud;

[0105] The fifth determining unit 132 is configured to determine whether a suspected ghost object is a ghost object according to the ghost points existing in the second point cloud when there are ghost points in the second point cloud.

[0106] The device for determining ghost objects provided in this embodiment can execute the above method embodiments, and its implementation principles and technical effects are similar, which will not be repeated here.

[0107] In one embodiment, please continue to refer to Figure 7 The fourth determination unit 131 is specifically used for: for each point in the second point cloud, determining whether there is a target point corresponding to the point in the first point cloud according to the laser detection information corresponding to the point, and the laser detection information corresponding to the target point is consistent with the laser detection information corresponding to the point. If there is a target point, the point is determined to be a ghost point.

[0108] The device for determining ghost objects provided in this embodiment can execute the above method embodiments, and its implementation principles and technical effects are similar, which will not be repeated here.

[0109] In one embodiment, please continue to refer to Figure 7 The fifth determination unit 132 is specifically used to determine that the suspected ghost object is a ghost object if the proportion of ghost points in the second point cloud is greater than a preset ratio threshold.

[0110] The device for determining ghost objects provided in this embodiment can execute the above method embodiments, and its implementation principles and technical effects are similar, which will not be repeated here.

[0111] In one embodiment, please continue to refer to Figure 7 The above device further includes: a fourth determination module 14 and a fifth determination module 15, wherein:

[0112] A fourth determination module 14 is used to determine the object closest to the laser radar based on the point cloud detected by the laser radar, and regard the object closest to the laser radar as the real object;

[0113] The fifth determination module 15 is used to determine the objects located on the extension line of the line connecting the laser radar and the real object based on the point cloud detected by the laser radar, and regard the objects located on the extension line as suspected ghost objects.

[0114] The device for determining ghost objects provided in this embodiment can execute the above method embodiments, and its implementation principles and technical effects are similar, which will not be repeated here.

[0115] In one embodiment, please continue to refer to Figure 7 The above device further comprises: a filtering module 16, wherein:

[0116] The filtering module 16 is used to filter out the ghost object by using a preset filtering algorithm when the suspected ghost object is determined to be a ghost object.

[0117] The device for determining ghost objects provided in this embodiment can execute the above method embodiments, and its implementation principles and technical effects are similar, which will not be repeated here.

[0118] For the specific definition of the device for determining ghost objects, please refer to the definition of the method for determining ghost objects in the above text, which will not be repeated here. Each module in the above-mentioned device for determining ghost objects can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the electronic device in the form of hardware, or can be stored in the memory of the electronic device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0119] Figure 8 1 is a block diagram of an electronic device 1300 according to an exemplary embodiment. For example, the electronic device 1300 may be a laser radar or a computer device.

[0120] Reference Figure 8 The electronic device 1300 may include one or more of the following components: a processing component 1302, a memory 1304, a power supply component 1306, a laser emitting component 1308, a laser receiving component 1310, an input / output (I / O) interface 1312, a detection component 1314, a scanning component 1316, a sensor component 1318, and a communication component 1320. The memory stores a computer program or instruction running on the processor.

[0121] The processing component 1302 generally controls the overall operation of the electronic device 1300, such as operations associated with display, data communication, data processing, and recording operations. The processing component 1302 may include one or more processors 1322 to execute instructions to complete all or part of the steps of the above method. In addition, the processing component 1302 may include one or more modules to facilitate the interaction between the processing component 1302 and other components.

[0122] The memory 1304 is configured to store various types of data to support operations on the electronic device 1300. Examples of such data include instructions for any application or method operating on the electronic device 1300, laser light, reflected light, etc. The memory 1304 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0123] The power supply component 1306 provides power to the various components of the electronic device 1300. The power supply component 1306 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the electronic device 1300.

[0124] The laser emitting assembly 1308 is an optical component for emitting laser light, and is usually composed of a laser, a beam expander, a beam shaper, and a modulator. The laser light generated by the laser can be processed by the beam expander and the beam shaper to form a beam with a certain shape and size. The modulator can modulate the laser light to prevent interference between different laser radars.

[0125] The laser receiving assembly 1310 is an optical component for receiving the laser signal reflected from the target object, and is usually composed of an optical filter, a photodetector, and an amplifier. The optical filter can filter out unnecessary light signals so that the photodetector can receive the effective signal reflected from the target object, and the amplifier can amplify the weak signal received by the photodetector to improve the detection accuracy and distance measurement range of the laser radar.

[0126] I / O interface 1312 provides an interface between processing component 1302 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include but are not limited to: a home button, a volume button, a start button, and a lock button.

[0127] The detection component 1314 is a component used to convert the optical signal received by the laser receiving component 1310 into an electrical signal, and is generally composed of a photodetector and an amplifier. The photodetector can convert the received optical signal into a weak electrical signal, and the amplifier can amplify the weak electrical signal to improve the detection accuracy and distance measurement range of the laser radar.

[0128] The scanning component 1316 is an optical component for controlling the scanning direction and angle of the laser light, and is usually composed of a scanning mirror, a stepper motor, an encoder, a controller, etc. The scanning mirror can convert the laser light into a scanning light to scan the target area, the stepper motor can control the rotation angle of the scanning mirror, the encoder can measure the rotation angle of the stepper motor, and the controller can adjust the control of the stepper motor.

[0129] The sensor assembly 1318 includes one or more sensors for providing various aspects of status assessment for the electronic device 1300. For example, the sensor assembly 1318 can detect the open / closed state of the electronic device 1300, the relative positioning of components, such as the display and keypad of the electronic device 1300, and the sensor assembly 1314 can also detect the position change of the electronic device 1300 or a component of the electronic device 1300, the presence or absence of user contact with the electronic device 1300, the orientation or acceleration / deceleration of the electronic device 1300, and the temperature change of the electronic device 1300. The sensor assembly 1318 may include a proximity sensor configured to detect the presence of a nearby object without any physical contact. The sensor assembly 1318 may also include an optical sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 1318 may also include an accelerometer, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0130] The communication component 1320 is configured to facilitate wired or wireless communication between the electronic device 1300 and other devices. The electronic device 1300 can access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 1320 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 1320 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0131] In an exemplary embodiment, the electronic device 1300 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to execute the above-mentioned method for determining ghost objects.

[0132] In an exemplary embodiment, a computer program product is also provided, and when the computer program is executed by a processor, the above method can be implemented. The computer program product includes one or more computer instructions. When these computer instructions are loaded and executed on a computer, part or all of the above method can be implemented in whole or in part according to the process or function described in the embodiment of the present disclosure.

[0133] It should be noted that the solutions described in this specification and the embodiments, if they involve the processing of personal information, will be processed on the premise of having a legal basis (such as obtaining the consent of the subject of personal information, or being necessary for the performance of a contract, etc.), and will only be processed within the scope of regulations or agreements. If a user refuses to process personal information other than the necessary information for basic functions, it will not affect the user's use of basic functions.

[0134] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in the embodiments of the present disclosure can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0135] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0136] The above-described embodiments only express several implementation methods of the embodiments of the present disclosure, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for those of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the embodiments of the present disclosure, and these all belong to the protection scope of the embodiments of the present disclosure. Therefore, the protection scope of the patent of the embodiments of the present disclosure shall be subject to the attached claims.

Claims

1. A method for determining a ghost object, characterized in that: The method comprises: Determine a first point cloud corresponding to a real object and a second point cloud corresponding to a suspected ghost object from the point cloud detected by the laser radar; Determine laser detection information corresponding to each point in the first point cloud and the second point cloud, wherein the laser detection information is used to characterize the emission direction of the target laser light that detects the corresponding point; Determine whether the suspected ghost object is a ghost object according to the laser detection information corresponding to each point in the first point cloud and the second point cloud.

2. The method according to claim 1, characterized in that The determining of laser detection information corresponding to each point in the first point cloud and the second point cloud includes: Determine an emitter identifier of a laser emitter to which a target laser light corresponding to each point in the first point cloud and the second point cloud belongs; Determine the horizontal position angle of the target laser light corresponding to each point in the first point cloud and the second point cloud; The transmitter identifier and the horizontal position angle corresponding to each point in the first point cloud and the second point cloud are used as the laser detection information corresponding to each point.

3. The method according to claim 2, characterized in that The determining the horizontal position angle of the target laser light corresponding to each point in the first point cloud and the second point cloud comprises: According to the position coordinates of each point in the first point cloud and the second point cloud in the spatial coordinate system and the position coordinates of the laser radar in the spatial coordinate system, the horizontal position angle of the target laser light corresponding to each point is determined.

4. The method according to any one of claims 1 to 3, characterized in that: The determining whether the suspected ghost object is a ghost object according to the laser detection information corresponding to each point in the first point cloud and the second point cloud includes: Determining whether there is a ghost point in the second point cloud according to laser detection information corresponding to each point in the first point cloud and the second point cloud; In the case that there are ghost points in the second point cloud, it is determined whether the suspected ghost object is the ghost object according to the ghost points in the second point cloud.

5. The method according to claim 4, characterized in that The determining whether there is a ghost point in the second point cloud according to the laser detection information corresponding to each point in the first point cloud and the second point cloud includes: For each point in the second point cloud, determine whether there is a target point corresponding to the point in the first point cloud based on the laser detection information corresponding to the point. The laser detection information corresponding to the target point is consistent with the laser detection information corresponding to the point. If the target point exists, determine that the point is a ghost point.

6. The method according to claim 4, characterized in that The determining whether the suspected ghost object is the ghost object according to the ghost points existing in the second point cloud includes: If the proportion of ghost points in the second point cloud is greater than a preset ratio threshold, the suspected ghost object is determined to be the ghost object.

7. The method according to any one of claims 1 to 3, characterized in that: Before determining the first point cloud corresponding to the real object and the second point cloud corresponding to the suspected ghost object from the point cloud detected by the laser radar, the method further includes: Determine the object closest to the laser radar based on the point cloud detected by the laser radar, and take the object closest to the laser radar as the real object; Based on the point cloud detected by the laser radar, an object located on an extension line of a line connecting the laser radar and the real object is determined, and the object located on the extension line is taken as the suspected ghost object.

8. The method according to claim 7, characterized in that The method further comprises: When it is determined that the suspected ghost object is the ghost object, the ghost object is filtered out using a preset filtering algorithm.

9. A device for determining ghost objects, applied to laser radar, characterized in that: The device comprises: A first determination module is used to determine a first point cloud corresponding to a real object and a second point cloud corresponding to a suspected ghost object from a point cloud detected by the laser radar; A second determination module is used to determine laser detection information corresponding to each point in the first point cloud and the second point cloud, wherein the laser detection information is used to characterize the emission direction of the target laser light that detects the corresponding point; The third determination module is used to determine whether the suspected ghost object is a ghost object according to the laser detection information corresponding to each point in the first point cloud and the second point cloud.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.

11. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

12. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images