Method for realizing horizontal installation angle detection compensation based on freespace road end 4D millimeter wave radar

Abstract

The present application relates to a kind of based on freespace road end 4D millimeter wave radar to realize horizontal installation angle detection compensation method, wherein, the method includes the following steps: 4D millimeter wave radar input original detection target;Collect all detection targets in the detection range of 4D millimeter wave radar and carry out point cloud filtering;Get region of interest ROI;According to the resolution size of self-defined setting grid, each detection target in region of interest ROI is updated according to the resolution size and is handled to occupy grid;The point cloud of being filtered with passable area detection is handled to online point cloud registration;Whether the pitch angle of the 4D millimeter wave radar detected needs to be compensated is judged, and corresponding processing is executed to the judgment result.The present application also relates to a corresponding system, device, processor and its storage medium.The method, system, device, processor and its storage medium of using the present application solve the problem that road end radar horizontal installation angle alignment process is complicated and not easy to operate.

Description

Technical Field

[0001] This invention relates to the field of intelligent transportation technology, and more particularly to the field of 4D millimeter-wave radar technology. Specifically, it relates to a method, system, device, processor, and computer-readable storage medium for horizontal installation angle detection and compensation based on Freespace roadside 4D millimeter-wave radar. Background Technology

[0002] In the field of intelligent transportation, the accuracy of roadside radar in sensing targets is often closely related to the horizontal angle of the radar installation. When the sensing system is initialized, the horizontal installation angle of the radar needs to be detected, calibrated, and compensated.

[0003] Currently, most radar horizontal installation angle calibration and compensation methods rely on joint calibration with other sensors (such as cameras, lidar, corner reflectors, etc.), or require the vehicle itself to continuously detect fixed reference objects in the environment (such as QR codes, guardrails, and the position information of other vehicles in the environment) to solve for the radar's horizontal installation angle using least-squares fitting of a straight line. While these methods can calibrate the radar's horizontal installation angle, they are cumbersome, inconvenient to operate, and the calibration results are greatly affected by external interference, often making real-time online operation impossible. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and provide a simple, convenient, and non-manually-intervention-required method, system, device, processor, and computer-readable storage medium for horizontal installation angle detection and compensation based on Freespace roadside 4D millimeter-wave radar.

[0005] To achieve the above objectives, the present invention provides a method, system, apparatus, processor, and computer-readable storage medium for horizontal installation angle detection and compensation based on a freespace roadside 4D millimeter-wave radar, as follows:

[0006] The method, system, device, processor, and computer-readable storage medium for horizontal installation angle detection and compensation based on Freespace roadside 4D millimeter-wave radar are characterized in that the method includes the following steps:

[0007] (1) Input the original target for detection into the 4D millimeter-wave radar;

[0008] (2) Collect all the detected targets within the detection range of the 4D millimeter-wave radar, and perform pass-through filtering by extracting the corresponding point cloud;

[0009] (3) Delineate the range of the passable detection area in the original detection target and obtain the region of interest (ROI);

[0010] (4) Based on the grid resolution size set by the user, update the grid occupancy of each detected target in the region of interest (ROI) according to the resolution size.

[0011] (5) Perform online point cloud registration processing on the point cloud after point cloud filtering and the point cloud of passable area detection to obtain the detection value of the horizontal installation angle of the 4D millimeter wave radar at the current moment.

[0012] (6) Determine whether the elevation angle of the 4D millimeter-wave radar detected at the current moment needs to be compensated, and perform corresponding processing on the determination result.

[0013] Preferably, step (2) specifically includes:

[0014] A 4D millimeter-wave radar is installed horizontally at a certain point on the roadside. It collects data on all targets within the radar's detection range and extracts the point cloud corresponding to the currently detected target from the original input data. The obtained point cloud is then subjected to pass-through filtering, and point clouds outside a 100m range are filtered out. This is represented as:

[0015] Assuming the current time is i, and n targets have been collected after point cloud filtering, let's denote them as the point cloud to be registered.

[0016] Preferably, step (4) specifically includes:

[0017] Based on the custom-defined raster resolution, the input raw detection targets are generated into an m*n raster map, and the position x of each detection target within the current detection range is set. i The target is projected onto a grid map based on its resolution, and the probability values ​​of the detected targets falling into the grid map are superimposed. The normalized probability values ​​of each grid cell are statistically analyzed to obtain the probability statistical value p of each grid cell. i This completes the update of the occupied grid.

[0018] Preferably, step (5) specifically includes the following steps:

[0019] (5.1) The points where the probability statistics are not zero are taken as the boundary points X of the passable area. i {x0,x1…x j Assuming that m points are retained after the grid update, these boundary points are combined to form the output point cloud of the passable region at time i.

[0020] (5.2) Use the output point cloud of the passable area as the target point cloud to be registered at the current time i. The point cloud obtained by passing through filtering the original detected targets of the 4D millimeter-wave radar is used as the input point cloud to be registered.

[0021] (5.3) The point cloud registration algorithm ICP nearest neighbor iterative algorithm is used to iterate N times until the point cloud to be registered is reached. After transformation T i (R i ,t i ) and target point cloud The Euclidean distance d between them is less than the preset threshold δ d The transformation after the last iteration is used as the final output T of the point cloud registration. i最终 (R i ,t i ), from transformation T i最终 (R i ,t i The rotational part extracts the change in pitch angle. The output of the current horizontal installation angle detection value is used to complete online point cloud registration.

[0022] Preferably, step (6) specifically includes the following steps:

[0023] (6.1) Obtain the compensation angle of the horizontal installation angle of the 4D millimeter-wave radar at the current moment.

[0024] (6.2) Determine the compensation angle Is it greater than the preset compensation threshold δ? pitch If it is greater than , no compensation will be performed; otherwise, proceed to step (6.3).

[0025] (6.3) The currently acquired compensation angle Convert to the compensation angle rotation matrix ΔR as follows: i :

[0026]

[0027] (6.4) The compensation angle rotation matrix ΔR i The following formula is applied to the corresponding position x of each of the original detection targets. i The above is used to obtain the compensated target point x. i ′:

[0028] x i ′=ΔR i *x i ;

[0029] (6.5) Input the detection target after horizontal installation angle compensation back into step (1) and repeat the process.

[0030] The system for horizontal installation angle detection and compensation based on Freespace roadside 4D millimeter-wave radar for implementing the above-described method is characterized in that the system comprises:

[0031] The horizontal installation angle detection module is used to perform pass-through filtering of the target point cloud and detection of the passable area within the detection range of the 4D millimeter-wave radar, based on the original target data input by the system. It then performs online point cloud registration based on the acquired point cloud to be registered and the passable area point cloud to obtain the compensation angle of the 4D millimeter-wave radar's horizontal installation angle at the current moment.

[0032] The horizontal installation angle compensation module is connected to the horizontal installation angle detection module and is used to perform online angle compensation processing for compensation angles that are less than the preset compensation threshold according to the system.

[0033] The main feature of this device, which uses Freespace roadside 4D millimeter-wave radar to achieve horizontal installation angle detection and compensation, is that the device includes:

[0034] A processor is configured to execute computer-executable instructions;

[0035] The memory stores one or more computer-executable instructions, which, when executed by the processor, implement the various steps of the method for horizontal installation angle detection and compensation based on the freespace roadside 4D millimeter-wave radar described above.

[0036] The processor for horizontal installation angle detection and compensation based on Freespace roadside 4D millimeter-wave radar is characterized in that the processor is configured to execute computer-executable instructions, which, when executed by the processor, implement the various steps of the method for horizontal installation angle detection and compensation based on Freespace roadside 4D millimeter-wave radar.

[0037] The computer-readable storage medium is characterized in that it stores a computer program that can be executed by a processor to implement the various steps of the method for horizontal installation angle detection and compensation based on the freespace roadside 4D millimeter-wave radar described above.

[0038] The present invention, employing a method, system, device, processor, and computer-readable storage medium for horizontal installation angle detection and compensation based on a Freespace roadside 4D millimeter-wave radar, solves the problems of cumbersome and difficult operation in aligning the horizontal installation angle of roadside radars. By detecting and compensating for the radar's horizontal installation angle in real time, the robustness and accuracy of radar sensing are increased. Since no other media are required, calibration costs and workload are reduced, improving efficiency. Furthermore, the radar horizontal installation angle detection proposed in this technical solution is based on matching Freespace point clouds and real-time point clouds to obtain the radar's horizontal installation angle, without needing to construct equations based on the number of radars and corner reflectors to solve for the parameters and determine the installation angle. This simplifies the calculation, eliminating the process of constructing and solving complex equations, and compared to other traditional radars (which do not output point clouds), it has better applicability. Attached Figure Description

[0039] Figure 1 This is a flowchart of the method for horizontal installation angle detection and compensation based on the freespace roadside 4D millimeter-wave radar of the present invention. Detailed Implementation

[0040] To more clearly describe the technical content of the present invention, the following description is provided in conjunction with specific embodiments.

[0041] Before describing the embodiments of the present invention in detail, it should be noted that, in the following, the terms “comprising,” “including,” or any other variations are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed or inherent to such process, method, article, or apparatus.

[0042] Before detailing this technical solution, the following explanations are provided for some of the English abbreviations used in this technical solution:

[0043] freespace: passable area

[0044] radar: Here it specifically refers to 4D millimeter-wave radar.

[0045] targets: targets output by millimeter-wave radar

[0046] ROI: Region of Interest

[0047] ICP: Iterative Clostest Point - Nearest Neighbor Iteration Point

[0048] T: Transform, rigid body transformation, including translation and rotation components.

[0049] R: Rotation, the rotational component in a rigid body transformation.

[0050] Please see Figure 1 As shown, this method for horizontal installation angle detection and compensation based on Freespace roadside 4D millimeter-wave radar includes the following steps:

[0051] (1) Input the original target for detection into the 4D millimeter-wave radar;

[0052] (2) Collect all the detected targets within the detection range of the 4D millimeter-wave radar, and perform pass-through filtering by extracting the corresponding point cloud;

[0053] (3) Delineate the range of the passable detection area in the original detection target and obtain the region of interest (ROI);

[0054] (4) Based on the grid resolution size set by the user, update the grid occupancy of each detected target in the region of interest (ROI) according to the resolution size.

[0055] (5) Perform online point cloud registration processing on the point cloud after point cloud filtering and the point cloud of passable area detection to obtain the detection value of the horizontal installation angle of the 4D millimeter wave radar at the current moment.

[0056] (6) Determine whether the elevation angle of the 4D millimeter-wave radar detected at the current moment needs to be compensated, and perform corresponding processing on the determination result.

[0057] In a preferred embodiment of the present invention, step (2) specifically comprises:

[0058] A 4D millimeter-wave radar is installed horizontally at a certain point on the roadside. It collects data on all targets within the radar's detection range and extracts the point cloud corresponding to the currently detected target from the original input data. The obtained point cloud is then subjected to pass-through filtering, and point clouds outside a 100m range are filtered out. This is represented as:

[0059] Assuming the current time is i, and n targets have been collected after point cloud filtering, let's denote them as the point cloud to be registered.

[0060] In a preferred embodiment of the present invention, step (4) specifically comprises:

[0061] Based on the custom-defined raster resolution, the input raw detection targets are generated into an m*n raster map, and the position x of each detection target within the current detection range is set. i The target is projected onto a grid map based on its resolution, and the probability values ​​of the detected targets falling into the grid map are superimposed. The normalized probability values ​​of each grid cell are statistically analyzed to obtain the probability statistical value p of each grid cell. i This completes the update of the occupied grid.

[0062] In a preferred embodiment of the present invention, step (5) specifically includes the following steps:

[0063] (5.1) The points where the probability statistics are not zero are taken as the boundary points X of the passable area. i {x0,x1…x j Assuming that m points are retained after the grid update, these boundary points are combined to form the output point cloud of the passable region at time i.

[0064] (5.2) Use the output point cloud of the passable area as the target point cloud to be registered at the current time i. The point cloud obtained by passing through filtering the original detected targets of the 4D millimeter-wave radar is used as the input point cloud to be registered.

[0065] (5.3) The point cloud registration algorithm ICP nearest neighbor iterative algorithm is used to iterate N times until the point cloud to be registered is reached. After transformation T i (R i ,t i ) and target point cloud The Euclidean distance d between them is less than the preset threshold δ d The transformation after the last iteration is used as the final output T of the point cloud registration. i最终 (R i ,t i ), from transformation T i最终 (R i ,t i The rotational part extracts the change in pitch angle. The output of the current horizontal installation angle detection value is used to complete online point cloud registration.

[0066] In a preferred embodiment of the present invention, step (6) specifically includes the following steps:

[0067] (6.1) Obtain the compensation angle of the horizontal installation angle of the 4D millimeter-wave radar at the current moment.

[0068] (6.2) Determine the compensation angle Is it greater than the preset compensation threshold δ? pitch If it is greater than , no compensation will be performed; otherwise, proceed to step (6.3).

[0069] (6.3) The currently acquired compensation angle Convert to the compensation angle rotation matrix ΔR as follows: i :

[0070]

[0071] (6.4) The compensation angle rotation matrix ΔR i The following formula is applied to the corresponding position x of each of the original detection targets. i The above is used to obtain the compensated target point x. i ′:

[0072] x i ′=ΔR i *x i ;

[0073] (6.5) Input the detection target after horizontal installation angle compensation back into step (1) and repeat the process.

[0074] In practical applications, this technical solution includes two main sub-modules: a radar horizontal installation angle detection module and a radar horizontal installation angle compensation module. The corresponding modules perform the following specific processing:

[0075] 1) Horizontal installation angle detection module

[0076] The horizontal installation angle detection module involves the following processing steps:

[0077] 1. Point Cloud Filtering. A radar is installed horizontally at a certain point on the roadside. All targets within the radar's detection range are collected. The point cloud corresponding to each target is extracted from the input targets. A pass-through filter is applied to the obtained point cloud to filter out point clouds outside a 100m range. Assume that at time i, n targets are collected after point cloud filtering, denoted as the point cloud to be registered.

[0078] 2. Freespace detection

[0079] i. Obtain the Region of Interest (ROI). Within the targets of the original radar input, define the area of ​​interest (ROI) as the passable region to be detected. For example, a three-dimensional region consisting of 100m directly in front (x-direction), 50m to the left and right (y-direction), and 20m forward (z-direction) is the ROI to be detected.

[0080] ii. Raster update. Based on the defined raster resolution size, for example, a raster occupies 50cm, generate an m*n raster map, and then update the x-coordinate of each target's location. i Based on the resolution, the target is projected onto a grid map, and the probability values ​​of the grid cells to which it falls are superimposed. Finally, the normalized probability values ​​of each grid cell are statistically analyzed to obtain the probability statistical value p for each grid cell. i ;

[0081] iii. Output the freespace point cloud. Use the points containing grid cells with non-zero probability statistics as the boundary points X of the passable area. i {x0,x1…x j Assuming that m points are retained after the grid update, these boundary points are combined to form the free space at time i, and the output point cloud is generated.

[0082] 3. Point cloud registration

[0083] The output point cloud of freespace is used as the target point cloud for registration at the current time i. The filtered point cloud from the original radar targets is used as the input point cloud to be registered. The point cloud registration algorithm, ICP nearest neighbor iterative algorithm, is used repeatedly N times until the point cloud to be registered is found. After transformation T i (R i ,t i ) and target point cloud The Euclidean distance d between them is less than a certain threshold δ d The transformation of the last iteration is used as the final output T of the point cloud registration. i (R i ,t i ), from transformation T i (R i ,t i The rotational part extracts the change in pitch angle. As the output of the horizontal installation angle detection module at the current moment;

[0084] 2) Horizontal installation angle compensation module

[0085] In step 1), the radar horizontal installation compensation angle at the current time i was obtained. Then, determine the compensation angle. Is it greater than the set compensation threshold δ? pitch If the value is greater than the value, no compensation is performed; otherwise, the compensation angle is converted into a compensation angle rotation matrix ΔR according to the following formula. i :

[0086]

[0087] Then this pitch compensation rotation matrix ΔR i The following formula is applied to each target position x in the original radar input. i Above, the compensated target point x is obtained. i ′

[0088] x i ′=ΔR i *x i

[0089] The input target after radar horizontal installation angle compensation is then input into the horizontal installation angle detection module, and so on in real time.

[0090] The system for implementing the above-described method, based on a freespace roadside 4D millimeter-wave radar, includes a horizontal mounting angle detection and compensation system, comprising:

[0091] The horizontal installation angle detection module is used to perform pass-through filtering of the target point cloud and detection of the passable area within the detection range of the 4D millimeter-wave radar, based on the original target data input by the system. It then performs online point cloud registration based on the acquired point cloud to be registered and the passable area point cloud to obtain the compensation angle of the 4D millimeter-wave radar's horizontal installation angle at the current moment.

[0092] The horizontal installation angle compensation module is connected to the horizontal installation angle detection module and is used to perform online angle compensation processing for compensation angles that are less than the preset compensation threshold according to the system.

[0093] This device, based on Freespace roadside 4D millimeter-wave radar, enables horizontal installation angle detection and compensation. The device comprises:

[0094] A processor is configured to execute computer-executable instructions;

[0095] The memory stores one or more computer-executable instructions, which, when executed by the processor, implement the various steps of the method for horizontal installation angle detection and compensation based on the freespace roadside 4D millimeter-wave radar described above.

[0096] The processor for horizontal installation angle detection and compensation based on Freespace roadside 4D millimeter-wave radar is configured to execute computer-executable instructions. When executed by the processor, the computer-executable instructions implement the various steps of the method for horizontal installation angle detection and compensation based on Freespace roadside 4D millimeter-wave radar.

[0097] The computer-readable storage medium contains a computer program that can be executed by a processor to implement the various steps of the method for horizontal installation angle detection and compensation based on the freespace roadside 4D millimeter-wave radar described above.

[0098] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of the preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as will be understood by those skilled in the art to which embodiments of the invention pertain.

[0099] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution device.

[0100] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0101] The storage media mentioned above can be read-only memory, disk, or optical disk, etc.

[0102] In the description of this specification, references to terms such as "an embodiment," "some embodiments," "example," "specific example," or "embodiment," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0103] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

[0104] This technical solution, based on a 4D millimeter-wave radar at the roadside in Freespace, achieves horizontal installation angle detection and compensation without requiring vehicle movement or other media. It only requires the radar to be installed at the roadside, without needing any specific reference object. The radar's raw output target is used to detect the boundary points of the passable Freespace area. The boundary point cloud of the passable Freespace area is then registered with the radar's raw output point cloud to detect the compensation angle for the current horizontal installation angle. This compensation angle is then added to the input of the next frame's point cloud. This process is repeated in real-time, resulting in a more accurate point cloud output from the radar to the downstream target detection module. This method of horizontal installation angle detection and compensation is simple to operate, convenient, requires no manual intervention, and offers high accuracy.

[0105] The present invention, employing a method, system, device, processor, and computer-readable storage medium for horizontal installation angle detection and compensation based on a Freespace roadside 4D millimeter-wave radar, solves the problems of cumbersome and difficult operation in aligning the horizontal installation angle of roadside radars. By detecting and compensating for the radar's horizontal installation angle in real time, the robustness and accuracy of radar sensing are increased. Since no other media are required, calibration costs and workload are reduced, improving efficiency. Furthermore, the radar horizontal installation angle detection proposed in this technical solution is based on matching Freespace point clouds and real-time point clouds to obtain the radar's horizontal installation angle, without needing to construct equations based on the number of radars and corner reflectors to solve for the parameters and determine the installation angle. This simplifies the calculation, eliminating the process of constructing and solving complex equations, and compared to other traditional radars (which do not output point clouds), it has better applicability.

[0106] In this specification, the invention has been described with reference to specific embodiments thereof. However, it will be apparent that various modifications and variations can be made without departing from the spirit and scope of the invention. Therefore, the specification and drawings should be considered illustrative rather than restrictive.

Claims

1. A method for horizontal installation angle detection and compensation based on Freespace road-end 4D millimeter-wave radar, characterized in that, The method includes the following steps: (1) Input the original target for detection into the 4D millimeter-wave radar; (2) Collect all the detected targets within the detection range of the 4D millimeter-wave radar, and perform pass-through filtering by extracting the corresponding point cloud; (3) Delineate the range of the passable detection area in the original detection target and obtain the region of interest (ROI); (4) Based on the custom-set grid resolution, update the grid occupancy of each detected target in the region of interest (ROI) according to the resolution. (5) Perform online point cloud registration processing on the point cloud after point cloud filtering and the point cloud of passable area detection to obtain the detection value of the horizontal installation angle of the 4D millimeter wave radar at the current moment; Step (5) specifically includes the following steps: (5.1) The points where the probability statistics are not zero are taken as the boundary points of the passable area. Assuming that after updating the occupied grid, there are If any point is retained, these boundary points are combined to form the passable area at the current moment. Output point cloud ; (5.2) Take the output point cloud of the passable area as the current time. Registered target point cloud The point cloud obtained by passing through filtering the original detected target of the 4D millimeter-wave radar is used as the input point cloud to be registered. ; (5.3) The point cloud registration algorithm ICP nearest neighbor iterative algorithm is used to iterate repeatedly. Next, until the point cloud to be registered. After transformation and target point cloud Euclidean distance between Error less than preset threshold The transformation after the last iteration is used as the final output of point cloud registration. From transformation Take the pitch angle from the rotating part Change As of the present moment The horizontal installation angle detection value is output to complete online point cloud registration; (6) Determine whether the elevation angle of the 4D millimeter-wave radar detected at the current moment needs to be compensated, and perform corresponding processing on the determination result; Step (6) specifically includes the following steps: (6.1) Obtain the compensation angle of the horizontal installation angle of the 4D millimeter-wave radar at the current moment. ; (6.2) Determine the compensation angle Is it greater than the preset compensation threshold? If it is greater than , no compensation will be performed; otherwise, proceed to step (6.3). (6.3) The currently acquired compensation angle Convert to a compensation angle rotation matrix as follows: : (6.4) The compensation angle rotation matrix is ​​described above. The following formula is applied to the corresponding position of each detection target of the original detection target. The above data is used to obtain the compensated target detection points. : ； (6.5) Input the detection target after horizontal installation angle compensation back into step (1) and repeat the process.

2. The method for horizontal installation angle detection and compensation based on freespace road-end 4D millimeter-wave radar according to claim 1, characterized in that, The specific steps (2) are as follows: A 4D millimeter-wave radar is installed horizontally at a certain location on the roadside. It collects data on all targets within the radar's detection range and extracts the point cloud corresponding to the currently detected target from the original input data. The obtained point cloud is then subjected to pass-through filtering, and point clouds outside a 100m range are filtered out. This is represented as: Assume the current time is time. After point cloud filtering, a total of [number] samples were collected. Each detected target is recorded as a point cloud to be registered. .

3. The method for horizontal installation angle detection and compensation based on freespace road-end 4D millimeter-wave radar according to claim 2, characterized in that, The specific steps (4) are as follows: Based on the custom-defined raster resolution, the input raw detection target is generated into a... A grid map of a certain size, and the location of each detected target within the current detection range. The data is projected onto a grid map based on the resolution, and the probability values ​​of the detected targets falling into the grid map are superimposed. Normalized probability values ​​are then statistically analyzed for each grid cell to obtain the probability statistics for each grid cell. This completes the update of the occupied grid.

4. A system for horizontal installation angle detection and compensation based on a freespace roadside 4D millimeter-wave radar for implementing the method of any one of claims 1 to 3, characterized in that, The system includes: The horizontal installation angle detection module is used to perform pass-through filtering of the target point cloud and detection of the passable area within the detection range of the 4D millimeter-wave radar, based on the original target data input by the system. It then performs online point cloud registration based on the acquired point cloud to be registered and the passable area point cloud to obtain the compensation angle of the 4D millimeter-wave radar's horizontal installation angle at the current moment. The horizontal installation angle compensation module is connected to the horizontal installation angle detection module and is used to perform online angle compensation processing for compensation angles that are less than the preset compensation threshold according to the system.

5. A device for horizontal installation angle detection and compensation based on Freespace road-end 4D millimeter-wave radar, characterized in that, The device includes: A processor is configured to execute computer-executable instructions; The memory stores one or more computer-executable instructions, which, when executed by the processor, implement the steps of the method for horizontal installation angle detection and compensation based on the freespace roadside 4D millimeter-wave radar as described in any one of claims 1 to 3.

6. A processor for horizontal installation angle detection and compensation based on Freespace roadside 4D millimeter-wave radar, characterized in that, The processor is configured to execute computer-executable instructions, which, when executed by the processor, implement the steps of the method for horizontal installation angle detection and compensation based on the freespace roadside 4D millimeter-wave radar as described in any one of claims 1 to 3.

7. A computer-readable storage medium, characterized in that, It stores a computer program that can be executed by a processor to implement the various steps of the method for horizontal installation angle detection and compensation based on the freespace roadside 4D millimeter-wave radar as described in any one of claims 1 to 3.

Images