An AGV double-radar installation error calibration method and device, and a storage medium

Abstract

This invention discloses a method, device, and storage medium for calibrating the installation error of dual radars in an AGV (Automated Guided Vehicle). The method includes: constructing a first calibration environment, acquiring corresponding point cloud data, calculating an error matrix, and correcting the theoretical installation parameters of one of the radars to achieve alignment between the two radars; constructing a second calibration environment, acquiring corresponding point cloud data, obtaining the angular relationship between the AGV's movement direction and the two radars, and further correcting the installation parameters to align the movement directions of the two radars and the AGV; and aligning the centroids of the two radars and the control points of the AGV based on corresponding centroid parameters. This method allows for the recalculation of the installation parameters of the dual radars, reducing the workload of dual radar installation, improving the accuracy of dual-radar vehicles, and enhancing the overall vehicle positioning accuracy.

Description

Technical Field

[0001] This invention relates to the field of AGV technology, and in particular to a method, equipment, and storage medium for calibrating the installation error of dual radars in AGVs. Background Technology

[0002] With the development of AGVs, their application scenarios are becoming increasingly diverse. In typical dual-radar AGVs, the parameters of the two radars installed on the AGV relative to the vehicle's control points need to be as precise as possible to ensure accurate positioning during navigation.

[0003] In existing technologies, the mechanical parameters of the AGV navigation radar are designed according to the mechanical theoretical values ​​for AGV positioning. However, due to manufacturing process reasons, such as the fit error between mechanical parts, the machining error of the mounting surface, the zero position error of the radar itself, and the non-overlapping scanning parameters between the two radars, the AGV cannot calculate the accurate position of the vehicle body control point during navigation, resulting in the control system being unable to correctly control the AGV to reach the accurate position. Summary of the Invention

[0004] The purpose of this invention is to provide a method, device, and storage medium for calibrating the installation error of dual radars in AGVs, so as to solve one or more technical problems existing in the prior art, and at least provide a beneficial option or create conditions.

[0005] The solution to the technical problem of this invention is to provide a method and equipment for calibrating the installation error of dual radars in AGVs, as well as a storage medium.

[0006] According to an embodiment of a first aspect of the present invention, a method for calibrating the installation error of dual radars in an AGV is provided, comprising the following steps:

[0007] The theoretical installation parameters corresponding to the two radars are obtained respectively, wherein the two radars are installed diagonally on the AGV vehicle;

[0008] Construct the first calibration environment, acquire point cloud data of corresponding scans from the two radars respectively, and calculate the error matrix based on one point cloud data.

[0009] Based on the error matrix, the theoretical installation parameters corresponding to another point cloud data are corrected to obtain the first corrected parameters, and the first theoretical installation parameters corresponding to a point cloud data are obtained.

[0010] When the AGV meets the set stopping conditions, a second calibration environment is constructed, and a first point cloud data of a radar scan is obtained. The AGV moves a set distance and obtains a second point cloud data of another radar scan.

[0011] Using the first point cloud data and the second point cloud data, a translation matrix is ​​calculated, and based on the translation matrix, the angular relationship between the AGV's moving direction and the dual radars is obtained.

[0012] Based on the angular relationship, the first correction parameter and the first theoretical installation parameter are corrected to obtain the second correction parameter and the third correction parameter, respectively.

[0013] Obtain the centroid parameters corresponding to the two radars, and calculate the installation error parameters corresponding to the two radars based on the second correction parameter, the third correction parameter, and the corresponding theoretical installation parameters. Correct the corresponding theoretical installation parameters based on the corresponding installation error parameters.

[0014] Furthermore, the construction of the second calibration environment when the AGV meets the set stopping conditions specifically includes:

[0015] A radar coordinate system is constructed, which includes: X degree of freedom, Y degree of freedom and YAW degree of freedom;

[0016] The AGV is moved with the positive direction of the X degree of freedom as the direction of movement. The AGV stops moving when the direction of the wheels of the AGV is consistent with the direction of movement of the AGV.

[0017] When the AGV stops moving, a second calibration environment is constructed.

[0018] Furthermore, the process of the AGV moving a predetermined distance to acquire second point cloud data from another radar scan specifically includes:

[0019] The AGV moves a set distance with the positive direction of the X-degree of freedom as the direction of movement;

[0020] When the AGV stops moving, acquire second point cloud data from another radar scan.

[0021] Furthermore, the step of calculating a translation matrix using the first and second point cloud data, and obtaining the angular relationship between the AGV's direction of movement and the dual radars based on the translation matrix, specifically includes:

[0022] Using the first point cloud data and the second point cloud data, a translation matrix is ​​calculated based on a point cloud matching algorithm. The translation matrix includes: first translation data and second translation data.

[0023] Using the formula for angular relationships:

[0024]

[0025] ,in, For the first translation data, Using the second translation data, the angular difference θ between the moving direction of the AGV and the YAW degree of freedom is calculated.

[0026] Furthermore, the step of correcting the first correction parameter and the first theoretical installation parameter according to the angular relationship to obtain the second correction parameter and the third correction parameter specifically includes:

[0027] Based on the angle difference, the first correction parameter and the first theoretical installation parameter are corrected respectively to obtain the second correction parameter and the third correction parameter.

[0028] Furthermore, the construction of the first calibration environment, acquiring point cloud data from the corresponding scans of the two radars, and calculating the error matrix based on one point cloud data specifically includes:

[0029] A first calibration environment is constructed, wherein the scanning areas of the dual radars have a common field of view, and the first calibration environment is located within the common field of view;

[0030] In the first calibration environment, one radar scans to obtain the third point cloud data, and another radar scans to obtain the fourth point cloud data;

[0031] Based on the point cloud matching algorithm, one point cloud data is used as a reference to process another point cloud data and calculate the error matrix.

[0032] Furthermore, the step of correcting the theoretical installation parameters corresponding to another point cloud data according to the error matrix to obtain the first correction parameter specifically includes:

[0033] Based on the error matrix, rotation and translation calculations are performed on the theoretical installation parameters corresponding to another point cloud data to obtain the first correction parameters.

[0034] Furthermore, the step of obtaining the centroid parameters corresponding to the two radars, and obtaining the installation error parameters corresponding to the two radars based on the second correction parameter, the third correction parameter, and the corresponding theoretical installation parameters, specifically includes:

[0035] Obtain the centroid parameters corresponding to the two radars, and subtract the corresponding centroid parameters from the second and third correction parameters to obtain the installation parameters corresponding to the two radars.

[0036] The installation error parameters for the dual radars are obtained by subtracting the corresponding installation parameters from the corresponding theoretical installation parameters.

[0037] According to a second aspect of the present invention, an electronic device is provided, comprising:

[0038] A memory for storing a program; a processor for executing the program stored in the memory, wherein when the processor executes the program stored in the memory, the processor is configured to execute an AGV dual-radar installation error calibration method as described in any one of the first aspects.

[0039] According to a third aspect of the present invention, a storage medium is provided, comprising: storing computer-executable instructions for performing an AGV dual-radar installation error calibration method as described in any one of the first aspects.

[0040] The beneficial effects of this invention are as follows: By constructing a first calibration environment, acquiring corresponding point cloud data, and calculating the error matrix, the theoretical installation parameters of one of the radars are corrected, achieving alignment between the two radars. By constructing a second calibration environment, acquiring corresponding point cloud data, the angular relationship between the AGV's movement direction and the two radars is obtained, and the installation parameters are corrected again to align the movement directions of the two radars and the AGV. Based on the corresponding centroid parameters, the centroids of the two radars and the control points of the AGV are aligned. Through the method of this invention, the installation parameters of the two radars can be recalculated, reducing the workload of installing the two radars, improving the accuracy of the two radar vehicles, and enhancing the overall vehicle positioning accuracy. Attached Figure Description

[0041] Figure 1 This is a schematic flowchart of an AGV dual-radar installation error calibration method provided in an embodiment of the present invention;

[0042] Figure 2 This is a schematic diagram of the dual radar scanning feature surface of an AGV dual radar installation error calibration method provided in another embodiment of the present invention.

[0043] Reference numerals: 100, common viewing area; 200, feature surface; 300, first radar scanning area; 400, second radar scanning area. Detailed Implementation

[0044] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and should not be construed as limiting the scope of the invention.

[0045] It should be noted that although functional modules are divided in the system diagram, in some cases, the steps shown or described may be executed in a different order than the module division or flowchart shown in the system. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0046] In the description of this invention, it should be noted that, unless otherwise explicitly defined, terms such as "setting," "installation," and "connection" should be interpreted broadly, and those skilled in the art can reasonably determine the specific meaning of the above terms in this invention in conjunction with the specific content of the technical solution.

[0047] According to an embodiment of the first aspect of the present invention, referring to Figure 1 In some embodiments of the present invention, an AGV dual-radar installation error calibration method includes the following steps:

[0048] S100 obtains the theoretical installation parameters for the two radars respectively, wherein the two radars are installed diagonally on the AGV.

[0049] In this embodiment, the theoretical installation parameters P1 of the first radar and the theoretical installation parameters P2 of the second radar are obtained from the mechanism design. The first and second radars are installed diagonally symmetrically on the AGV, therefore, there are two relatively overlapping areas between the scanning areas of the first and second radars. These two relatively overlapping areas constitute the common field of view of the radars. The first and second radars can be lidar.

[0050] S200: Construct the first calibration environment, acquire point cloud data of corresponding scans from the two radars respectively, and calculate the error matrix based on one point cloud data.

[0051] In this embodiment, the AGV is moved to a first calibration environment, which can be a location with strong feature surfaces or features placed around the AGV. In the first calibration environment, a first radar and a second radar scan to acquire corresponding point cloud data. One of the point cloud data is used as the reference data; that is, either the point cloud data scanned by the first radar or the point cloud data scanned by the second radar can be used as the reference data. Based on the reference data, the other point cloud data is processed to calculate an error matrix. The error matrix is ​​a transformation matrix T, which includes a translation matrix and a rotation matrix.

[0052] S300, based on the error matrix, corrects the theoretical installation parameters corresponding to another point cloud data to obtain the first corrected parameters, and obtains the first theoretical installation parameters corresponding to a point cloud data.

[0053] In this embodiment, based on the error matrix obtained in S200, the theoretical installation parameters P2 of the second radar are corrected using the point cloud data scanned by the first radar as reference data. This results in the first corrected data P. 21 The theoretical installation parameters of the first radar are not modified in any way; the first theoretical installation parameter P1 is obtained.

[0054] When the point cloud data scanned by the second radar is used as the reference data, the theoretical installation parameters P1 of the first radar are corrected. This yields the first corrected data P. 11 The theoretical installation parameters of the second radar are not modified in any way; the first theoretical installation parameter P2 is obtained.

[0055] Through steps S200 and S300 above, based on the error matrix, the first and second radars are aligned, and their relative installation parameters are corrected to ensure overlap. In other words, it can be assumed that the radar coordinate systems of the first and second radars are now consistent.

[0056] S400: When the AGV meets the set stopping conditions, a second calibration environment is constructed, and the first point cloud data of a radar scan is obtained. The AGV moves a set distance and obtains the second point cloud data of another radar scan.

[0057] In this embodiment, the AGV is moved, and it is determined whether the AGV meets the set stopping conditions. If so, a second calibration environment is constructed. The second calibration environment can be a location with a strong feature surface or features placed around the AGV. When a strong feature surface is used as the calibration environment, this feature surface can be set in the direction of movement of the AGV.

[0058] In the second calibration environment, one radar scans and acquires the first point cloud data. After recording, the AGV is slowly moved forward a predetermined distance, ranging from 0.8m to 1.5m. Once the AGV has moved completely, the other radar scans again to acquire the second point cloud data.

[0059] The S500 calculates a translation matrix using the first and second point cloud data, and obtains the angular relationship between the AGV's moving direction and the dual radars based on the translation matrix.

[0060] In this embodiment, the translation matrix between the first and second point cloud data obtained in S400 is calculated. Translation data within the translation matrix is ​​acquired, and based on this data, the AGV's direction of movement in S400 and the angular relationship between the two radars are calculated.

[0061] As can be seen from S200 and S300, the two radars are already aligned; therefore, in S500, the two radars can be regarded as a whole.

[0062] S600, based on the angular relationship, corrects the first correction parameter and the first theoretical installation parameter to obtain the second correction parameter and the third correction parameter respectively.

[0063] In this embodiment, based on the angular relationship obtained in S500, the first correction parameter obtained in S300 is further corrected to obtain the second correction parameter. The first theoretical installation parameter obtained in S300 without any processing is also corrected to obtain the third correction parameter. The installation parameters of the two radars are then corrected again based on the obtained angular relationship to align the radar with the AGV vehicle's movement direction.

[0064] For example: when the first correction data P 11 When the first theoretical installation parameter P2 is the theoretical installation parameter of the second radar, and the first installation parameter P2 is the corrected installation parameter of the first radar, the first correction data P is adjusted according to the obtained angle relationship. 11 The first theoretical installation parameter P2 is then corrected to obtain the second corrected parameter P of the first radar. 12 And the third correction parameter P of the second radar 21 .

[0065] Or, when the first correction data P 21 The corrected installation parameters for the second radar are given. When the first theoretical installation parameter P1 is the theoretical installation parameter for the first radar, the first corrected data P is adjusted based on the obtained angular relationship. 21 The first theoretical installation parameter P1 is corrected to obtain the second corrected parameter P of the second radar. 22 And the third correction parameter P of the first radar 11 .

[0066] S700: Obtain the centroid parameters corresponding to the two radars, and calculate the installation error parameters corresponding to the two radars based on the second correction parameter, the third correction parameter, and the corresponding theoretical installation parameters. Correct the corresponding theoretical installation parameters based on the corresponding installation error parameters.

[0067] In this embodiment, the installation parameters, namely the second correction parameters and the third correction parameters, are obtained through the above steps S100 to S600 after aligning the two radars with each other and with the moving direction of the AGV. However, from a mechanical design perspective, the centroids of the two radar points are symmetrical to the theoretical center point of the vehicle body. That is, in the AGV field, the center point of the line connecting the installation points of the two radars is generally the center point of the AGV. Therefore, the centroids of the two radars can be taken as the control points. Thus, it is necessary to calculate the centroids of the radar control points.

[0068] Based on the second and third correction parameters obtained in S600, and the theoretical installation parameters P1 and P2 of the first and second radars obtained in S100, the centroids and errors of the two radars are calculated to obtain the first installation error parameter Δ1 of the first radar and the second installation error parameter Δ2 of the second radar. Using the obtained first and second installation error parameters Δ1 and Δ2, the theoretical installation parameters P1 and P2 of the first and second radars obtained in S100 are corrected. Therefore, the above installation parameters are obtained based on the theoretical installation parameters P1 and P2 of the first and second radars, and the obtained error values ​​are errors specific to the theoretical installation parameters P1 and P2 of the first and second radars. Therefore, it is necessary to correct the theoretical installation parameters based on the error values.

[0069] The method described in this invention allows for the recalculation of dual-radar installation parameters, reducing the workload of dual-radar installation, improving the accuracy of dual-radar vehicles, and enhancing the overall vehicle positioning accuracy. Compared to existing technologies, this invention also considers the influence of the vehicle's movement direction on the installation parameters of the two radars and corrects these parameters accordingly. Even AGVs of similar models may have slight differences in some mechanical parts, resulting in slight errors in the installation of dual radars on different AGVs. This application considers the relationship between the dual-radar installation parameters and the AGV's movement direction, making the two radars installed on the AGV more closely aligned with the actual operation of the AGV, thus improving the overall vehicle positioning accuracy.

[0070] In some embodiments of the present invention, step S200 specifically includes the following steps:

[0071] S210, Set the first calibration environment. The scanning areas of the two radars have overlapping areas, i.e., common viewing areas. The first calibration environment must be within the common viewing area.

[0072] In this embodiment, the AGV is moved to a location with strong feature surfaces or features are placed around the AGV. (See reference...) Figure 2 The first radar scanning area 300 and the first radar scanning area 400 overlap, forming a common viewing area 100. The feature surface 200 needs to be located within the common viewing area 100 of the AGV, with a distance between the AGV and the vehicle controlled between 1 meter and 10 meters. The distance can be adjusted according to the size of the feature surface 200. In other words, the first calibration environment is located within a range of 1 to 10 meters from the AGV. Feature surfaces that are too far away will amplify uneven ground during radar scanning, leading to calculation accuracy issues. Point cloud data within 1 meter will exhibit some distortion due to radar characteristics; therefore, a range with good radar data characteristics is chosen, specifically the 1-10 meter range.

[0073] S220: In the first calibration environment, one radar scans the first calibration environment to obtain the third point cloud data, and the remaining radar also scans the first calibration environment to obtain the third point cloud data.

[0074] In this embodiment, in the first calibration environment, the first radar scans the feature surface or feature to obtain the third point cloud data Q3, and the second radar scans the feature surface or feature to obtain the fourth point cloud data Q4. Alternatively, the second radar can scan the feature surface or feature to obtain the third point cloud data Q3, and the first radar can scan the feature surface or feature to obtain the fourth point cloud data Q4.

[0075] S230, based on a point cloud matching algorithm, uses one point cloud data as the reference data to process the other point cloud data and calculate the error matrix between the two point cloud data.

[0076] In this embodiment, point cloud data corresponding to two radar scans are acquired. One of the point cloud data is used as the reference data; that is, either the third point cloud data Q3 or the fourth point cloud data Q4 can be used as the reference data. Based on the point cloud matching algorithm, the other point cloud data is translated according to the reference data, and the error matrix when the overlap between the third point cloud data Q3 and the fourth point cloud data Q4 is the highest is calculated. The error matrix is ​​a transformation matrix T, which includes a translation matrix and a rotation matrix.

[0077] In some embodiments of the present invention, the process of obtaining the first correction parameter based on the error matrix in S300 specifically includes:

[0078] S310, based on the error matrix obtained in S200, perform rotation and translation calculations on the theoretical installation parameters corresponding to another point cloud data to obtain the first correction parameter.

[0079] In this embodiment, based on the error matrix obtained in S200, when the third point cloud data Q3 is used as the reference data, i.e., the point cloud data scanned by the first radar is used as the reference data, the theoretical installation parameters corresponding to the fourth point cloud data Q4 are rotated and translated, i.e., the theoretical installation parameters P2 of the second radar are corrected. This yields the first corrected data P. 21 The theoretical installation parameters of the first radar are not modified in any way, thus obtaining the first theoretical installation parameter P1.

[0080] When the fourth point cloud data Q4 is used as the reference data, the third point cloud data Q3 is rotated and translated. That is, when the point cloud data scanned by the second radar is used as the reference data, the theoretical installation parameters P1 of the first radar are corrected. This yields the first corrected data P. 11 The theoretical installation parameters of the second radar are not modified in any way, thus obtaining the first theoretical installation parameter P2.

[0081] The first correction parameter and the first theoretical installation parameter are the installation parameters of the two radars after the current correction. Through the above steps S210, S220, S230, and S310, based on the error matrix, the first radar and the second radar are aligned, and the relative installation parameters of the two radars are corrected to ensure the overlap between the two radars. That is, it can be considered that the radar coordinate systems of the first radar and the second radar are consistent at this time.

[0082] In some embodiments of the present invention, step S400 specifically includes the following steps:

[0083] S410, Establish the radar coordinate system, which includes: X degree of freedom, Y degree of freedom, and YAW degree of freedom.

[0084] In this embodiment, since the installation parameters between the two radars are corrected through the above steps S210, S220, S230, and S310, thereby aligning the two radars, the first radar and the second radar are already aligned in this embodiment. The coordinate systems of the first radar and the second radar are consistent. Therefore, the establishment of the radar coordinate system is not limited to which radar it is established on; it can be the coordinate system of the first radar or the coordinate system of the second radar. The X-degree of freedom, Y-degree of freedom, and YAW-degree of freedom can also be regarded as the X-axis, Y-axis, and YAW-axis.

[0085] S420: The AGV moves in the positive direction of the X-axis. After moving a certain distance, the AGV stops moving once the direction of all wheels on the AGV is consistent with the direction of movement.

[0086] In this embodiment, if there are no manufacturing errors, the X-axis of the radar coordinate system should be parallel to the direction of movement of the AGV. After the AGV moves forward a certain distance, ensure that the direction of all wheels on the AGV is consistent with the direction of movement of the AGV. When the two directions are consistent, the AGV stops moving and stays in place.

[0087] S430: After the AGV stops, a second calibration environment is constructed, and a radar scans to obtain the first point cloud data.

[0088] In this embodiment, after the AGV stops, a second calibration environment is constructed. This second calibration environment can be a location with a strong feature surface or features placed around the AGV. When a strong feature surface is used as the calibration environment, this feature surface can be located in the AGV's direction of movement. Based on the modified first correction parameters and first theoretical installation parameters from S210, S220, S230, and S310, one radar scans in the second calibration environment and records the first point cloud data Q1. In this embodiment, the prerequisite for one radar to scan is that the two radars are already aligned, and one of the theoretical installation parameters has been modified.

[0089] S440 moves again based on S430, in the same direction as S420, moving along the positive X-axis for the AGV to travel the set distance.

[0090] In this embodiment, the AGV moves in the positive direction of the X-axis, and the distance the AGV moves is set, with the value ranging from 0.8 meters to 1.5 meters.

[0091] In S450, when the AGV of S440 stops moving, another radar scans to acquire second point cloud data.

[0092] In this embodiment, after the AGV has moved the predetermined distance, another radar scans and acquires second point cloud data Q2. This radar can be either a first radar or a second radar.

[0093] In some embodiments of the present invention, step S500 specifically includes the following steps:

[0094] S510: Based on the first point cloud data Q1 and the second point cloud data Q2 obtained in S400, the translation matrix between the first point cloud data Q1 and the second point cloud data Q2 is calculated using a point cloud matching algorithm, and the first translation data and the second translation data are extracted from the translation matrix.

[0095] In this embodiment, the first point cloud data Q1 and the second point cloud data Q2 obtained through S400 are used to calculate the translation matrix between the first point cloud data Q1 and the second point cloud data Q2 through a point cloud matching algorithm. From the translation matrix, the Y-axis translation data and the X-axis translation data are extracted, that is, the first translation data and the second translation data are extracted.

[0096] It should be noted that when a rigid body moves in a plane, if the direction of rigid body motion is unique, the direction formed by the translation amount of the translation matrix is ​​equal to the direction of rigid body motion relative to the origin (0, 0) in the radar coordinate system.

[0097] S520, based on the angle relationship formula:

[0098]

[0099] ,in, For the first translation data, The second translation data is given, where θ is the angle difference between the YAW axis in the radar coordinate system and the direction of movement of the AGV.

[0100] In this embodiment, based on the first translation data and the second translation data obtained in S510, the angle difference θ between the YAW axis in the radar coordinate system and the moving direction of the AGV is obtained through the angle relationship formula.

[0101] In some embodiments of the present invention, S600 specifically includes:

[0102] S610: Based on the angle difference θ between the YAW axis and the moving direction of the AGV in the radar coordinate system obtained in S520, the first correction parameter is corrected to obtain the second correction parameter. The first theoretical installation parameter obtained in S300 is corrected to obtain the third correction parameter.

[0103] In this embodiment, the angle difference θ obtained in S520 is substituted into the first correction parameter for rotation transformation to obtain a second correction parameter parallel to the movement direction of the AGV vehicle. The angle difference θ obtained in S520 is then substituted into the first theoretical installation parameter for rotation transformation to obtain a third correction parameter parallel to the movement direction of the AGV vehicle.

[0104] For example: when the first correction data P 11 When the first theoretical installation parameter P2 is the theoretical installation parameter of the second radar, and the first installation parameter P2 is the corrected installation parameter of the first radar, the first correction data P is adjusted according to the obtained angle relationship. 11 The first theoretical installation parameter P2 is then corrected to obtain the second corrected parameter P of the first radar. 12 And the third correction parameter P of the second radar 21 .

[0105] Or, when the first correction data P 21 The corrected installation parameters for the second radar are given. When the first theoretical installation parameter P1 is the theoretical installation parameter for the first radar, the first corrected data P is adjusted based on the obtained angular relationship. 21 The first theoretical installation parameter P1 is corrected to obtain the second corrected parameter P of the second radar. 22 And the third correction parameter P of the first radar 11 .

[0106] The second and third correction parameters are the installation parameters of the two radars after the current correction. Using the above steps S410, S420, S430, S440, S450, S510, S520, and S610, based on the angular difference θ between the YAW axis in the radar coordinate system and the AGV's direction of movement, the second and third correction parameters, parallel to the AGV's body movement direction, are obtained. This aligns the two radars with the AGV's body movement direction, reducing the machining accuracy requirements of the radar mounting surface in the X, Y, and YAW degrees of freedom, and improving the overall vehicle positioning accuracy.

[0107] In some embodiments of the present invention, the process of obtaining the installation error parameters in S700 specifically includes the following steps:

[0108] S710: Obtain the centroid parameters of the first radar and the second radar. Based on the second correction parameters and the third correction parameters obtained in S610, subtract the corresponding centroid parameters of the first radar and the second radar to obtain the final first installation parameters of the first radar and the second installation parameters of the second radar.

[0109] In this embodiment, the first centroid parameter G1 of the first radar and the second centroid parameter G2 of the second radar are known from the mechanical design. Based on the second and third correction parameters obtained in S610, the corresponding first centroid parameter G1 and second centroid parameter G2 are subtracted. This yields the final first installation parameters of the first radar and the second installation parameters of the second radar.

[0110] For example: when P is the third correction parameter of the first radar 11 And the second correction parameter P of the second radar 22 At that time, the third correction parameter P 11 Subtracting the first centroid parameter G1, we obtain the first installation parameter P of the first radar. 12 The second correction parameter P 22 Subtracting the second centroid parameter G2, we obtain the second installation parameter P of the second radar. 23 .

[0111] Alternatively, when P is the second correction parameter of the first radar. 12 And the third correction parameter P of the second radar 21 At that time, the second correction parameter P will be... 12 Subtracting the first centroid parameter G1, we obtain the first installation parameter P of the first radar. 13 The third correction parameter P 21 Subtracting the second centroid parameter G2, we obtain the second installation parameter P of the second radar. 22 .

[0112] S720, the first installation error parameter Δ1 is obtained by subtracting the first installation parameter obtained in S710 from the theoretical installation parameter P1 of the first radar, and the second installation error parameter Δ2 is obtained by subtracting the second installation parameter obtained in S710 from the theoretical installation parameter P2 of the second radar.

[0113] In this embodiment, the installation parameters after aligning the two radars with each other and with the direction of vehicle movement are subtracted from the theoretical installation data P1 and P2 obtained from the mechanism to obtain error parameters Δ1 and Δ2. The obtained first installation error parameter Δ1 and second installation error parameter Δ2 are used to correct the theoretical installation parameters P1 and P2 of the first radar obtained in S100. Therefore, the above installation parameters are obtained based on the theoretical installation parameters P1 and P2 of the first radar, and the obtained error values ​​are errors specific to the theoretical installation parameters P1 and P2 of the first radar. Therefore, it is necessary to correct the theoretical installation parameters based on the error values.

[0114] In a specific embodiment of the present invention:

[0115] Obtain the theoretical installation parameters P1 of the first radar and the theoretical installation parameters P2 of the second radar;

[0116] To set up the first calibration environment, the scanning areas of the two radars must overlap, i.e., the common field of view area. The first calibration environment must be within the common field of view area.

[0117] In the first calibration environment, the first radar scans the feature surface or feature to obtain the third point cloud data Q3, and the second radar scans the feature surface or feature to obtain the fourth point cloud data Q4.

[0118] Based on the point cloud matching algorithm, the third point cloud data Q3 is used as the reference data, and the fourth point cloud data Q4 is processed to calculate the transformation matrix T when the overlap between the two point cloud data is the highest.

[0119] Based on the transformation matrix T, rotation and translation calculations are performed on the theoretical installation parameters P2 of the second radar to obtain the first correction parameter P. 21 .

[0120] A radar coordinate system is established, including X-degree of freedom, Y-degree of freedom, and YAW-degree of freedom. The AGV moves in the positive direction of the X-degree of freedom. After moving a certain distance, ensuring that the direction of all wheels on the AGV is consistent with the direction of movement, the AGV stops moving. A second calibration environment is then constructed based on the first correction parameter P. 21 Based on the theoretical installation parameters P1 of the first radar, the first radar performs a scan and obtains the first point cloud data Q1.

[0121] The AGV moves again in the positive direction of the X-axis, moving a set distance, where the set distance ranges from 0.8 meters to 1.5 meters. When the AGV stops moving, the second radar scans and acquires the second point cloud data Q2.

[0122] Using a point cloud matching algorithm, the translation matrix between the first point cloud data Q1 and the second point cloud data Q2 is calculated, and the first translation data is extracted from the translation matrix. Second translation data

[0123] Using the formula for angular relationships:

[0124]

[0125] The angle difference θ between the YAW axis and the moving direction of the AGV in the radar coordinate system is calculated.

[0126] Based on the angular difference θ between the YAW axis and the AGV's direction of movement in the radar coordinate system, the first correction parameter P is adjusted. 21 After correction, the second correction parameter P is obtained. 22 The first theoretical installation parameter P1 is corrected to obtain the third corrected parameter P. 11 .

[0127] From the mechanical design, we can know the first centroid parameter G1 of the first radar and the second centroid parameter G2 of the second radar, and then the third correction parameter P. 11 Subtracting the first centroid parameter G1, we obtain the first installation parameter P of the first radar. 12 The second correction parameter P 22 Subtracting the second centroid parameter G2, we obtain the second installation parameter P of the second radar. 23 ;

[0128] Set the first installation parameter P 12 The first installation error parameter Δ1 is obtained by subtracting the theoretical installation parameter P1 from the first radar. The second installation parameter P is then used. 23 The second installation error parameter Δ2 is obtained by subtracting the theoretical installation parameter P2 of the second radar.

[0129] The theoretical installation parameters P1 of the first radar are corrected according to the first installation error parameter Δ1, and the theoretical installation parameters P2 of the second radar are corrected according to the second installation error parameter Δ2.

[0130] According to an embodiment of a second aspect of the present invention, an electronic device includes: a memory for storing a program; and a processor for executing the program stored in the memory, wherein when the processor executes the program stored in the memory, the processor is configured to execute an AGV dual-radar installation error calibration method according to the first aspect.

[0131] The processor and memory can be connected via a bus or other means.

[0132] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs, such as the AGV dual-radar installation error calibration method described in the embodiments of the present invention. The processor implements the AGV dual-radar installation error calibration method of the first aspect of the present invention by running the non-transitory software program and instructions stored in the memory.

[0133] The memory may include a program storage area and a parameter storage area. The program storage area may store the operating system and application programs required for at least one function. The parameter storage area may store the above-described method for calibrating the installation error of the AGV dual radar. Furthermore, the memory may include high-speed random access memory and non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory may optionally include memory remotely located relative to the processor, which can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0134] The non-transient software program and instructions required to implement the above-described terminal selection method are stored in a memory. When executed by one or more processors, they are used to implement the AGV dual radar installation error calibration method of the first aspect of the present invention.

[0135] According to an embodiment of the third aspect of the present invention, the present invention also provides a storage medium storing computer-executable instructions for executing an AGV dual-radar installation error calibration method according to the first aspect.

[0136] It will be understood by those skilled in the art that all or some of the steps and systems in the methods disclosed above can be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components can be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, parameter structures, program modules, or other parameters). Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, as is known to those skilled in the art, communication media typically include computer-readable instructions, parameter structures, program modules, or other parameters in modulation parameter signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0137] The preferred embodiments of the present invention have been described in detail above, but the present invention is not limited to the embodiments described. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention, and these equivalent modifications or substitutions are all included within the scope defined by the claims of this application.

Claims

1. A method for calibrating the installation error of dual radars in an AGV, characterized in that, include: The theoretical installation parameters corresponding to the two radars are obtained respectively, wherein the two radars are installed diagonally on the AGV vehicle; Construct the first calibration environment, acquire point cloud data of corresponding scans from the two radars respectively, and calculate the error matrix based on one point cloud data. Based on the error matrix, the theoretical installation parameters corresponding to another point cloud data are corrected to obtain the first corrected parameters, and the first theoretical installation parameters corresponding to a point cloud data are obtained. When the AGV meets the set stopping conditions, a second calibration environment is constructed, and a first point cloud data of a radar scan is obtained. The AGV moves a set distance and obtains a second point cloud data of another radar scan. Using the first point cloud data and the second point cloud data, a translation matrix is ​​calculated, and based on the translation matrix, the angular relationship between the AGV's moving direction and the dual radars is obtained. Based on the angular relationship, the first correction parameter and the first theoretical installation parameter are corrected to obtain the second correction parameter and the third correction parameter, respectively. Obtain the centroid parameters corresponding to the two radars, and calculate the installation error parameters corresponding to the two radars based on the second correction parameter, the third correction parameter and the corresponding theoretical installation parameters. Correct the corresponding theoretical installation parameters based on the corresponding installation error parameters. The construction of the second calibration environment when the AGV meets the set stopping conditions specifically includes: A radar coordinate system is constructed, which includes: X degree of freedom, Y degree of freedom and YAW degree of freedom; The AGV is moved with the positive direction of the X degree of freedom as the direction of movement. The AGV stops moving when the direction of the wheels of the AGV is consistent with the direction of movement of the AGV. When the AGV stops moving, a second calibration environment is constructed.

2. The method for calibrating the installation error of dual radars in an AGV according to claim 1, characterized in that, The AGV trolley moves a set distance to acquire second point cloud data from another radar scan, specifically including: The AGV moves a set distance with the positive direction of the X-degree of freedom as the direction of movement; When the AGV stops moving, acquire second point cloud data from another radar scan.

3. The method for calibrating the installation error of dual radars in an AGV according to claim 1, characterized in that, The step of calculating a translation matrix using the first and second point cloud data, and obtaining the angular relationship between the AGV's movement direction and the dual radars based on the translation matrix, specifically includes: Using the first point cloud data and the second point cloud data, a translation matrix is ​​calculated based on a point cloud matching algorithm. The translation matrix includes: first translation data and second translation data. Using the formula for angular relationships: ， in, For the first translation data, Using the second translation data, the angular difference between the AGV's direction of movement and the YAW degree of freedom is calculated. .

4. The method for calibrating the installation error of dual radars in an AGV according to claim 3, characterized in that, The step of correcting the first correction parameter and the first theoretical installation parameter according to the angular relationship to obtain the second correction parameter and the third correction parameter specifically includes: Based on the angle difference, the first correction parameter and the first theoretical installation parameter are corrected respectively to obtain the second correction parameter and the third correction parameter.

5. The method for calibrating the installation error of dual radars in an AGV according to claim 1, characterized in that, The construction of the first calibration environment, acquiring point cloud data from the corresponding scans of the two radars, and calculating the error matrix based on one point cloud data point specifically includes: A first calibration environment is constructed, wherein the scanning areas of the dual radars have a common field of view, and the first calibration environment is located within the common field of view; In the first calibration environment, one radar scans to obtain the third point cloud data, and another radar scans to obtain the fourth point cloud data; Based on the point cloud matching algorithm, one point cloud data is used as a reference to process another point cloud data and calculate the error matrix.

6. The method for calibrating the installation error of dual radars in an AGV according to claim 1, characterized in that, The step of correcting the theoretical installation parameters corresponding to another point cloud data according to the error matrix to obtain the first correction parameter specifically includes: Based on the error matrix, rotation and translation calculations are performed on the theoretical installation parameters corresponding to another point cloud data to obtain the first correction parameters.

7. The method for calibrating the installation error of dual radars in an AGV according to claim 1, characterized in that, The process of obtaining the centroid parameters corresponding to the two radars, and obtaining the installation error parameters corresponding to the two radars based on the second correction parameter, the third correction parameter, and the corresponding theoretical installation parameters, specifically includes: Obtain the centroid parameters corresponding to the two radars, and subtract the corresponding centroid parameters from the second and third correction parameters to obtain the installation parameters corresponding to the two radars. The installation error parameters for the dual radars are obtained by subtracting the corresponding installation parameters from the corresponding theoretical installation parameters.

8. An electronic device, characterized in that, include: Memory, used to store programs; A processor is configured to execute a program stored in the memory. When the processor executes the program stored in the memory, the processor is configured to perform an AGV dual-radar installation error calibration method as described in any one of claims 1 to 7.

9. A storage medium, characterized in that, include: The device stores computer-executable instructions for performing an AGV dual-radar installation error calibration method as described in any one of claims 1 to 7.

Images