Method for filtering cluster-shaped noise points of automatic driving radar in rainy days and related products

Abstract

The embodiment of the invention relates to the technical field of automatic driving, and provides a method for filtering cluster-shaped noise points of an automatic driving radar in rainy days and related products, and the method comprises the steps: building a preset first number of distance grids with the radar as an original point in a vehicle driving process; obtaining the time sequence model and the self-vehicle speed obtained through construction, and marking the specific area grids according to the time sequence model and the self-vehicle speed; if the filtering grid is marked and occupied, determining that the filtering grid has a real obstacle; if the filtering grids are not marked and occupied, whether obstacles exist in the filtering grids or not is judged; and if the obstacles exist in the filtering grids, determining that the filtering grids have clustered noise points, and filtering the clustered noise points. According to the embodiment of the invention, cluster-shaped noisy points formed by the radar in rainy days can be filtered under the condition that real obstacles are prevented from being filtered, and smooth automatic driving of the vehicle and safety of automatic driving are ensured.

Description

Technical Field

[0001] This application relates to the field of autonomous driving technology, and in particular to a method for filtering cluster noise in rainy weather using autonomous driving radar, a device for filtering cluster noise in rainy weather using autonomous driving radar, a corresponding vehicle, and a corresponding computer-readable storage medium. Background Technology

[0002] The advantage of LiDAR lies in its real-world ranging capabilities, which are independent of the scene. LiDAR is commonly used in autonomous driving to ensure its safety. However, in rainy weather, water droplets on the radar surface can cause ranging problems, resulting in noise circles around the radar, known as cluster noise. Filtering these cluster noises is beneficial for ensuring the safety of autonomous driving. Summary of the Invention

[0003] This application provides a method and related products for filtering cluster noise points of autonomous driving radar in rainy weather. It can filter cluster noise points formed by radar in rainy weather without filtering real obstacles, ensuring smooth autonomous driving and the safety of autonomous driving.

[0004] In one aspect, embodiments of this application provide a method for filtering clustered noise points in rainy weather for autonomous driving radar, the method comprising:

[0005] During vehicle operation, a preset first number of distance grids are established with the radar as the origin; the preset first number of distance grids includes a specific area grid.

[0006] The constructed time series model and vehicle speed are obtained, and the specific region grid is marked according to the time series model and the vehicle speed; the specific region grid includes filter grids, and the number of specific region grids is at least greater than the number of filter grids, wherein the filter grids are a preset second number of grids located around the axle center of the vehicle in the preset first number of distance grids;

[0007] If the filter cell is marked as occupied, it is determined that there is a real obstacle in the filter cell;

[0008] If the filter cell is not marked as occupied, then determine whether there is an obstacle in the filter cell;

[0009] If there are obstacles in the filter grid, it is determined that there are clustered noise points in the filter grid, and the clustered noise points are filtered.

[0010] In another aspect, embodiments of this application provide a filtering device for clustered noise points in rainy weather for autonomous driving radar, the device comprising:

[0011] A grid division module is used to establish a preset first number of distance grids with the radar as the origin during vehicle movement; the preset first number of distance grids includes a specific area grid.

[0012] A grid marking module is used to acquire the constructed time series model and vehicle speed, and to mark the grid of the specific region according to the time series model and the vehicle speed; the grid of the specific region includes filter grids, and the number of the specific region grids is at least greater than the number of filter grids, wherein the filter grids are a preset second number of grids located around the axle center of the vehicle in the preset first number of distance grids;

[0013] An obstacle determination module is used to determine that a real obstacle exists in the filter cell when the filter cell is marked as occupied;

[0014] The cluster noise filtering module is used to determine whether there is an obstacle in the filter cell when the filter cell is not marked as occupied; if there is an obstacle in the filter cell, it is determined that there is cluster noise in the filter cell, and the cluster noise is filtered.

[0015] In another aspect, embodiments of this application also provide a vehicle, including: a processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein the computer program, when executed by the processor, implements the method for filtering rain-related cluster noise points of autonomous driving radar as described in any one of the claims.

[0016] In another aspect, embodiments of this application also provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method for filtering cluster noise points in rainy weather for autonomous driving radar as described in any one of the claims.

[0017] In another aspect, embodiments of this application also provide a computer program product containing instructions that, when run on a computer, cause the computer to perform the filtering method for rain-related cluster noise in autonomous driving radar described in the above aspects.

[0018] The autonomous driving radar rain-induced cluster noise filtering method and related products provided in this application establish a preset first number of distance grids with the radar as the origin during vehicle operation. The constructed distance grids may contain specific region grids. Then, based on the constructed time series model and the vehicle speed, the specific region grids are marked. The specific region grids may contain filter grids, and the number of specific region grids is at least greater than the number of filter grids. The filter grids are mainly a preset second number of grids located around the center of the vehicle axle in the distance grids. In the marked specific region grids, if a filter grid is marked and occupied, it is determined that there is a real obstacle in the filter grid; if a filter grid is not marked and occupied, it can be determined whether there is an obstacle in the filter grid. If there is an obstacle in the filter grid, it indicates that there is cluster noise in the filter grid. At this time, the cluster noise can be filtered. This method can filter the cluster noise formed by the radar in rainy weather without filtering real obstacles, ensuring smooth autonomous driving and the safety of autonomous driving. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating the steps of a method for filtering clustered noise points in rainy weather using autonomous driving radar, as provided in an embodiment of this application.

[0020] Figure 2 This is a schematic diagram of the process of the rain cluster noise filtering method provided in the embodiments of this application;

[0021] Figure 3 This is a structural block diagram of a filtering device for clustered noise points in rainy weather provided in an embodiment of this application;

[0022] Figure 4 This is a structural block diagram of a vehicle provided in an embodiment of this application;

[0023] Figure 5 This is a structural block diagram of a computer-readable storage medium provided in an embodiment of this application. Detailed Implementation

[0024] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0025] Reference Figure 1 The diagram illustrates a flowchart of a method for filtering cluster noise points in rainy weather using autonomous driving radar, as provided in an embodiment of this application. Specifically, the method may include the following steps:

[0026] Step S101: During vehicle operation, establish a preset first number of distance grids with the radar as the origin.

[0027] In rainy weather, radar can experience ranging problems due to water droplets adhering to its surface, resulting in noise circles around the radar, known as cluster noise. Filtering out cluster noise helps ensure the safety of autonomous driving.

[0028] In this embodiment, the radar's detection space can be divided into grids, specifically by establishing a predetermined number of range grids with the radar as the origin. Each range grid is the smallest unit after dividing the detection space into grids, and each range grid corresponds to a specific small area in physical space.

[0029] Specifically, the gridding of the radar detection space can be achieved using the radar's calibrated scanning range. Optionally, distance grids can be constructed based on the radar's calibrated scanning range and a preset grid size. It should be noted that the preset grid size can be set according to actual needs, and this application embodiment does not impose any limitations on it.

[0030] In some embodiments of this application, the radar calibration scanning range can be used to indicate the first scanning distance of the radar in a preset first direction and the second scanning distance of the radar in a preset second direction. At this time, the number of grids in the preset first direction can be calculated using the first scanning distance and the preset grid size, and the number of grids in the preset second direction can be calculated using the second scanning distance and the preset grid size. Then, with the radar as the origin, a preset first number of distance grids is established, wherein the preset first number can be the product of the number of grids in the preset first direction and the number of grids in the preset second direction.

[0031] Optionally, the preset first direction can be the forward direction of the radar; the preset second direction can be the horizontal direction of the radar, that is, the left or right direction of the radar.

[0032] For example, assuming the radar's calibrated scanning range is forward (front_span meters) and to the left and right (right_left_span meters), a total of front_num distance grids can be established with the radar as the origin, extending forward (front_span meters) and to the left and right (right_left_span meters) in total, with a total of right_left_num distance grids. The preset first number can be front_num * right_left_num, meaning a total of front_num * right_left_num distance grids are established. The value of front_num can be calculated based on front_span and the preset grid size, and the value of right_left_num can also be calculated based on right_left_span and the preset grid size; this embodiment does not impose any limitations on this.

[0033] Among them, the first number of distance grids established can include specific area grids. The specific area can refer to the sensitive area. The specific area grids represent the grids that may be marked as occupied, which can usually be the center area of ​​the grid.

[0034] A specific region grid may contain filter grids, which represent the grids where clustered noise may exist. Typically, the number / area of ​​specific region grids will be at least greater than the number / area of ​​filter grids. Filter grids may refer to a predetermined second number of grids located around the center of the vehicle's axle within a predetermined first number of distance grids.

[0035] Optionally, the preset second quantity can be determined based on the radar's height difference range and the preset grid size. For example, the measured cluster noise points are mainly concentrated within a range of z meters, y meters to the left, and x meters to the right from the radar center. Assuming the number of filter grids is filter_num, the value of filter_num can be determined based on xyz and the preset grid size. This application embodiment does not limit this.

[0036] In this embodiment of the application, among the preset first number of distance grids, the remaining grids other than the preset second number of filter grids can be used to obtain time sequence information, that is, to play the role of obtaining time sequence information, specifically to record the historical data of obstacles.

[0037] In a preferred embodiment of this application, index information can be calculated based on the vehicle's direction of travel, and then a time series model can be established.

[0038] Specifically, the step direction of data processing can be determined based on the vehicle's direction of travel, such as forward, backward, left or right turns, thus ensuring that the step direction is consistent with the direction of travel. The starting cell for data processing can also be determined from the remaining cells. Then, according to the starting cell and the step direction, the occupancy status of each cell at different timestamps can be recorded to obtain the index information of each cell. The index information of each cell can then be used to construct a time series model.

[0039] Step S102: Obtain the constructed time series model and vehicle speed, and mark the grid cells in a specific region based on the time series model and vehicle speed.

[0040] In some embodiments of this application, for target obstacles detected by radar, the target obstacles can be tracked by a time series model. The time series model can be used to indicate the index change of the target obstacle in two consecutive frames of data, that is, the position change of the target obstacle in two consecutive frames of data.

[0041] Optionally, when tracking target obstacles, in addition to considering the obstacle's movement speed and convex hull region to predict the occupancy status of the specific area of ​​the obstacle, the vehicle's speed can also be incorporated. That is, a temporal model can be used to track target obstacles, and based on the obstacle's movement speed, convex hull region, and the vehicle's speed, a comprehensive rendering of the specific area of ​​the obstacle can be performed.

[0042] It should be noted that the rendering mentioned in the embodiments of this application refers to the marking at the data level, specifically the marking as occupied, which is used to indicate that a certain area grid belongs to the actual coverage area of ​​the actual obstacle. Being marked as occupied means that it is occupied by the obstacle.

[0043] Step S103: If the filter cell is marked as occupied, it is determined that there is a real obstacle in the filter cell.

[0044] For the marking results of a specific area grid, it can be determined whether the filter grid contained in the specific area grid is marked as occupied. In one case, if the filter grid is marked as occupied, it can be determined that there is a real obstacle in the physical area corresponding to the filter grid, rather than noise. In this case, the real obstacle in the filter grid is not filtered, so that the radar can accurately identify the real obstacle and control the vehicle to avoid the real obstacle in time, thus ensuring the safety of autonomous driving.

[0045] Step S104: If the filter cell is not marked as occupied, determine whether there is an obstacle in the filter cell.

[0046] Step S105: If there are obstacles in the filter grid, it is determined that there are clustered noise points in the filter grid, and the clustered noise points are filtered.

[0047] In another scenario, if a filter grid is not marked as occupied, it means that there are no real obstacles in the physical area corresponding to that filter grid. In this case, it can be further determined whether there are obstacles in the filter grid. If there are obstacles in the filter grid after filtering out the absence of real obstacles, it means that there are clustered noise points in the filter grid. In this case, the clustered noise points can be filtered to prevent the radar from identifying them, thus ensuring the accuracy of radar identification and the safety of autonomous driving.

[0048] In some embodiments of this application, reference is made to Figure 2 The diagram illustrates the process of the rain cluster noise filtering method provided in the embodiments of this application.

[0049] In practical applications, cluster noise can include both high-altitude cluster noise and non-high-altitude cluster noise. High-altitude cluster noise mainly occurs when there are not many water droplets; when the water droplets reach a certain level, they will form cluster noise with an undulating height of z_low on the ground, i.e., non-high-altitude cluster noise. In the embodiments of this application, when filtering cluster noise in rainy weather, such as... Figure 2 As shown, it can be divided into two processing logic links.

[0050] Radar can detect both high-altitude and non-high-altitude obstacles, such as... Figure 2 As shown, for filtering high-altitude cluster noise, this embodiment of the application can use a time-series model to time-track high-altitude obstacles and obtain the vehicle speed comprehensive rendering grid to mark the grid in a specific area; for filtering non-high-altitude cluster noise, this embodiment of the application can use a time-series model to time-track non-high-altitude obstacles and obtain the vehicle speed comprehensive rendering grid to mark the grid in a specific area.

[0051] In some embodiments of this application, regardless of whether it is for high-altitude cluster noise or non-high-altitude cluster noise, the target obstacle can be tracked by a time-series model, and the grid of a specific region can be comprehensively rendered based on the movement speed and convex hull region of the target obstacle, as well as the vehicle speed.

[0052] It should be noted that the rendering mentioned in the embodiments of this application refers to the marking at the data level, specifically the marking as occupied, which is used to indicate that a certain area grid belongs to the actual coverage area of ​​the actual obstacle.

[0053] Specifically, a temporal model can be used to track the target obstacle, which changes position in two consecutive frames of data. The movement speed of the target obstacle can be calculated by combining the time interval between the two consecutive frames, i.e., the two-frame rate interval. Then, based on the movement speed of the target obstacle, the convex hull region, and the vehicle speed, specific regions of the grid can be marked.

[0054] Optionally, the target obstacle can first be filtered based on its movement speed to obtain the filtered target obstacle. Then, the specific area grid where the filtered target obstacle is located can be marked based on the convex hull region of the filtered target obstacle and the vehicle speed.

[0055] The convex hull region refers to the smallest convex polygon that surrounds the obstacle. It is mainly used to determine the actual coverage area of ​​the obstacle. For example, the convex hull region formation process can be represented by generating the smallest convex polygon from the point cloud data or boundary points of the broken obstacle through a convex hull algorithm, such as QuickHull. This application embodiment does not limit this.

[0056] Filtering of target obstacles involves filtering stationary objects and abnormal speeds to eliminate interference from stationary or abnormally fast objects. For example, a preset speed threshold can be set; obstacles below the preset speed threshold are considered stationary objects, such as the ground or fixed facilities, and can be excluded. A maximum speed threshold can be set; obstacles exceeding the threshold may be noise or false detections, and can also be excluded.

[0057] Determining the specific region grid where the filtered target obstacle is located involves determining the physical coordinates of the filtered target obstacle based on its convex hull region. This can be achieved by traversing all physical coordinates within the convex hull and mapping them to their corresponding grid indices to determine the covering grid. Then, based on the vehicle speed and the time interval between two consecutive frames of data, the coordinates of the previously determined filtered target obstacle can be corrected to obtain the specific region grid where the filtered target obstacle is located. Finally, the specific region grid where the filtered target obstacle is located is marked as occupied.

[0058] In practical applications, being marked as occupied means being occupied by an obstacle. Real obstacles usually exist in multiple consecutive frames of data and continuously occupy the corresponding cells; while cluster noise is a temporary false signal. Specifically, if the cell corresponding to it was not marked as occupied in the time interval between two consecutive frames of data in the previous filter_num, and an obstacle is only detected in the current frame, then the currently detected obstacle is determined to be cluster noise, and cluster noise can be filtered.

[0059] For example, such as Figure 2As shown, for filtering high-altitude cluster noise, the front end, i.e., the suspended obstacles detected by radar (i.e., high-altitude obstacles), can be obtained, and specific areas of grid cells can be rendered according to the obstacle speed and convex hull region. The rendered grid cells can be marked as occupied. It is then determined whether the filter_num grid cell, i.e., the rendered grid cell, has been rendered. If the rendered grid cell is marked as occupied, the filtering process ends. If the rendered grid cell is not marked as occupied, but there are obstacles in the rendered grid cell, it indicates that it is a high-altitude cluster noise, and high-altitude cluster noise can be filtered in this case. If the rendered grid cell is not marked as occupied and there are no obstacles in the rendered grid cell, the filtering process ends.

[0060] For filtering non-high-altitude clustered noise, the front end can be obtained, i.e., non-suspended obstacles detected by radar (i.e., non-high-altitude obstacles). Specific regions of grid cells are rendered based on obstacle velocity and convex hull area. Rendered grid cells can be marked as occupied. The filter_num grid cell (i.e., the rendered grid cell) is checked. If the rendered grid cell is marked as occupied, the filtering process ends. If the rendered grid cell is not marked as occupied but contains obstacles, it indicates that it is a non-high-altitude clustered noise, and filtering of non-high-altitude clustered noise can be performed. If the rendered grid cell is not marked as occupied and contains no obstacles, the filtering process ends.

[0061] In this embodiment, a first preset number of distance grids are established with the radar as the origin during vehicle operation. The constructed distance grids may include specific region grids. Then, the specific region grids are marked according to the constructed time series model and the vehicle speed. The specific region grids may include filter grids, and the number of specific region grids is at least greater than the number of filter grids. The filter grids are mainly a second preset number of grids located around the center of the vehicle axle in the distance grids. If a filter grid is marked and occupied in the marked specific region grids, it is determined that there is a real obstacle in the filter grid. If a filter grid is not marked and occupied, it can be determined whether there is an obstacle in the filter grid. If there is an obstacle in the filter grid, it means that there are cluster noise points in the filter grid. At this time, the cluster noise points can be filtered out. This can filter out the cluster noise points formed by the radar in rainy weather without filtering out real obstacles, ensuring smooth autonomous driving and the safety of autonomous driving.

[0062] It should be noted that, for the sake of simplicity, the method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments of this application are not limited to the described order of actions, because according to the embodiments of this application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of this application.

[0063] Reference Figure 3 The diagram illustrates a structural block diagram of a rain-related noise filtering device for autonomous driving radar, as provided in an embodiment of this application. Specifically, it may include the following modules:

[0064] The grid division module 301 is used to establish a preset first number of distance grids with the radar as the origin during vehicle movement; the preset first number of distance grids includes a specific area grid;

[0065] The grid marking module 302 is used to obtain the constructed time series model and vehicle speed, and to mark the grids in a specific region according to the time series model and vehicle speed; the grids in the specific region include filter grids, and the number of grids in the specific region is at least greater than the number of filter grids, wherein the filter grids are a preset second number of grids located around the axle center of the vehicle in a preset first number of distance grids;

[0066] The obstacle determination module 303 is used to determine whether there is a real obstacle in the filter cell when the filter cell is marked as occupied;

[0067] The cluster noise filtering module 304 is used to determine whether there are obstacles in the filter grid when the filter grid is not marked as occupied; if there are obstacles in the filter grid, it is determined that there are cluster noises in the filter grid, and the cluster noises are filtered.

[0068] In some embodiments of this application, the grid division module 301 may include the following sub-modules:

[0069] The distance grid establishment submodule is used to obtain the radar's calibrated scanning range; the scanning range is used to indicate the radar's first scanning distance in a preset first direction and the radar's second scanning distance in a preset second direction; using the first scanning distance and a preset grid size, the number of grids in the preset first direction is calculated, and using the second scanning distance and a preset grid size, the number of grids in the preset second direction is calculated; with the radar as the origin, a preset first number of distance grids is established; the preset first number is the product of the number of grids in the preset first direction and the number of grids in the preset second direction.

[0070] In some embodiments of this application, the preset second quantity is determined based on the radar's altitude difference range and the preset grid size, and the remaining grids in the preset first quantity of distance grids, excluding the preset second quantity of filter grids, are used to acquire timing information;

[0071] The apparatus provided in this application embodiment may further include the following modules:

[0072] The time series model construction module is used to determine the step direction of data processing based on the vehicle's direction of travel, and to determine the starting cell for data processing from the remaining cells; the step direction is consistent with the direction of travel; according to the starting cell and the step direction, the occupancy status of each cell at different timestamps is recorded to obtain the index information of each cell; the time series model is constructed using the index information of each cell.

[0073] In some embodiments of this application, the grid marking module 302 may include the following sub-modules:

[0074] The grid marking submodule is used to acquire target obstacles detected by radar; target obstacles include high-altitude obstacles and non-high-altitude obstacles; target obstacles are tracked through a time series model, and grids in specific areas are marked based on the movement speed and convex hull area of ​​the target obstacles, as well as the vehicle speed.

[0075] In some embodiments of this application, the grid marker submodule may include the following units:

[0076] The grid marking unit is used to track target obstacles through a time-series model. It tracks the positional changes of the target obstacle in two consecutive frames of data through the time-series model and calculates the movement speed of the target obstacle by combining the time interval between the two consecutive frames of data. Based on the movement speed of the target obstacle, the convex hull region, and the vehicle speed, it marks the grid in a specific region.

[0077] In some embodiments of this application, the grid marker unit may include the following sub-units:

[0078] The grid marking subunit is used to filter target obstacles based on their movement speed to obtain filtered target obstacles; and to mark the specific area grid where the filtered target obstacles are located based on the convex hull region of the filtered target obstacles and the vehicle speed. The specific steps for marking the specific area grid where the filtered target obstacles are located based on the convex hull region of the filtered target obstacles and the vehicle speed include: determining the physical coordinates of the filtered target obstacles based on their convex hull region; correcting the coordinates of the filtered target obstacles based on the vehicle speed and the time interval between two consecutive frames of data to obtain the specific area grid where the filtered target obstacles are located; and marking the specific area grid where the filtered target obstacles are located as occupied.

[0079] In this embodiment, a first preset number of distance grids are established with the radar as the origin during vehicle operation. The constructed distance grids may include specific region grids. Then, the specific region grids are marked according to the constructed time series model and the vehicle speed. The specific region grids may include filter grids, and the number of specific region grids is at least greater than the number of filter grids. The filter grids are mainly a second preset number of grids located around the center of the vehicle axle in the distance grids. If a filter grid is marked and occupied in the marked specific region grids, it is determined that there is a real obstacle in the filter grid. If a filter grid is not marked and occupied, it can be determined whether there is an obstacle in the filter grid. If there is an obstacle in the filter grid, it means that there are cluster noise points in the filter grid. At this time, the cluster noise points can be filtered out. This can filter out the cluster noise points formed by the radar in rainy weather without filtering out real obstacles, ensuring smooth autonomous driving and the safety of autonomous driving.

[0080] As the device embodiment is basically similar to the method embodiment, the description is relatively simple, and relevant parts can be found in the description of the method embodiment.

[0081] This application also provides a vehicle, as shown in the embodiments. Figure 4 The provided vehicle 400 includes a memory 410, a processor 420, and a computer program 411 stored in the memory 410 and capable of running on the processor 420. When the computer program 411 is executed by the processor, it implements the various processes of the above-described embodiment of the method for filtering cluster noise points in rainy weather by autonomous driving radar, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0082] This application also provides a computer-readable storage medium, see embodiments thereof. Figure 5 The computer-readable storage medium 500 provides a computer program 411. When the computer program 411 is executed by the processor, it implements the various processes of the above-described method embodiment for filtering cluster noise points in rainy weather using autonomous driving radar, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0083] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0084] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of the embodiments of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or modules is not necessarily limited to those steps or modules explicitly listed, but may include other steps or modules not explicitly listed or inherent to these processes, methods, products, or devices. The division of modules in the embodiments of this application is merely a logical division; in actual applications, there may be other division methods. For example, multiple modules may be combined into or integrated into another system, or some features may be ignored or not performed. Additionally, the shown or discussed mutual coupling or direct coupling or communication connection may be through some interface, and the indirect coupling or communication connection between modules may be electrical or other similar forms, none of which are limited in the embodiments of this application. Furthermore, the modules or sub-modules described as separate components may or may not be physically separated, may or may not be physical modules, or may be distributed among multiple circuit modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the embodiments of this application.

[0085] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0086] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0087] In the embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, apparatuses, or modules, and may be electrical, mechanical, or other forms.

[0088] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0089] Furthermore, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium.

[0090] In the above embodiments, the implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, in the form of a computer program product.

[0091] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0092] This application describes embodiments with reference to flowchart illustrations and / or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, generate instructions for implementing the flowchart illustrations. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0093] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing terminal device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 a process or multiple processes and / or boxes Figure 1 The functions specified in one or more boxes; these computer program instructions may also be loaded onto a computer or other programmable data processing terminal equipment to cause a series of operational steps to be performed on the computer or other programmable terminal equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable terminal equipment for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0094] Although preferred embodiments of the present application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of the present application.

[0095] Finally, it should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation portals are provided for users to choose to authorize or refuse.

[0096] The technical solutions provided in the embodiments of this application have been described in detail above. Specific examples have been used in the embodiments of this application to illustrate the principles and implementation methods of the embodiments of this application. The description of the above embodiments is only for the purpose of helping to understand the methods and core ideas of the embodiments of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the embodiments of this application. Therefore, the content of this specification should not be construed as a limitation on the embodiments of this application.

Claims

1. A method for filtering clustered noise points in autonomous driving radar during rainy weather, characterized in that, The method includes: During vehicle operation, a preset first number of distance grids are established with the radar as the origin; the preset first number of distance grids includes a specific area grid. The constructed time series model and vehicle speed are obtained, and the specific region grid is marked according to the time series model and the vehicle speed; the specific region grid includes filter grids, and the number of specific region grids is at least greater than the number of filter grids, wherein the filter grids are a preset second number of grids located around the axle center of the vehicle in the preset first number of distance grids; If the filter cell is marked as occupied, it is determined that there is a real obstacle in the filter cell; If the filter cell is not marked as occupied, then determine whether there is an obstacle in the filter cell; If there are obstacles in the filter grid, it is determined that there are clustered noise points in the filter grid, and the clustered noise points are filtered.

2. The method according to claim 1, characterized in that, The step of establishing a preset first number of distance grids with the radar as the origin includes: Obtain the scanning range calibrated by the radar; the scanning range is used to indicate the first scanning distance of the radar in a preset first direction and the second scanning distance of the radar in a preset second direction; Using the first scanning distance and the preset grid size, the number of grids in the preset first direction is calculated, and using the second scanning distance and the preset grid size, the number of grids in the preset second direction is calculated. With the radar as the origin, a preset first number of distance grids are established; the preset first number is the product of the number of grids in the preset first direction and the number of grids in the preset second direction.

3. The method according to claim 1, characterized in that, The preset second quantity is determined based on the radar's altitude difference range and a preset grid size; the remaining grids in the preset first quantity of range grids, excluding the preset second quantity of filter grids, are used to acquire timing information; the method further includes: Based on the vehicle's direction of travel, the step direction for data processing is determined, and the starting cell for data processing is determined from the remaining cells; the step direction is consistent with the direction of travel. According to the starting grid and the stepping direction, record the occupancy status of each grid at different timestamps to obtain the index information of each grid; A time series model is constructed using the index information of each grid.

4. The method according to claim 1 or 3, characterized in that, The step of marking the specific region grid according to the time series model and the vehicle speed includes: The radar detects target obstacles; the target obstacles include high-altitude obstacles and non-high-altitude obstacles. The target obstacle is tracked using the time-series model, and the specific region grid is marked based on the target obstacle's movement speed and convex hull region, as well as the vehicle's speed.

5. The method according to claim 4, characterized in that, The step of tracking the target obstacle using the time-series model, and marking the specific region grid based on the target obstacle's movement speed and convex hull region, as well as the vehicle's speed, includes: The target obstacle is tracked using the time-series model, and the position change of the target obstacle in two consecutive frames of data is tracked using the time-series model. The movement speed of the target obstacle is calculated by combining the time interval between the two consecutive frames of data. The specific area grid is marked based on the movement speed and convex hull area of ​​the target obstacle, as well as the vehicle speed.

6. The method according to claim 5, characterized in that, The process of marking the specific region grid based on the movement speed and convex hull region of the target obstacle, and the vehicle speed, includes: The target obstacle is filtered based on its movement speed to obtain a filtered target obstacle; Based on the convex hull region of the filtered target obstacle and the vehicle speed, the specific area grid where the filtered target obstacle is located is marked.

7. The method according to claim 6, characterized in that, The process of marking specific grid areas containing the filtered target obstacles based on the convex hull region of the filtered target obstacles and the vehicle speed includes: Based on the convex hull region of the filtered target obstacle, determine the physical coordinates of the filtered target obstacle. Based on the vehicle speed and the time interval between two consecutive frames of data, the coordinates of the filtered target obstacle are corrected to obtain the specific area grid where the filtered target obstacle is located, and the specific area grid where the filtered target obstacle is located is marked as occupied.

8. A filtering device for clustered noise points in rainy weather for autonomous driving radar, characterized in that, The device includes: A grid division module is used to establish a preset first number of distance grids with the radar as the origin during vehicle movement; the preset first number of distance grids includes a specific area grid. A grid marking module is used to acquire the constructed time series model and vehicle speed, and to mark the grid of the specific region according to the time series model and the vehicle speed; the grid of the specific region includes filter grids, and the number of the specific region grids is at least greater than the number of filter grids, wherein the filter grids are a preset second number of grids located around the axle center of the vehicle in the preset first number of distance grids; An obstacle determination module is used to determine that a real obstacle exists in the filter cell when the filter cell is marked as occupied; The cluster noise filtering module is used to determine whether there is an obstacle in the filter cell when the filter cell is not marked as occupied; if there is an obstacle in the filter cell, it is determined that there is cluster noise in the filter cell, and the cluster noise is filtered.

9. A vehicle, characterized in that, include: A processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein the computer program, when executed by the processor, implements the method for filtering rain-related cluster noise in autonomous driving radar as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, which, when executed by a processor, implements the method for filtering rain-related cluster noise points in autonomous driving radar as described in any one of claims 1 to 7.

Images