An external parameter calibration method, device, equipment and storage medium

Abstract

The present application discloses an external parameter calibration method, apparatus, device, and storage medium, including: obtaining first feature information and first pose information of a first sensor, and second feature information and second pose information of a second sensor, where the first feature information includes third feature information and fourth feature information, and the second feature information includes fifth feature information and sixth feature information; determining a first error parameter based on the first pose information, the second pose information, and an initial external parameter; when the first error parameter is greater than a first preset error parameter, determining a calibrated external parameter based on the third feature information and the fifth feature information; determining a second error parameter based on the fourth feature information, the sixth feature information, and the calibrated external parameter; and when the second error parameter is less than or equal to a second preset error parameter, outputting the calibrated external parameter. This external parameter calibration method can obtain data of multiple sensors in real time, online monitor and calibrate external parameters, and improve the accuracy of multi-sensor external parameter calibration.

Description

Technical Field

[0001] This application relates to the technical field of data processing, and particularly to an external parameter calibration method, device, equipment and storage medium. Background Art

[0002] Autonomous driving technology can use multi-sensor perception and artificial intelligence technology to collaboratively complete the vehicle's planning and control tasks, thereby achieving automatic control of the vehicle without the participation of a driver. SLAM, that is, simultaneous localization and mapping, is related to whether an autonomous driving vehicle can safely and accurately perform tasks in a specific scenario, and has a greater impact on autonomous driving technology.

[0003] Compared with the single sensor used in the SLAM system, using multi-sensor fusion localization technology can make the entire SLAM system have better robustness, and the accuracy of the external parameter calibration between multi-sensors directly affects the normal operation of the SLAM system.

[0004] Most existing SLAM systems can only complete the offline calibration of multi-sensor external parameters, and cannot solve problems such as pose offset generated during the operation of the sensor, resulting in low accuracy of the external parameters between the calibrated multi-sensors. Summary of the Invention

[0005] Embodiments of this application provide an external parameter calibration method, device, equipment and storage medium to improve the accuracy of external parameter calibration.

[0006] In a first aspect, embodiments of this application provide an external parameter calibration method, and the method includes:

[0007] Based on the data of the first sensor and the data of the second sensor, obtain the first feature information, the first pose information of the first sensor, the second feature information, and the second pose information of the second sensor, where the first feature information includes the third feature information and the fourth feature information, and the second feature information includes the fifth feature information and the sixth feature information;

[0008] Based on the first pose information, the second pose information, and the initial external parameters, determine the first error parameter;

[0009] When the first error parameter is greater than the first preset error parameter, based on the third feature information and the fifth feature information, determine the calibrated external parameters;

[0010] Based on the fourth feature information, the sixth feature information, and the calibrated external parameters, determine the second error parameter;

[0011] When the second error parameter is less than or equal to the second preset error parameter, output the calibrated external parameters.

[0012] In a possible implementation, determining the calibration extrinsic parameters based on the third feature information and the fifth feature information includes:

[0013] Performing feature verification on the third feature information and the fifth feature information;

[0014] When the third feature information meets the first feature verification condition and the fifth feature information meets the second feature verification condition, determining the calibration extrinsic parameters by using the hand-eye calibration method;

[0015] Otherwise, determining the calibration extrinsic parameters by using the mutual information MI method.

[0016] In a possible implementation, determining the calibration extrinsic parameters by using the MI method includes:

[0017] Based on the third feature information, determining the gray probability distribution of the first sensor;

[0018] Based on the fifth feature information and the initial extrinsic parameters, determining the intensity probability distribution of the second sensor;

[0019] Based on the gray probability distribution and the intensity probability distribution, determining the MI, and performing iterative solution on the MI to determine the maximum value of the MI;

[0020] Based on the maximum value of the MI, determining the calibration extrinsic parameters.

[0021] In a possible implementation, determining the first error parameter based on the first pose information, the second pose information, and the initial extrinsic parameters includes:

[0022] Based on the first pose information and the second pose information, determining the pose matrix;

[0023] Based on the pose matrix and the initial extrinsic parameters, determining the first error parameter.

[0024] In a possible implementation, determining the second error parameter based on the fourth feature information, the sixth feature information, and the calibration extrinsic parameters includes:

[0025] Using the random sample consensus RANSAC method to form feature pairs from the fourth feature information and the sixth feature information, where the feature pairs include target fourth feature information and target sixth feature information;

[0026] Based on the target fourth feature information, the target sixth feature information, and the calibration extrinsic parameters, determining the second error parameter.

[0027] Second aspect, an external parameter calibration device is provided in an embodiment of the present application. The device includes: an acquisition module, a first error verification module, an external parameter calibration module, a second error verification module, and an output module;

[0028] The acquisition module is configured to obtain first feature information, first pose information of the first sensor, second feature information, and second pose information of the second sensor based on data of the first sensor and data of the second sensor. Wherein, the first feature information includes third feature information and fourth feature information, and the second feature information includes fifth feature information and sixth feature information;

[0029] The first error verification module is configured to determine a first error parameter based on the first pose information, the second pose information, and an initial external parameter;

[0030] The external parameter calibration module is configured to determine a calibrated external parameter based on the third feature information and the fifth feature information when the first error parameter is greater than a first preset error parameter;

[0031] The second error verification module is configured to determine a second error parameter based on the fourth feature information, the sixth feature information, and the calibrated external parameter;

[0032] The output module is configured to output the calibrated external parameter when the second error parameter is less than or equal to a second preset error parameter.

[0033] In a possible implementation manner, the external parameter calibration module is specifically configured to perform feature verification on the third feature information and the fifth feature information;

[0034] When the third feature information meets a first feature verification condition and the fifth feature information meets a second feature verification condition, the calibrated external parameter is determined by using a hand-eye calibration method;

[0035] Otherwise, the calibrated external parameter is determined by using a mutual information MI method.

[0036] In a possible implementation manner, the second error verification module is specifically configured to use a random sample consensus RANSAC method to form feature pairs from the fourth feature information and the sixth feature information. Wherein, the feature pairs include target fourth feature information and target sixth feature information;

[0037] The second error parameter is determined based on the target fourth feature information, the target sixth feature information, and the calibrated external parameter.

[0038] Third aspect, an external parameter calibration device is provided in an embodiment of the present application. The device includes: a memory and a processor;

[0039] The memory is used to store relevant program codes;

[0040] The processor is used to call the program codes to execute the external parameter calibration method according to any implementation manner of the first aspect above.

[0041] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, which is used to store a computer program, and the computer program is used to execute the external parameter calibration method according to any implementation manner of the first aspect above.

[0042] Thus, the embodiments of the present application have the following beneficial effects:

[0043] In the above implementation manner of the embodiment of the present application, based on the data of the first sensor and the data of the second sensor, the first feature information, the first pose information of the first sensor, the second feature information, and the second pose information of the second sensor are obtained. Among them, the first feature information includes the third feature information and the fourth feature information, and the second feature information includes the fifth feature information and the sixth feature information; based on the first pose information, the second pose information, and the initial external parameters, the first error parameter is determined; when the first error parameter is greater than the first preset error parameter, based on the third feature information and the fifth feature information, the calibrated external parameters are determined; based on the fourth feature information, the sixth feature information, and the calibrated external parameters, the second error parameter is determined; when the second error parameter is less than or equal to the second preset error parameter, the calibrated external parameters are output. The external parameter calibration method provided by the embodiment of the present application can obtain the data of multiple sensors in real time and monitor and calibrate the external parameters in real time during the operation of the autonomous vehicle, improving the accuracy of the external parameter calibration of multiple sensors. Description of the Drawings

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments provided in the present application. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0045] Figure 1 It is a flowchart of an external parameter calibration method in an embodiment of the present application;

[0046] Figure 2 It is a schematic diagram of an external parameter calibration device in an embodiment of the present application;

[0047] Figure 3 It is a schematic diagram of an external parameter calibration device in an embodiment of the present application. Detailed Embodiments

[0048] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. The described embodiments are only exemplary implementation manners of the present application, not all implementation manners. Those skilled in the art can obtain other embodiments without creative work in combination with the embodiments of the present application, and these embodiments are also within the protection scope of the present application.

[0049] The problem of Simultaneous Localization And Mapping (SLAM) can be described as follows: When a robot is placed at an unknown location in an unknown environment, is there a way for the robot to gradually depict a complete map of this environment while moving? A complete map means being able to move to every accessible corner of the room without being blocked. That is, in an unknown environment, starting from an unknown location and moving, the robot locates itself based on its position and the map during the movement, and at the same time builds an incremental map based on its self-localization to achieve autonomous positioning and navigation.

[0050] When calibrating the external parameters of a multi-sensor fusion SLAM system, the commonly used sensors are cameras and lidar. When the camera and lidar return the collected data, they are both based on their own built-in coordinate systems. Therefore, the problem of calibrating the external parameters between the lidar and the camera can be described as the rotation and translation between two coordinate systems.

[0051] Most existing SLAM systems can only complete the offline calibration of the external parameters of multiple sensors and cannot solve problems such as pose offsets generated during the operation of the sensors, resulting in low accuracy of the external parameters between the calibrated multiple sensors.

[0052] Pose represents position and orientation. Any rigid body in a spatial coordinate system (O-XYZ) can be uniquely represented by its position and orientation. Among them, position: the x, y, and z coordinates, and orientation: the angle rx between the rigid body and the OX axis, the angle ry between the rigid body and the OY axis, and the angle rz between the rigid body and the OZ axis.

[0053] Based on this, the embodiments herein provide an external parameter calibration method to improve the accuracy of external parameter calibration. Specifically, in implementation, first, based on the data of the first sensor and the data collected by the second sensor, the first feature information, the first pose information of the first sensor, the second feature information, and the second pose information of the second sensor are obtained. Among them, the first feature information includes the third feature information and the fourth feature information, and the second feature information includes the fifth feature information and the sixth feature information. Based on the first pose information, the second pose information, and the initial external parameters, the first error parameter is determined. When the first error parameter is greater than the first preset error parameter, the calibrated external parameters are determined based on the third feature information and the fifth feature information. Based on the fourth feature information, the sixth feature information, and the calibrated external parameters, the second error parameter is determined. When the second error parameter is less than or equal to the second preset error parameter, the calibrated external parameters are output. The external parameter calibration method provided by the embodiments of the present application can, when the autonomous driving vehicle is working, obtain the data of multiple sensors in real time, and monitor and calibrate the external parameters in real time online, improving the accuracy of multi-sensor external parameter calibration.

[0054] Next, the external parameter calibration method provided by the embodiments of the present application will be described with reference to the accompanying drawings.

[0055] See Figure 1 , Figure 1 which is a flowchart of an external parameter calibration method in the embodiments of the present application.

[0056] The method specifically includes the following steps:

[0057] S101: Based on the data of the first sensor and the data of the second sensor, obtain the first feature information, the first pose information of the first sensor, the second feature information, and the second pose information of the second sensor. Among them, the first feature information includes the third feature information and the fourth feature information, and the second feature information includes the fifth feature information and the sixth feature information.

[0058] In the embodiments of the present application, the first sensor is taken as a camera as an example, and the second sensor is taken as a lidar as an example for illustration.

[0059] In a possible implementation, the data collected by the first sensor and the second sensor can be input into the SLAM function module of the autonomous driving system. The SLAM function module processes the data to obtain the first feature information, the first pose information of the first sensor, the second feature information, and the second pose information of the second sensor. Among them, the first feature information includes the third feature information and the fourth feature information, and the second feature information includes the fifth feature information and the sixth feature information.

[0060] In this embodiment, the third feature information and the fifth feature information are low-level feature information, the fourth feature information and the sixth feature information are high-level feature information. The third feature information of the first sensor includes: feature points (corner points), and the corner points are extreme points, that is, points that are particularly prominent in certain aspects. The fourth feature information includes: structural line features, that is, the edge contour line features of buildings or other objects in the natural environment. The fifth feature information of the second sensor includes: feature points (corner points), and the corner points are extreme points, that is, points that are particularly prominent in certain aspects. The sixth feature information envelope: structural line features.

[0061] S102: Based on the first pose information, the second pose information, and the initial extrinsic parameters, determine the first error parameter.

[0062] In a possible implementation manner, the first error verification module can perform pose alignment based on the first pose information, the second pose information, and the initial extrinsic parameters to determine the first error parameter.

[0063] Specifically, based on the first pose information and the second pose information, determine the pose matrix; then, based on the pose matrix and the initial extrinsic parameters, determine the first error parameter.

[0064] In this embodiment, let T represent the initial extrinsic parameters, pose_1 represent the first pose information, which can be denoted as pose_2 represent the second pose information, which can be denoted as E1 represent the first error parameter. This embodiment provides a calculation method for determining the first error parameter, that is, based on the first pose information and the second pose information, determine the pose matrix as Then, based on the pose matrix and the initial extrinsic parameters, determine the first error parameter

[0065] S103: When the first error parameter is greater than the first preset error parameter, based on the third feature information and the fifth feature information, determine the calibrated extrinsic parameters.

[0066] When the first error parameter is less than or equal to the first preset error parameter, it indicates that the initial extrinsic parameters are within the allowable error range and there is no need to recalibrate the initial extrinsic parameters; when the first error parameter is greater than the first preset error parameter, it is necessary to recalibrate the initial extrinsic parameters, and based on the third feature information of the first sensor and the fifth feature information of the second sensor, determine the calibrated extrinsic parameters.

[0067] Since the pose information includes position information and attitude information, when verifying the first error parameter, the error parameter matrix can be divided into two dimensions, that is, displacement error and angle error, and respectively determine whether the displacement error and the angle error exceed the preset error values.

[0068] In a possible implementation, feature verification is performed on the third feature information and the fifth feature information. When the third feature information meets the first feature verification condition and the fifth feature information meets the second feature verification condition, the hand-eye calibration method is used to determine the calibration extrinsic parameters; otherwise, the method of mutual information (MI) is used to determine the calibration extrinsic parameters.

[0069] In specific implementation, the third feature information is quantified, and it is judged whether the number of feature points after quantization is greater than or equal to 200 feature points per 640*480 pixel image. If so, the third feature information meets the first feature verification condition; the fifth feature information is quantified, and it is judged whether the number of features after quantization is greater than 50 features included in each frame of point cloud of a lidar with 64 beams and a resolution of 0.2*0.2 degrees. If so, the fifth feature information meets the second feature verification condition.

[0070] When the third feature information meets the first feature verification condition and the fifth feature information meets the second feature verification condition, the hand-eye calibration method is used to determine the calibration extrinsic parameters. If at least one of the above two feature verification conditions is not met, the MI method based on intensity and gray value is used to determine the calibration extrinsic parameters.

[0071] Next, the principle of hand-eye calibration will be introduced in combination with specific embodiments. For example, when a person's eyes see an object and wants to grab it with their hand, the brain needs to know the coordinate relationship between the eyes and the hand. If the brain is compared to B, the eyes are compared to A, and the hand is compared to C, if the relationship between A and B is known and the relationship between B and C is known, then the relationship between C and A can be known, that is, the coordinate relationship between the hand and the eye.

[0072] MI is a useful information measure in information theory, which can be regarded as the amount of information contained in one random variable about another random variable, or the reduction in uncertainty of a random variable due to knowing another random variable.

[0073] When using the MI method to determine the calibration extrinsic parameters, first, based on the third feature information, the gray probability distribution of the first sensor is determined, that is, its gray probability distribution is statistically calculated according to the image data; based on the fifth feature information and the initial extrinsic parameters, the intensity probability distribution of the second sensor is determined, that is, the point cloud data is projected onto the pixel coordinates through the initial extrinsic parameters, and its intensity probability distribution is statistically calculated; the value of MI is determined based on the obtained gray probability distribution and intensity probability distribution, and MI is iteratively solved to determine the maximum value of MI; the extrinsic parameters corresponding to when MI is the maximum value are the calibration extrinsic parameters.

[0074] It should be noted that the calibration extrinsic parameters determined by the above method of the embodiment are a rough calibration method, and the extrinsic parameters of the rough calibration still need to be error-checked below.

[0075] S104: Determine the second error parameter based on the fourth feature information, the sixth feature information, and the calibrated extrinsic parameters.

[0076] After obtaining the calibrated extrinsic parameters, determine the second error parameter based on the fourth feature information, the sixth feature information, and the calibrated extrinsic parameters.

[0077] In a possible implementation, use the Random Sample Consensus (RANSAC) method to form a feature pair from the fourth feature information and the sixth feature information, where the feature pair includes the target fourth feature information and the target sixth feature information; determine the second error parameter based on the target fourth feature information, the target sixth feature information, and the calibrated extrinsic parameters.

[0078] The RANSAC method is an algorithm that calculates the mathematical model parameters of the data based on a sample data set containing abnormal data to obtain the effective sample data.

[0079] Specifically, use the K-NearestNeighbor (KNN) classification algorithm to classify the fourth feature information and the sixth feature information, and form a feature pair from the target fourth feature information and the target sixth feature information.

[0080] Let E2 represent the second error parameter, feature_1 represent the target fourth feature information, feature_2 represent the target sixth feature information, and T_fine represent the calibrated extrinsic parameters. Then, in this embodiment, one way to calculate the second error parameter is: E2 = T_fine * feature_1 - feature_2.

[0081] S105: When the second error parameter is less than or equal to the second preset error parameter, output the calibrated extrinsic parameters.

[0082] When the second error parameter is less than or equal to the second preset error parameter, it indicates that the calibration parameters meet the requirements of fine calibration, and the calibrated extrinsic parameters are output. If the second error parameter is greater than the second preset error parameter, it indicates that the calibrated extrinsic parameters do not meet the requirements of fine calibration, and it is necessary to re-obtain the data of the first sensor and the second sensor to re-calibrate the initial extrinsic parameters.

[0083] The extrinsic parameter calibration method provided by the embodiments of the present application can obtain the data of multiple sensors in real time and monitor and calibrate the extrinsic parameters in real time online when the autonomous vehicle is working, improving the accuracy and real-time performance of the extrinsic parameter calibration of multiple sensors.

[0084] Although the operations are depicted in a particular order, this should not be construed as requiring that the operations be performed in the particular order shown or in sequential order. In certain circumstances, multitasking and parallel processing may be advantageous. It should be understood that the various steps recited in the method embodiments of the present application may be performed in a different order and / or in parallel. In addition, method embodiments may include additional steps and / or omit performing the steps shown. The scope of protection of the present application is not limited in this regard.

[0085] Based on the above method embodiments, an external parameter calibration device is further provided in an embodiment of the present application. Refer to Figure 2 , Figure 2 which is a schematic diagram of the external parameter calibration device in the embodiment of the present application.

[0086] As Figure 2 shown, the device 200 includes an acquisition module 201, a first error verification module 202, an external parameter calibration module 203, a second error verification module 204, and an output module 205;

[0087] The acquisition module 201 is configured to acquire first feature information, first pose information of the first sensor, second feature information, and second pose information of the second sensor based on data of the first sensor and data of the second sensor, where the first feature information includes third feature information and fourth feature information, and the second feature information includes fifth feature information and sixth feature information;

[0088] The first error verification module 202 is configured to determine a first error parameter based on the first pose information, the second pose information, and the initial external parameters;

[0089] The external parameter calibration module 203 is configured to determine calibrated external parameters based on the third feature information and the fifth feature information when the first error parameter is greater than a first preset error parameter;

[0090] The second error verification module 204 is configured to determine a second error parameter based on the fourth feature information, the sixth feature information, and the calibrated external parameters;

[0091] The output module 205 is configured to output the calibrated external parameters when the second error parameter is less than or equal to a second preset error parameter.

[0092] The first error verification module 202 is specifically configured to determine a pose matrix based on the first pose information and the second pose information; and determine the first error parameter based on the pose matrix and the initial external parameters.

[0093] The external parameter calibration module 203 is specifically configured to perform feature verification on the third feature information and the fifth feature information; when the third feature information meets the first feature verification condition and the fifth feature information meets the second feature verification condition, the hand-eye calibration method is used to determine the calibrated external parameters; otherwise, the mutual information MI method is used to determine the calibrated external parameters.

[0094] The external parameter calibration module 203 is specifically configured to determine the gray probability distribution of the first sensor based on the third feature information; determine the intensity probability distribution of the second sensor based on the fifth feature information and the initial external parameters; determine the MI based on the gray probability distribution and the intensity probability distribution, perform iterative solution on the MI to determine the maximum value of the MI; determine the calibrated external parameters based on the maximum value of the MI.

[0095] The second error verification module 204 is specifically configured to use the random sample consensus RANSAC method to form feature pairs from the fourth feature information and the sixth feature information, where the feature pairs include the target fourth feature information and the target sixth feature information; determine the second error parameter based on the target fourth feature information, the target sixth feature information, and the calibrated external parameters.

[0096] For the beneficial effects of the external parameter calibration device provided in the embodiments of the present application, refer to the above method embodiments, which will not be elaborated here.

[0097] Based on the above method embodiments and device embodiments, the embodiments of the present application also provide an external parameter calibration device, refer to Figure 3 , Figure 3 is a schematic diagram of an external parameter calibration device in the embodiments of the present application,

[0098] The device 300 includes: a memory 301 and a processor 302;

[0099] The memory 301 is used to store relevant program codes;

[0100] The processor 302 is used to call the program codes to execute the external parameter calibration method described in the above method embodiments.

[0101] In addition, the embodiments of the present application also provide a computer-readable storage medium, and the computer-readable storage medium is used to store a computer program, and the computer program is used to execute the external parameter calibration method described in the above method embodiments.

[0102] It should be noted that the terms "first" and "second" used in the present application are used to distinguish similar objects, rather than to describe a specific order or sequence. It should be understood that such terms can be interchanged under appropriate circumstances, which is only a way of distinguishing objects with the same attributes when describing the embodiments of the present application.

[0103] The modules involved in the embodiments described in this application can be implemented in software or in hardware. In some cases, the name of a module does not constitute a limitation on the module itself.

[0104] It should be noted that the computer-readable medium described in this application can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of a computer-readable storage medium can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this application, a computer-readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in conjunction with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, and this computer-readable signal medium can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on a computer-readable medium can be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0105] Each embodiment in this specification is described in a progressive manner. Similar parts between each embodiment can be referred to each other, and the key point of each embodiment is to illustrate the differences from other embodiments. In particular, for the apparatus embodiments, since they are basically similar to the method embodiments, they are described relatively simply, and the relevant parts can be referred to the corresponding descriptions in the method embodiments. The apparatus embodiments described above are merely illustrative. The units or modules described as separate components may or may not be physically separated, and the components shown as units or modules may or may not be physical modules, that is, they can be located in one place or distributed to multiple network units. Some or all of the units or modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. A person of ordinary skill in the art can understand and implement it without creative efforts.

[0106] The above are only exemplary embodiments of the present application and do not impose any formal restrictions on the present application. Equivalent changes or modifications made to the above embodiments all fall within the protection scope of the present application.

Claims

1. An external parameter calibration method, characterized in that, The method includes: Based on the data of the first sensor and the data of the second sensor, obtain the first feature information, the first pose information of the first sensor, the second feature information, and the second pose information of the second sensor. Among them, the first feature information includes the third feature information and the fourth feature information, and the second feature information includes the fifth feature information and the sixth feature information; the third feature information and the fifth feature information are low-level feature information, and the fourth feature information and the sixth feature information are high-level feature information; the third feature information includes feature points; the fourth feature information includes structural line features; the fifth feature information includes feature points; the sixth feature information includes structural line features; Based on the first pose information, the second pose information, and the initial extrinsic parameters, determine the first error parameter; When the first error parameter is greater than the first preset error parameter, based on the third feature information and the fifth feature information, determine the calibration extrinsic parameters; Based on the fourth feature information, the sixth feature information, and the calibration extrinsic parameters, determine the second error parameter; When the second error parameter is less than or equal to the second preset error parameter, output the calibration extrinsic parameters.

2. The method according to claim 1, wherein The determining the calibration extrinsic parameters based on the third feature information and the fifth feature information includes: Perform feature verification on the third feature information and the fifth feature information; When the third feature information meets the first feature verification condition and the fifth feature information meets the second feature verification condition, use the hand-eye calibration method to determine the calibration extrinsic parameters; Otherwise, use the mutual information MI method to determine the calibration extrinsic parameters.

3. The method according to claim 2, characterized in that, The using the MI method to determine the calibration extrinsic parameters includes: Based on the third feature information, determine the gray probability distribution of the first sensor; Based on the fifth feature information and the initial extrinsic parameters, determine the intensity probability distribution of the second sensor; Based on the gray probability distribution and the intensity probability distribution, determine the MI, perform iterative solution on the MI, and determine the maximum value of the MI; Based on the maximum value of the MI, determine the calibration extrinsic parameters.

4. The method according to claim 1, characterized in that, The determining the first error parameter based on the first pose information, the second pose information, and the initial extrinsic parameters includes: Based on the first pose information and the second pose information, determine the pose matrix; Based on the pose matrix and the initial extrinsic parameters, determine the first error parameter.

5. The method according to claim 1, characterized in that, The determining the second error parameter based on the fourth feature information, the sixth feature information, and the calibration extrinsic parameters includes: Use the random sample consensus RANSAC method to form feature pairs from the fourth feature information and the sixth feature information, where the feature pairs include target fourth feature information and target sixth feature information; Based on the target fourth feature information, the target sixth feature information, and the calibration extrinsic parameters, determine the second error parameter.

6. An external parameter calibration device, characterized in that, The device includes: an acquisition module, a first error verification module, an extrinsic parameter calibration module, a second error verification module, and an output module; The acquisition module is configured to acquire first feature information, first pose information of the first sensor, second feature information, and second pose information of the second sensor based on data of the first sensor and data of the second sensor, where the first feature information includes third feature information and fourth feature information, and the second feature information includes fifth feature information and sixth feature information; the third feature information and the fifth feature information are low-level feature information, and the fourth feature information and the sixth feature information are high-level feature information; the third feature information includes feature points; the fourth feature information includes structure line features; the fifth feature information includes feature points; the sixth feature information includes structure line features; The first error verification module is configured to determine a first error parameter based on the first pose information, the second pose information, and an initial external parameter; The external parameter calibration module is configured to determine a calibrated external parameter based on the third feature information and the fifth feature information when the first error parameter is greater than a first preset error parameter; The second error verification module is configured to determine a second error parameter based on the fourth feature information, the sixth feature information, and the calibrated external parameter; The output module is configured to output the calibrated external parameter when the second error parameter is less than or equal to a second preset error parameter.

7. The device according to claim 6, characterized in that, The external parameter calibration module is specifically configured to perform feature verification on the third feature information and the fifth feature information; When the third feature information meets a first feature verification condition and the fifth feature information meets a second feature verification condition, the calibrated external parameter is determined by using a hand-eye calibration method; Otherwise, the calibrated external parameter is determined by using the mutual information MI method.

8. The device according to claim 6, characterized in that, The second error verification module is specifically configured to use the random sample consensus (RANSAC) method to form feature pairs from the fourth feature information and the sixth feature information, where the feature pairs include target fourth feature information and target sixth feature information; Based on the target fourth feature information, the target sixth feature information, and the calibrated external parameter, the second error parameter is determined.

9. An external parameter calibration device, characterized in that, The device includes: a memory and a processor; The memory is configured to store relevant program codes; The processor is configured to call the program codes to execute the external parameter calibration method according to any one of claims 1 to 5.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium is configured to store a computer program, and the computer program is configured to execute the external parameter calibration method according to any one of claims 1 to 5.

Images