Parameter calibration method, controller, drone detection system and storage medium

Abstract

The present invention relates to the field of unmanned aerial vehicle technology, and in particular to a parameter calibration method, a controller, an unmanned aerial vehicle detection system and a storage medium. The method is applied to the controller, and the method includes: obtaining target data of a radar vision system, the radar vision system is connected to the controller in communication, and includes a radar device and a camera device. The target data includes azimuth data and pitch angle data of the radar device, and azimuth data and pitch angle data of the camera device. Based on the target data, the target azimuth difference and target pitch angle difference of the radar device and the camera device are calculated, and the target azimuth difference and target pitch angle difference are sent to the radar vision system, so that the radar vision system replaces the current azimuth difference and pitch angle difference with the target azimuth difference and target pitch angle difference. The embodiment of the present invention can realize automatic calibration of the radar vision system, save time and labor, improve calibration efficiency, and ensure the accuracy of the calibration result.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicles, and in particular to a parameter calibration method, a controller, an unmanned aerial vehicle detection system and a storage medium. Background Art

[0002] With the development of Internet technology and drone technology, the threat of drones has become increasingly prominent. Among them, the radar vision system, as a means of anti-drone detection technology, can be used for anti-drone detection. The radar vision system is a detection system composed of phased array radar and PTZ camera equipment. Phased array radar has the advantages of fast scanning and high resolution, and PTZ camera equipment can track drones through visual lock, with advantages such as clear and intuitive observation.

[0003] In existing radar vision systems, calibration of phased array radar and PTZ camera equipment often requires professionals to use professional equipment to perform a lot of manual operations, which takes a lot of time and is inefficient. There may be inaccurate calibration results due to accidental factors, which reduces the performance of the radar vision system and the accuracy of drone detection results. Summary of the invention

[0004] In view of this, an embodiment of the present invention provides a parameter calibration method, a controller, a UAV detection system and a storage medium, aiming to solve the technical problem of calibrating a radar vision system in the prior art that it takes a lot of time and is inefficient.

[0005] To solve the above technical problems, the embodiments of the present invention provide the following technical solutions:

[0006] In a first aspect, an embodiment of the present invention provides a parameter calibration method, which is applied to a controller, and the method includes:

[0007] Acquiring target data of a radar vision system, wherein the radar vision system is communicatively connected to the controller and comprises a radar device and a camera device, and the target data comprises azimuth data and elevation data of the radar device, and azimuth data and elevation data of the camera device;

[0008] Calculating a target azimuth difference and a target elevation difference between the radar device and the camera device based on the target data;

[0009] The target azimuth angle difference and the target elevation angle difference are sent to the radar vision system, so that the radar vision system replaces the current azimuth angle difference and the current elevation angle difference with the target azimuth angle difference and the target elevation angle difference.

[0010] In some embodiments, acquiring target data of a radar vision system includes:

[0011] Acquire a first quantity of first azimuth angle data and first elevation angle data from the radar device, and acquire a second quantity of second azimuth angle data and second elevation angle data from the camera device;

[0012] According to the timestamps of the azimuth data and the elevation data, the first azimuth data and the second azimuth data with the same timestamp are combined into a data pair, and the first pitch angle data and the second pitch angle data with the same timestamp are combined into a data pair to obtain the target data, which includes an azimuth data pair and an elevation data pair.

[0013] In some embodiments, the step of calculating the target azimuth difference and the target elevation difference between the radar device and the camera device based on the target data includes:

[0014] A linear regression algorithm is used to fit the azimuth angle data pair and the pitch angle data pair to obtain a target azimuth angle difference and a target pitch angle difference. The linear regression algorithm includes one or more of a least squares algorithm, a weighted least squares algorithm, a minimum absolute deviation regression algorithm, a ridge regression algorithm, an elastic net regression algorithm, and a lasso regression algorithm.

[0015] In some embodiments, acquiring a first quantity of first azimuth angle data and first elevation angle data from the radar device comprises:

[0016] Extracting first azimuth angle data and first elevation angle data based on a standard target object in each frame of radar data from a plurality of frames of radar data detected and acquired by the radar device, until a first number of first azimuth angle data and first elevation angle data are extracted;

[0017] The step of acquiring a second amount of second azimuth angle data and second pitch angle data from the camera device comprises:

[0018] From the plurality of frames of image data captured by the camera device, second azimuth angle data and second elevation angle data based on the standard target object are extracted from each frame of image data until a second number of second azimuth angle data and second elevation angle data are extracted.

[0019] In some embodiments, after acquiring a first quantity of first azimuth angle data and first elevation angle data from the radar device, and acquiring a second quantity of second azimuth angle data and second elevation angle data from the camera device, acquiring target data of the radar vision system further includes:

[0020] Filtering target angle data that meets a data screening condition from the first azimuth angle data and the first elevation angle data of the radar device, the data screening condition comprising an azimuth angle being less than or equal to a first angle threshold and an elevation angle being less than or equal to a second angle threshold, the target angle data comprising azimuth angle data and elevation angle data;

[0021] With the timestamp of the target angle data as the horizontal axis and the target angle data as the vertical axis, a target function corresponding to the target angle data is constructed, and the target angle data of the target function is smoothed by exponential weighted averaging to obtain the target azimuth angle data and target elevation angle data of the radar device.

[0022] In some embodiments, the acquiring target data of the radar vision system further includes:

[0023] The target azimuth data and the target elevation data of the radar device are normalized to obtain first target data, wherein the first target data is the azimuth data and the elevation data of the radar device after screening, smoothing and normalization.

[0024] In some embodiments, the acquiring target data of the radar vision system further includes:

[0025] The second azimuth angle data and the second elevation angle data of the camera device are normalized to obtain second target data, wherein the second target data is the normalized azimuth angle data and the elevation angle data of the camera device.

[0026] In a second aspect, an embodiment of the present invention provides a controller, including:

[0027] A processor and a memory communicatively connected to the processor;

[0028] The memory stores computer program instructions executable by the processor, and when the computer program instructions are called by the processor, the processor executes any one of the parameter calibration methods described in the first aspect, or executes the steps in any one of the implementations of any one of the parameter calibration methods described in the first aspect.

[0029] In a third aspect, an embodiment of the present invention provides a drone detection system, comprising:

[0030] The controller according to the second aspect; and

[0031] The radar vision system is connected to the controller in communication, and includes a radar device and a camera device. The radar device is used to detect the target object to obtain radar data, and the camera device is used to detect the target object to obtain image data.

[0032] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which computer program instructions executable by a processor are stored, and when the computer program instructions are called by the processor, the processor executes any one of the parameter calibration methods described in the first aspect, or executes the steps in any one of the implementations of any one of the parameter calibration methods described in the first aspect.

[0033] The embodiments of the present invention have the following beneficial effects: Different from the prior art, the parameter calibration method provided by the embodiments of the present invention is applied to a controller, and the method includes: acquiring target data of a radar vision system, wherein the radar vision system is communicatively connected to the controller, and includes a radar device and a camera device, and the target data includes azimuth data and elevation data of the radar device, and azimuth data and elevation data of the camera device; based on the target data, a target azimuth difference and a target elevation difference of the radar device and the camera device are calculated, and the target azimuth difference and the target elevation difference are sent to the radar vision system, so that the radar vision system replaces the current azimuth difference and the elevation difference with the target azimuth difference and the target elevation difference.

[0034] The embodiment of the present invention obtains the azimuth data and elevation data of the radar device and the camera device in the radar vision system, calculates the target azimuth difference and the target elevation difference of the radar device and the camera device, and enables the radar device and the camera device to adjust the current azimuth difference and the elevation difference to the target azimuth difference and the target elevation difference, respectively, thereby realizing automatic calibration of the radar vision system, saving time and labor, improving calibration efficiency, and ensuring the accuracy of the calibration result. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following is a brief introduction to the drawings required for the description of the prior art or the embodiments. Obviously, the drawings described below only illustrate certain embodiments of the present invention and should not be regarded as limiting the scope of protection. For those of ordinary skill in the art, other related drawings can also be obtained based on these drawings without creative work.

[0036] Figure 1 is a schematic diagram of an application scenario of a parameter calibration method provided by some embodiments of the present invention;

[0037] Figure 2 is a schematic diagram of the structure of a drone detection system provided by some embodiments of the present invention;

[0038] Figure 3 is a schematic diagram of the structure of a controller provided by some embodiments of the present invention;

[0039] Figure 4is a flow chart of a parameter calibration method provided by some embodiments of the present invention;

[0040] Figure 5 is a schematic diagram of the relative position relationship (azimuth) between a target object and a radar device and a camera device in some embodiments of the present invention;

[0041] Figure 6 is a schematic diagram of the relative position relationship (pitch angle) between the target object and the radar device and the camera device in some embodiments of the present invention;

[0042] Figure 7 It is a schematic diagram of the structure of a parameter calibration device provided in some embodiments of the present invention. DETAILED DESCRIPTION

[0043] In order to make the purpose and advantages of the embodiments of the present invention easier to understand, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. The detailed description of the embodiments of the present invention in the drawings below does not limit the scope of protection claimed by the present invention, but only represents the selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0044] It should be noted that, if no conflict is constituted, the various technical features involved in the embodiments of the present invention described below can be combined with each other, and are all within the scope of protection of the present invention. In addition, although the functional modules are divided in the device or structural schematic diagram, and the logical order is shown in the flow chart, in some cases, the steps shown or described can be performed in a different order than the module division in the device, or in a different order than in the flow chart. In addition, the "first", "second", "third" and other similar expressions used herein do not limit the data and execution order, but are only for the purpose of convenience of explanation and to distinguish between the same or similar items with basically the same functions and effects, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features.

[0045] Unless otherwise defined, the technical terms and scientific terms used in this specification have the same meanings as those generally understood by those of ordinary skill in the art of the present invention. The terms used in this specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. It should be understood that the term "and / or" used in this specification includes any and all combinations of one or more of the related listed items.

[0046] See also Figure 1 , Figure 1A schematic diagram of an application scenario of a parameter calibration method provided by some embodiments of the present invention is schematically shown.

[0047] like Figure 1 As shown, the application scenario includes a radar device 10 and a camera device 20, and the radar device 10 and the camera device 20 are respectively connected to a controller ( Figure 1 The radar device 10 and the camera device 20 are communicatively connected (not shown), wherein the radar device 10 and the camera device 20 form a radar vision system.

[0048] Specifically, the radar device 10 emits radar signals to detect target objects within the detectable range and obtain radar data. The camera device 20 can fix and shoot target objects within the photographable range, and / or track and shoot target objects within the photographable range to obtain image data, which includes picture data and / or video data. Obviously, the target object can be a drone, a bird, or any other type of object, such as Figure 1 Shown are bird 31, drone 32 and drone 33.

[0049] In the radar data obtained by detection and the image data obtained by shooting, the radar data and image data containing the same target object and shot and detected at the same time are screened out as reference radar data and reference image data, respectively, to obtain a set of target detection data. According to the above conditions, the radar data and image data are screened multiple times, and finally several sets of target detection data are screened out, and then the several sets of target detection data are processed to obtain the azimuth angle data and the pitch angle data of the radar device 10, and the azimuth angle data and the pitch angle data of the camera device 20, so as to infer the specific position of the target object according to the azimuth angle data and the pitch angle data of the radar device 10, and the azimuth angle data and the pitch angle data of the camera device 20.

[0050] The inventors have found that when detecting small targets, the azimuth data and elevation data measured by the radar device 10 have large fluctuations. In addition, when running for a long time, the performance of the radar device 20 will decay, which will also affect the accuracy of the measured azimuth data and elevation data.

[0051] In view of this, an embodiment of the present invention provides a parameter calibration method, which obtains azimuth data and elevation data of a radar device and azimuth data and elevation data of a camera device in a radar vision system, and calculates target azimuth difference and target elevation difference of the radar device and the camera device according to the azimuth data and elevation data of the radar device and the azimuth data and elevation data of the camera device, so that the radar device and the camera device adjust the azimuth difference and the elevation difference to be the target azimuth difference and the target elevation difference, respectively, thereby realizing automatic calibration of the radar vision system, saving time and labor, improving calibration efficiency, and ensuring the accuracy of the calibration result, so that the radar vision system can obtain accurate azimuth data and elevation data of the radar device when detecting a target object, thereby improving the performance of the radar vision system and the accuracy of the target object detection result.

[0052] In the embodiment of the present invention, the radar device 10 may be a phased array radar, a Doppler radar, a synthetic aperture radar, a reverse synthetic aperture radar, etc., and the camera device 20 may be a PTZ (Pan-Tilt-Zoom) camera (a camera with omnidirectional movement and lens zoom and zoom control functions), a panoramic camera, a pan-tilt camera, etc. It is easy to understand that the radar device 10 and the camera device 20 may also be any other suitable equipment, device or component, and the embodiment of the present invention does not impose any limitation on this.

[0053] It is worth noting that, in some embodiments, the controller may be a single-chip microcomputer, a microcontroller or other suitable types of equipment, devices or components. For example, the controller may be a LabVIEW controller, which is a controller based on the LabVIEW graphical programming environment. Those skilled in the art may select any suitable controller according to actual needs, and the embodiments of the present invention do not impose any limitation on this.

[0054] It should be understood that Figure 1 This is just a schematic illustration of a situation where the radar vision system detects a target object. Figure 1 The radar device 10 and the camera device 20 in the application scenario shown do not impose any restrictions on the structure, type, quantity, etc. of the radar device and the camera device in other application scenarios or embodiments. For example, in some other application scenarios or embodiments, the radar device and the camera device may include multiple devices, and the locations of the radar device and the camera device may also be different from those in the embodiment shown in FIG. Figure 1 In addition, the radar device and the camera device in some other application scenarios or embodiments may also have the same Figure 1 The radar device 10 and the camera device 20 shown have different configurations.

[0055] See also Figure 2 , Figure 2The structural diagram of the drone detection system in some embodiments of the present invention is schematically shown.

[0056] like Figure 2 As shown, the drone detection system 1000 includes a controller 100 and a radar vision system 200, which is connected to the controller 100 through a network, and the radar vision system 200 includes a radar device 210 and a camera device 220. It can be understood that examples of the aforementioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0057] Specifically, the radar device 210 is used to detect the target object to obtain radar data, and send the radar data to the controller 100 , and the camera device 220 is used to detect the target object to obtain image data, and send the image data to the controller 100 .

[0058] The controller 110 is used to process the radar data and the image data to obtain the azimuth data and the pitch angle data of the radar device 210 and the azimuth data and the pitch angle data of the camera device 220, so as to infer the specific position of the target object based on the azimuth data and the pitch angle data of the radar device 210 and the azimuth data and the pitch angle data of the camera device 220, so as to achieve the purpose of detecting the target object.

[0059] In some embodiments, the controller 110 is also used to calculate the target azimuth angle difference and target pitch angle difference of the radar device 210 and the camera device 220 based on the azimuth angle data and pitch angle data of the radar device 210 and the azimuth angle data and pitch angle data of the camera device 220, and control the radar device 210 and the camera device 220 to adjust the current azimuth angle difference and pitch angle difference to the target azimuth angle difference and target pitch angle difference, respectively, thereby realizing automatic calibration of the radar vision system 200.

[0060] To facilitate understanding of the parameter calibration method provided by the embodiment of the present invention, the controller provided by the embodiment of the present invention is firstly introduced in detail.

[0061] See also Figure 3 , Figure 3 The schematic diagram shows the structure of the controller provided by some embodiments of the present invention.

[0062] like Figure 3 As shown, the controller 100 includes at least one processor 110 and a memory 120 that are communicatively connected. Figure 3In the example, the bus system 130 is connected to a processor. The various components in the controller 100 are coupled together through the bus system 130, and the bus system 130 is used to realize the connection and communication between the various components. It is easy to understand that the bus system 130 includes not only a data bus, but also a power bus, a control bus, and a status signal bus. However, for the sake of clarity and brevity, Figure 3 In the figure, various buses are labeled as bus system 130. It can be understood that Figure 3 The structure shown in the embodiment is only for illustration and does not impose any limitation on the structure of the controller. Figure 3 The structures shown may have more or fewer components, or may have the same Figure 3 Different configurations of the structure shown.

[0063] Specifically, the processor 110 is used to provide computing and control capabilities to control the controller 100 to perform corresponding tasks, such as controlling the above-mentioned controller 100 to perform any parameter calibration method provided in the embodiment of the present invention, or to perform any step in any possible implementation of any parameter calibration method provided in the embodiment of the present invention. Those skilled in the art can understand that the processor 110 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.

[0064] The memory 120, as a non-transient computer-readable storage medium, can be used to store non-transient software programs, non-transient computer executable programs, instructions and modules, for example, programs, instructions and modules corresponding to the parameter calibration method in the embodiment of the present invention. In some embodiments, the memory 120 may include a storage program area and a storage data area, the storage program area may store an operating system, an application required for at least one function, and the storage data area may store data created according to the use of the processor 110, etc. The processor 110 executes various functional applications and data processing of the controller 100 by running the non-transient software programs, instructions and modules stored in the memory 120 to implement any parameter calibration method provided in the embodiment of the present invention, or execute any of the steps in any possible implementation of any parameter calibration method provided in the embodiment of the present invention. In some embodiments, the memory 120 may include a high-speed random access memory and may also include a non-transient memory. For example, at least one disk storage device, a flash memory device or other non-transient solid-state storage device. In some embodiments, the memory 120 may also include a memory remotely arranged relative to the processor 110, and these remotely arranged memories may be connected to the processor 110 through a communication network. It is understandable that examples of the above-mentioned communication network include but are not limited to the Internet, corporate intranet, local area network, mobile communication network and combinations thereof.

[0065] It can be understood from the above that the implementation and execution subject of any parameter calibration method provided in the embodiment of the present invention can be any suitable type of controller with certain computing and control capabilities, for example, it can be implemented and executed by the above-mentioned controller 100. In some feasible implementations, any parameter calibration method provided in the embodiment of the present invention can be implemented by a processor calling a computer program instruction stored in a memory.

[0066] The parameter calibration method provided by the embodiment of the present invention will be described in detail below in conjunction with the exemplary application and implementation of the controller provided by the embodiment of the present invention.

[0067] See also Figure 4 , Figure 4 The flowchart of the parameter calibration method provided by some embodiments of the present invention is schematically shown.

[0068] Those skilled in the art will appreciate that the parameter calibration method provided in the embodiment of the present invention can be applied to the above controller (eg, controller 100). Specifically, the execution subject of the parameter calibration method is one or at least two processors of the controller.

[0069] like Figure 4 As shown, the parameter calibration method includes but is not limited to the following steps S100-S300:

[0070] S100: Acquire target data of a radar vision system, wherein the radar vision system is in communication with a controller and includes a radar device and a camera device, and the target data includes azimuth data and elevation data of the radar device, and azimuth data and elevation data of the camera device.

[0071] It can be understood that the azimuth angle data indicates the direction of the target object relative to the horizontal plane of the radar device or the camera device, and the elevation angle data indicates the direction of the target object relative to the vertical plane of the radar device or the camera device. Figure 5 and Figure 6 , Figure 5 The diagram schematically shows the relative position relationship (azimuth) between the target object and the radar device and the camera device. Figure 6 A schematic diagram showing the relative positional relationship (elevation angle) between the target object, the radar device, and the camera device. Figure 5 In the figure, the plane 30 is parallel to the horizontal plane where the radar device 10 and the camera device 20 are located, so the plane 30 can be used as the horizontal plane of the radar device 10 and the camera device 20, so that the azimuth data of the target object 32 relative to the radar device 10 and the camera device 20 are angles A1 and A2 respectively. Figure 6 In FIG. 1 , the vertical planes of the radar device 10 and the camera device 20 are plane F1 and plane F2, respectively, so that the pitch angle data of the target object 32 relative to the radar device 10 and the camera device 20 are angles B1 and B2, respectively. Figure 5 and Figure 6 The clockwise direction is taken as the positive direction (ie, the positive azimuth direction or the pitch angle direction). In some other embodiments, the counterclockwise direction may also be taken as the positive direction, and the embodiments of the present invention do not impose any limitation on this.

[0072] Specifically, the radar device detects radar data of surrounding objects by emitting radar signals (e.g., electromagnetic waves) and receiving radar signals reflected by objects, wherein the radar data includes position data (e.g., angle, height, and distance relative to the radar device) and motion data (e.g., speed relative to the radar device). The radar data is calculated to obtain azimuth data and pitch data of the radar device based on the target object, wherein the azimuth data and the pitch data represent the position of the target object in the radar field of view.

[0073] Specifically, the camera device captures visual information of an object by shooting images or video streams, and calculates azimuth data and elevation data of the camera device based on the feature points or target objects based on the relative position data (e.g., angle, height, and distance relative to the camera device) and relative motion data (e.g., speed relative to a radar device) of the feature points or target objects in the image data (i.e., images or video streams), wherein the azimuth data and elevation data represent the position of the feature points or target objects in the camera field of view.

[0074] Exemplarily, the radar device and the camera device are connected to the controller through wired communication or wireless communication (such as CAN bus, Ethernet, Wi-Fi, Bluetooth, etc.). After obtaining the azimuth data and the elevation data, the radar device and the camera device transmit their respective azimuth data and the elevation data to the controller through wired communication or wireless communication, so that the controller obtains the azimuth data and the elevation data of the radar device, and the azimuth data and the elevation data (i.e., the target data) of the camera device.

[0075] Of course, after the radar device detects and obtains radar data of surrounding objects, and the camera device captures image data (i.e., image or video stream), they may not process the radar data and image data, but directly transmit the radar data and image data to the controller, which processes the radar data and image data to calculate the azimuth data and elevation data of the radar device and the azimuth data and elevation data of the camera device (i.e., target data), thereby obtaining the target data of the radar vision system.

[0076] In some embodiments, obtaining target data of a radar vision system includes but is not limited to the following steps S110-S120:

[0077] S110: Acquire a first quantity of first azimuth angle data and first elevation angle data from the radar device, and acquire a second quantity of second azimuth angle data and second elevation angle data from the camera device.

[0078] Specifically, the radar device collects and obtains first azimuth angle data and first elevation angle data until a first quantity of first azimuth angle data and first elevation angle data is obtained, and the camera device captures and obtains second azimuth angle data and second elevation angle data until a second quantity of second azimuth angle data and second elevation angle data is obtained.

[0079] After the radar device collects and obtains a first quantity of first azimuth angle data and first elevation angle data, and the camera device captures and obtains a second quantity of second azimuth angle data and second elevation angle data, the radar device transmits the respectively obtained azimuth angle data and elevation angle data (i.e., the first azimuth angle data, the first elevation angle data, the second azimuth angle data, and the second elevation angle data) to a controller connected to the communication through a wired communication method or a wireless communication method (e.g., CAN bus, Ethernet, Wi-Fi, and Bluetooth), so that the controller obtains the first quantity of first azimuth angle data and first elevation angle data of the radar device, and the second quantity of second azimuth angle data and second elevation angle data of the camera device.

[0080] It can be understood that the first azimuth data, the first elevation data, the second azimuth data, and the second elevation data are all accompanied by a timestamp, and the timestamp indicates the time when the azimuth data or the elevation data is collected. In addition, those skilled in the art can determine the first quantity (i.e., the total number of the first azimuth data and the first elevation data) and the second quantity (i.e., the total number of the second azimuth data and the second elevation data) according to actual needs, and the first quantity and the second quantity can be the same or different, and the embodiment of the present invention does not impose any limitation on this.

[0081] In some embodiments, obtaining a first quantity of first azimuth angle data and first elevation angle data from a radar device specifically includes but is not limited to the following steps S1101:

[0082] S1101: extracting first azimuth angle data and first elevation angle data based on a standard target object in each frame of radar data from a plurality of frames of radar data detected and acquired by a radar device, until a first number of first azimuth angle data and first elevation angle data are extracted.

[0083] It is understandable that the standard target object is a fixed target object with a known shape and position and located within the field of view of the radar device and the camera device, and is a target object for comparison or calibration. Those skilled in the art can select any suitable object as the standard target object according to actual needs, and can set the standard target object at any suitable position according to actual needs, and the embodiments of the present invention do not make any limitation on this.

[0084] In some embodiments, the radar device sends several frames of radar data (including the standard target object) detected and acquired directly to the controller, which processes the radar data, determines the position of the standard target object through feature matching or target recognition algorithm, and calculates the first azimuth data and first elevation data based on the standard target object in each frame of radar data (that is, the azimuth and elevation of the standard target object relative to the radar device) according to the position of the standard target object, thereby extracting the first azimuth data and first elevation data based on the standard target object in each frame of radar data, until a sufficient number of first azimuth data and first elevation data are extracted to obtain a first number of first azimuth data and first elevation data.

[0085] For example, the radar device captures and acquires 30 frames of radar data, and extracts azimuth data and elevation data based on the standard target object in each frame of radar data according to the position of the standard target object, thereby extracting 30 azimuth data and 30 elevation data, and obtaining a first number (i.e., 60) of first azimuth data and first elevation data.

[0086] In some embodiments, obtaining a second amount of second azimuth angle data and second pitch angle data from the camera device specifically includes but is not limited to the following steps S1102:

[0087] S1102: extracting second azimuth angle data and second elevation angle data based on a standard target object in each frame of image data from a plurality of frames of image data captured by a camera device, until a second number of second azimuth angle data and second elevation angle data are extracted.

[0088] In some embodiments, the camera device directly sends several frames of image data (i.e., images or video streams, the image data including the standard target object) captured and acquired to the controller, which processes the image data, identifies the position information of the standard target object through image processing, target detection algorithm, deep learning model or computer vision method (such as Hough transform, feature point matching, etc.), calculates the second azimuth data and second pitch angle data based on the standard target object in each frame of image data according to the position information of the standard target object, thereby extracting the second azimuth data and second pitch angle data based on the standard target object in each frame of image data (i.e., the azimuth and pitch angle of the standard target object relative to the camera device), until a sufficient number of second azimuth data and second pitch angle data are extracted to obtain a first number of second azimuth data and second pitch angle data.

[0089] For example, the camera device captures 15 frames of image data, and extracts azimuth data and pitch angle data based on the standard target object in each frame of image data according to the position of the standard target object, thereby extracting 15 azimuth data and 15 pitch angle data, and obtaining a second number (i.e., 30) of second azimuth data and second pitch angle data.

[0090] S120: According to the timestamps of the azimuth data and the elevation data, the first azimuth data and the second azimuth data with the same timestamp are combined into a data pair, and the first elevation data and the second elevation data with the same timestamp are combined into a data pair to obtain target data, wherein the target data includes an azimuth data pair and an elevation data pair.

[0091] Wherein, the azimuth data and the pitch angle data are both accompanied by corresponding timestamps. The first azimuth data and the second azimuth data, and the first pitch angle data and the second pitch angle data are obtained and paired according to the timestamps of the azimuth data (i.e., the first azimuth data and the second azimuth data) and the pitch angle data (i.e., the first pitch angle data and the second pitch angle data). That is, the first azimuth data and the second azimuth data with the same timestamp are combined into a group of data pairs, and the first pitch angle data and the second pitch angle data with the same timestamp are combined into a group of data pairs, thereby obtaining several groups of azimuth data pairs and several groups of pitch angle data pairs (i.e., target data).

[0092] S200: Calculate the target azimuth angle difference and the target elevation angle difference between the radar device and the camera device based on the target data.

[0093] In some embodiments, for the radar device and the camera device, the units of the azimuth angle data and the elevation angle data are degrees (°). Of course, the units of the azimuth angle data and the elevation angle data can also be expressed in radians.

[0094] Specifically, the azimuth data of the radar device is subtracted from the azimuth data of the camera device to obtain the target azimuth difference between the radar device and the camera device. It is understandable that the azimuth difference needs to take into account its periodicity, for example, the mapping between 0° and 360°, that is, it needs to be adjusted according to the actual range of the target azimuth difference so that the target azimuth difference is within the range of -180° to +180°. For example, if the target azimuth difference is greater than 180°, it is necessary to adjust the target azimuth difference to a smaller value, that is, to subtract 360°. If the target azimuth difference is less than -180°, it is necessary to adjust the target azimuth difference to a larger value, that is, to add 360°. It should be understood that the azimuth difference is limited to between -180° and +180° in order to ensure that the minimum difference angle between the two azimuths is calculated to avoid unnecessary large angle differences within a range of 360 degrees.

[0095] Specifically, the pitch angle data of the radar device is subtracted from the pitch angle data of the camera device to obtain the target azimuth difference of the radar device and the camera device. It is understandable that it is also necessary to ensure that the target azimuth difference is within a reasonable range (i.e., the target pitch angle difference is between -90° and +90°). If the target pitch angle difference is greater than 90°, it is necessary to subtract 180° from the target pitch angle difference. If the target pitch angle difference is less than -90°, it is necessary to add 180° to the target pitch angle difference so that the target pitch angle difference is within the range of -90° to +90°. It should be understood that the pitch angle difference is limited to between -90° and +90° because the maximum pitch angle of an object is ±90° in physics. Objects beyond the range of -90° to +90° are actually in an inverted or completely reversed state, which is physically unattainable and has no practical significance.

[0096] In some embodiments, the target data includes multiple groups of azimuth data and elevation data of radar devices, and azimuth data and elevation data of camera devices. The azimuth data of the camera device is subtracted from the azimuth data of each group of radar devices to obtain several groups of azimuth differences between radar devices and camera devices. The average of the several groups of azimuth differences is calculated to obtain the target azimuth difference between the radar device and the camera device.

[0097] Similarly, the pitch angle data of the camera device is subtracted from the pitch angle data of each group of radar devices to obtain several groups of pitch angle differences between the radar device and the camera device, and then the average of the several groups of pitch angle differences is calculated to obtain the target pitch angle difference between the radar device and the camera device.

[0098] Of course, any other suitable method may be used to calculate the target azimuth angle difference and the target elevation angle difference of the radar device and the camera device, and the embodiment of the present invention does not impose any limitation on this.

[0099] In some embodiments, a linear regression algorithm may be used to perform fitting calculations on the azimuth angle data and the elevation angle data, respectively, to obtain a target azimuth angle difference and a target elevation angle difference.

[0100] In some embodiments, based on the target data, the target azimuth angle difference and the target elevation angle difference of the radar device and the camera device are calculated, including but not limited to the following steps S210:

[0101] S210: Using a linear regression algorithm, respectively perform fitting calculations on the azimuth angle data pair and the pitch angle data pair to obtain a target azimuth angle difference and a target pitch angle difference, wherein the linear regression algorithm includes one or more of a least squares algorithm, a weighted least squares algorithm, a least absolute deviation regression algorithm, a ridge regression algorithm, an elastic net regression algorithm, and a lasso regression algorithm.

[0102] It can be understood that the linear regression algorithm includes one or more of the least squares algorithm, the weighted least squares algorithm, the least absolute deviation regression algorithm, the ridge regression algorithm, the elastic net regression algorithm and the lasso regression algorithm. The linear regression algorithm is to find the regression coefficient by minimizing the square difference between the predicted value and the actual value, so as to achieve the fitting to obtain the target azimuth angle difference and the target pitch angle difference. Of course, those skilled in the art can also use any other suitable algorithm, and the embodiment of the present invention does not make any limitation to this.

[0103] In some embodiments, after obtaining target data (including: a plurality of azimuth angle data pairs and a plurality of elevation angle data pairs), a linear regression algorithm is used to fit the azimuth angle data pairs and the elevation angle data pairs to calculate the target azimuth angle difference and the target elevation angle difference. The azimuth angle data pairs are represented as [Ax t ,Ay t ]), the pitch angle data is expressed as ([Px t ,Py t ]), where t is a timestamp index, indicating the azimuth angle data or elevation angle data of the radar device and the camera device at the same moment.

[0104] In some embodiments, if the least squares algorithm is used to fit the azimuth angle data pair and the pitch angle data pair, the azimuth angle data pair and the pitch angle data pair are respectively substituted into the formula corresponding to the least squares algorithm, and after multiple fitting calculations, the square difference between the predicted value and the actual value is minimized, thereby obtaining the regression coefficient, wherein the formula corresponding to the least squares algorithm is:

[0105]

[0106] Among them, x i is the first azimuth angle data or the first elevation angle data of the radar device, y i is the second azimuth angle data or the second pitch angle data of the camera device, β0 and β1 are regression coefficients, and the regression coefficient β0 obtained by fitting is the target azimuth angle difference or the target pitch angle difference.

[0107] In some embodiments, if a weighted least squares algorithm is used to fit the azimuth data pair and the elevation data pair, the azimuth data pair and the elevation data pair are respectively substituted into the formula corresponding to the weighted least squares algorithm, and after multiple fitting calculations, the square difference between the predicted value and the actual value is minimized, thereby obtaining the regression coefficient. The weighted least squares method is similar to the least squares method, except that the weighted least squares method assigns a weight to each data point. Among them, the formula corresponding to the weighted least squares algorithm is:

[0108]

[0109] Among them, xi is the first azimuth angle data or the first elevation angle data of the radar device, y i is the second azimuth angle data or the second elevation angle data of the camera device, w i is the weight of each data point, β0 and β1 are regression coefficients, and the regression coefficient β0 obtained by fitting is the target azimuth angle difference or the target pitch angle difference.

[0110] In some embodiments, if the minimum absolute deviation regression algorithm is used to fit the azimuth data pair and the pitch angle data pair, the azimuth data pair and the pitch angle data pair are respectively substituted into the formula corresponding to the minimum absolute deviation regression algorithm, and after multiple fitting calculations, the absolute difference between the predicted value and the actual value is minimized, thereby obtaining the regression coefficient. The minimum absolute deviation regression algorithm is similar to the least squares method, except that the minimum absolute deviation regression algorithm solves the regression coefficient by minimizing the absolute difference between the actual value and the predicted value. Among them, the formula corresponding to the minimum absolute deviation regression algorithm is:

[0111]

[0112] Among them, x i is the first azimuth angle data or the first elevation angle data of the radar device, y i is the second azimuth angle data or the second pitch angle data of the camera device, β0 and β1 are regression coefficients, and the regression coefficient β0 obtained by fitting is the target azimuth angle difference or the target pitch angle difference.

[0113] In some embodiments, if the ridge regression algorithm is used to fit the azimuth data pair and the pitch angle data pair, the azimuth data pair and the pitch angle data pair are respectively substituted into the formula corresponding to the ridge regression algorithm, and after multiple fitting calculations, the absolute difference between the predicted value and the actual value is minimized, thereby obtaining the regression coefficient. The ridge regression algorithm is similar to the least squares method, except that the ridge regression algorithm adds an L2 regularization term to prevent overfitting. Among them, the formula corresponding to the ridge regression algorithm is:

[0114]

[0115] Among them, x i is the first azimuth angle data or the first elevation angle data of the radar device, y i is the second azimuth angle data or the second pitch angle data of the camera device, λ is the parameter of the L2 regularization term, β0, β1 and β i are all regression coefficients, and the regression coefficient β0 obtained by fitting is the target azimuth angle difference or the target pitch angle difference.

[0116] In some embodiments, if the lasso regression algorithm is used to fit the azimuth data pair and the pitch angle data pair, the azimuth data pair and the pitch angle data pair are respectively substituted into the formula corresponding to the lasso regression algorithm, and after multiple fitting calculations, the absolute difference between the predicted value and the actual value is minimized, thereby obtaining the regression coefficient. The lasso regression algorithm is similar to the least squares method, except that the lasso regression algorithm adds an L1 regularization term, and some specific regression coefficients are made to tend to zero through L1 regularization. Among them, the formula corresponding to the lasso regression algorithm is:

[0117]

[0118] Among them, x i is the first azimuth angle data or the first elevation angle data of the radar device, y i is the second azimuth angle data or the second pitch angle data of the camera device, λ is the parameter of the L1 regularization term, β0, β1 and β i are all regression coefficients, and the regression coefficient β0 obtained by fitting is the target azimuth angle difference or the target pitch angle difference.

[0119] In some embodiments, if the elastic net regression algorithm is used to fit the azimuth data pair and the pitch angle data pair, the azimuth data pair and the pitch angle data pair are respectively substituted into the formula corresponding to the elastic net regression algorithm, and after multiple fitting calculations, the square difference between the predicted value and the actual value is minimized, thereby obtaining the regression coefficient. The elastic net regression algorithm is a combination of the ridge regression algorithm and the lasso regression algorithm, and uses L1 and L2 regularization terms, which is suitable for high-dimensional data. Among them, the formula corresponding to the elastic net regression algorithm is:

[0120]

[0121] Among them, x i is the first azimuth angle data or the first elevation angle data of the radar device, y i is the second azimuth angle data or the second pitch angle data of the camera device, λ1 and λ2 are the parameters of the L1 regularization term and the L2 regularization term, respectively, β0, β1 and β i are all regression coefficients, and the regression coefficient β0 obtained by fitting is the target azimuth angle difference or the target pitch angle difference.

[0122] S300: Sending the target azimuth angle difference and the target elevation angle difference to the radar vision system, so that the radar vision system replaces the current azimuth angle difference and the current elevation angle difference with the target azimuth angle difference and the target elevation angle difference.

[0123] Specifically, after calculating and obtaining the target azimuth difference and target elevation difference of the radar device and the camera device, the target azimuth difference and target elevation difference are sent to the radar vision system through a wired communication method or a wireless communication method (such as CAN bus, Ethernet, Wi-Fi, Bluetooth, etc.), so that the radar vision system replaces the current azimuth difference and elevation difference with the target azimuth difference and target elevation difference. After receiving the target azimuth difference and target elevation difference sent by the controller, the radar vision system adjusts the positions of the radar device and the camera device so that the difference between the azimuth data of the standard target object based on the radar device and the camera device is the target azimuth difference, and the difference between the elevation data of the standard target object based on the radar device and the camera device is the target elevation difference, thereby completing the replacement of the current azimuth difference and elevation difference with the target azimuth difference and the target elevation difference, respectively.

[0124] In some embodiments, when the current azimuth angle difference and the pitch angle difference are replaced with the target azimuth angle difference and the target pitch angle difference, the position of the camera device remains unchanged, and only the position of the radar device is adjusted, so that the difference between the azimuth angle data of the radar device and the camera device based on the standard target object is the target azimuth angle difference, and the difference between the pitch angle data of the radar device and the camera device based on the standard target object is the target pitch angle difference.

[0125] Of course, in some embodiments, the position of the radar device may remain unchanged, and only the position of the camera device may be adjusted, so that the difference between the azimuth data of the standard target object based on the radar device and the camera device is the target azimuth difference, and the difference between the elevation data of the standard target object based on the radar device and the camera device is the target elevation difference.

[0126] In some embodiments, the positions of the radar device and the camera device are adjusted so that the difference between the azimuth data of the standard target object of the radar device and the camera device is the target azimuth difference, and the difference between the elevation data of the standard target object of the radar device and the camera device is the target elevation difference.

[0127] In some embodiments, after obtaining the first azimuth data and the first elevation data of the radar device and obtaining the second azimuth data and the second elevation data of the camera device, the azimuth data and the elevation data need to be preprocessed (e.g., screening, smoothing, and normalization, etc.), so as to improve the accuracy of calculating the target azimuth difference and the target elevation difference.

[0128] Exemplarily, in some embodiments, after obtaining a first quantity of first azimuth angle data and first elevation angle data from the radar device, and obtaining a second quantity of second azimuth angle data and second elevation angle data from the camera device, obtaining target data of the radar vision system specifically includes but is not limited to the following steps S111-S112:

[0129] S111: Filter out target angle data that meets data screening conditions from the first azimuth angle data and the first elevation angle data of the radar device. The data screening conditions include that the azimuth angle is less than or equal to a first angle threshold and the elevation angle is less than or equal to a second angle threshold. The target angle data includes azimuth angle data and elevation angle data.

[0130] Among them, the data screening conditions include the azimuth angle being less than or equal to the first angle threshold and the pitch angle being less than or equal to the second angle threshold. Technical personnel in this field can set the first angle threshold and the second angle threshold according to actual needs. For example, in some embodiments, the first angle threshold is set to 5° and the second angle threshold is set to 8°. The embodiments of the present invention do not impose any limitations on this.

[0131] Specifically, after acquiring the first azimuth data and the first elevation data of the radar device, the first azimuth data and the first elevation data are respectively compared with the data screening condition, that is, the first azimuth data is compared with the first angle threshold, if the azimuth angle in the first azimuth data is less than or equal to the first angle threshold, it indicates that the first azimuth data meets the condition, and the first azimuth data is screened out as the target angle data. And the first elevation data is compared with the first angle threshold, if the elevation angle in the first elevation data is less than or equal to the first angle threshold, it indicates that the first elevation data meets the condition, and the first elevation data is screened out as the target angle data. It can be understood that the target angle data includes azimuth data (i.e., the first azimuth data that meets the data screening condition) and elevation data (i.e., the first elevation data that meets the data screening condition).

[0132] S112: With the timestamp of the target angle data as the horizontal axis and the target angle data as the vertical axis, a target function corresponding to the target angle data is constructed, and the target angle data of the target function is smoothed by exponential weighted averaging to obtain the target azimuth angle data and target pitch angle data of the radar device.

[0133] Specifically, the timestamp of the filtered target angle data (i.e., the first azimuth angle data and the first elevation angle data of the radar device that meets the data filtering conditions) is used as the horizontal axis, and the corresponding filtered target angle data is used as the vertical axis to establish a target function corresponding to the target angle data, thereby constructing a target function corresponding to the target angle data, and the target function includes the timestamp, the first azimuth angle data, and the first elevation angle data.

[0134] Next, the target angle data in the objective function is smoothed to remove noise and smooth the data. Specifically, the target angle data is smoothed using the Exponential Moving Average (EMA) method to obtain the target azimuth data and target elevation data of the radar device after smoothing, wherein the azimuth data and the elevation data are smoothed respectively, and the target azimuth data and the target elevation data have good smoothness and accuracy.

[0135] In some embodiments, the exponentially weighted average formula is as follows:

[0136] y t =αx t +(1-α)y t-1

[0137] Among them, y t is the smoothed target angle data (i.e. azimuth data or elevation data), x t is the target angle data (i.e. azimuth data or pitch angle data) at the current time point, α is the smoothing coefficient, which determines the weight of historical data. The larger the α value, the greater the weight of the current data, and the smaller the α value, the greater the weight of the historical data. t-1 It is the target angle data at the previous time point.

[0138] Exemplarily, in some embodiments, obtaining target data of a radar vision system specifically includes but is not limited to the following steps S113:

[0139] S113: Normalizing the target azimuth data and the target elevation data of the radar device to obtain first target data, wherein the first target data is the azimuth data and the elevation data of the radar device after screening, smoothing and normalization.

[0140] Specifically, the target azimuth data and target elevation data of the radar device are normalized to ensure that the target azimuth data and target elevation data of the radar device have a uniform scale in subsequent calculations. The purpose of normalization is to scale the target azimuth data and target elevation data to a specific range, for example, the range may be [0, 1], [-1, 1], [-π, π] or [0, π], etc., so that target azimuth data or target elevation data of different scales can be compared and processed at the same scale.

[0141] It should be understood that the following two common normalization methods are used: minimum-maximum normalization (scaling the data to the range of [0, 1], [-1, 1], [-π, π] or [0, π]) and Z-Score standardization (converting the data to a standard normal distribution with a mean of 0 and a standard deviation of 1).

[0142] In an embodiment of the present invention, a minimum-maximum normalization method is used to normalize the target azimuth data and the target elevation data of the radar device respectively. The minimum-maximum normalization method is simple and suitable for processing angle data, and has a good normalization effect on the angle data.

[0143] Specifically, the target azimuth data and the target elevation data are respectively substituted into the normalization formula, and after calculation by the normalization formula, the corresponding first target data is output, wherein the first target data is the azimuth data and elevation data of the radar device after screening, smoothing and normalization. In some embodiments, the normalization formula is as follows:

[0144]

[0145] Among them, Axmin and Axmax are the minimum and maximum values ​​of the target azimuth data of the radar device, Pxmin and Pxmax are the minimum and maximum values ​​of the target elevation data of the radar device, A x [i] and P x [i] are target azimuth angle data and target elevation angle data, A′ x [i] and P′ x [i] are the normalized azimuth angle data and elevation angle data respectively.

[0146] Through the above formula, the target azimuth data and target elevation data of the radar device are scaled to the range of [0, 1], thereby obtaining the azimuth data and elevation data (i.e., the first target data) of the radar device that have been screened, smoothed and normalized.

[0147] Exemplarily, in some embodiments, obtaining target data of a radar vision system further includes but is not limited to the following steps S114:

[0148] S114: normalizing the second azimuth angle data and the second elevation angle data of the camera device to obtain second target data, wherein the second target data is the normalized azimuth angle data and the elevation angle data of the camera device.

[0149] Specifically, the second azimuth angle data and the second pitch angle data of the camera device are normalized to ensure that the second azimuth angle data and the second pitch angle data of the camera device have a uniform scale in subsequent calculations. The purpose of normalization is to scale the second azimuth angle data and the second pitch angle data to a specific range, for example, the range may be [0, 1], [-1, 1], [-π, π] or [0, π], etc., so that the second azimuth angle data or the second pitch angle data of different scales can be compared and processed at the same scale.

[0150] It should be understood that the following two common normalization methods are used: minimum-maximum normalization (scaling the data to the range of [0, 1], [-1, 1], [-π, π] or [0, π]) and Z-Score standardization (converting the data to a standard normal distribution with a mean of 0 and a standard deviation of 1).

[0151] In an embodiment of the present invention, a minimum-maximum normalization method is used to normalize the second azimuth angle data and the second pitch angle data of the camera device respectively. The minimum-maximum normalization method is simple and suitable for processing angle data, and has a good normalization effect on the angle data.

[0152] Specifically, the second azimuth angle data and the second pitch angle data are respectively substituted into the normalization formula, and after calculation by the normalization formula, the corresponding second target data is output, and the second target data is the normalized azimuth angle data and pitch angle data of the camera device. In some embodiments, the normalization formula is as follows:

[0153]

[0154] Wherein, Aymin and Aymax are respectively the minimum value and the maximum value in the second azimuth angle data of the camera device, Pymin and Pymax are respectively the minimum value and the maximum value in the second pitch angle data of the camera device, A y [i] and P y [i] are the second azimuth angle data and the second elevation angle data, A′ y [i] and P′ y [i] are the normalized azimuth and elevation data respectively.

[0155] By using the above formula, the second azimuth angle data and the second elevation angle data of the camera device are scaled to the range of [0, 1], thereby obtaining the normalized azimuth angle data and elevation angle data (ie, the second target data) of the camera device.

[0156] In summary, the parameter calibration method provided by the embodiment of the present invention is applied to a controller, and the method includes: acquiring target data of a radar vision system, wherein the radar vision system is communicatively connected to the controller, and includes a radar device and a camera device, and the target data includes azimuth data and elevation data of the radar device, and azimuth data and elevation data of the camera device; based on the target data, a target azimuth difference and a target elevation difference of the radar device and the camera device are calculated, and the target azimuth difference and the target elevation difference are sent to the radar vision system, so that the radar vision system replaces the current azimuth difference and the elevation difference with the target azimuth difference and the target elevation difference.

[0157] The embodiment of the present invention obtains the azimuth data and elevation data of the radar device and the camera device in the radar vision system, calculates the target azimuth difference and the target elevation difference of the radar device and the camera device, and enables the radar device and the camera device to adjust the current azimuth difference and the elevation difference to the target azimuth difference and the target elevation difference, respectively, thereby realizing automatic calibration of the radar vision system, saving time and labor, improving calibration efficiency, and ensuring the accuracy of the calibration result.

[0158] As another aspect of an embodiment of the present invention, an embodiment of the present invention provides a parameter calibration device, wherein the parameter calibration device may be a software module, the software module includes a plurality of instructions, which are stored in a memory, and a processor may access the memory and call the instructions for execution to complete the parameter calibration method described in each of the above embodiments.

[0159] In some possible implementations, the parameter calibration device may also be constructed by hardware devices. For example, the parameter calibration device may be constructed by one or more chips, and the chips may work in coordination with each other to implement the parameter calibration methods described in the above implementations. In some embodiments, the parameter calibration device may also be constructed by various logic devices, such as: a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a single-chip microcomputer, a field programmable gate array (FPGA), an ARM (Acorn RISC Machine) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, or any combination of these devices or components.

[0160] See also Figure 7 , Figure 7 The schematic diagram of the structure of the parameter calibration device provided by some embodiments of the present invention is shown schematically. It is easy to understand that the parameter calibration device can be configured in the controller.

[0161] Specifically, Figure 7 As shown, the parameter calibration device 500 includes an acquisition module 510 , a calculation module 520 and a sending module 530 .

[0162] The acquisition module 510 is used to acquire target data of the radar vision system, wherein the radar vision system is connected to the controller in communication, and includes a radar device and a camera device, and the target data includes azimuth data and elevation data of the radar device, and azimuth data and elevation data of the camera device. The calculation module 520 is used to calculate the target azimuth difference and target elevation difference of the radar device and the camera device based on the target data. The sending module 530 is used to send the target azimuth difference and target elevation difference to the radar vision system, so that the radar vision system replaces the current azimuth difference and elevation difference with the target azimuth difference and target elevation difference.

[0163] In some embodiments, the acquisition module 510 is specifically used to: acquire a first quantity of first azimuth data and first elevation data from a radar device, and acquire a second quantity of second azimuth data and second elevation data from a camera device; according to the timestamps of the azimuth data and the elevation data, form a data pair with the first azimuth data and the second azimuth data with the same timestamp; and form a data pair with the first pitch angle data and the second pitch angle data with the same timestamp to obtain target data, wherein the target data includes an azimuth data pair and an elevation angle data pair.

[0164] In some embodiments, the calculation module 520 is specifically used to: use a linear regression algorithm to perform fitting calculations on the azimuth data pairs and the pitch angle data pairs, respectively, to obtain the target azimuth angle difference and the target pitch angle difference, and the linear regression algorithm includes one or more of a least squares algorithm, a weighted least squares algorithm, a minimum absolute deviation regression algorithm, a ridge regression algorithm, an elastic net regression algorithm, and a lasso regression algorithm.

[0165] In some embodiments, the acquisition module 510 is also specifically used to: extract the first azimuth angle data and the first elevation angle data based on the standard target object in each frame of radar data from several frames of radar data acquired by detection by the radar device, until a first quantity of first azimuth angle data and first elevation angle data are extracted; and extract the second azimuth angle data and the second elevation angle data based on the standard target object in each frame of image data from several frames of image data acquired by capture by the camera device, until a second quantity of second azimuth angle data and second elevation angle data are extracted.

[0166] In some embodiments, the acquisition module 510 is further specifically used to: after acquiring a first quantity of first azimuth data and first elevation data from the radar device, and acquiring a second quantity of second azimuth data and second elevation data from the camera device, filter out target angle data that meets the data filtering condition from the first azimuth data and the first elevation data of the radar device, the data filtering condition including that the azimuth angle is less than or equal to the first angle threshold and the elevation angle is less than or equal to the second angle threshold, and the target angle data includes the azimuth data and the elevation data. The smoothing module 550 is used to filter out target angle data that meets the data filtering condition from the first azimuth data and the first elevation data of the radar device, the data filtering condition including that the azimuth angle is less than or equal to the first angle threshold and the elevation angle is less than or equal to the second angle threshold, and the target angle data includes the azimuth data and the elevation data.

[0167] In some embodiments, the acquisition module 510 is also specifically used to: normalize the target azimuth data and target elevation data of the radar device to obtain first target data, wherein the first target data is the azimuth data and elevation data of the radar device after screening, smoothing and normalization.

[0168] In some embodiments, the acquisition module 510 is further specifically used to normalize the second azimuth data and the second elevation data of the camera device to obtain second target data, wherein the second target data is the normalized azimuth data and the elevation data of the camera device.

[0169] It should be noted that, for the sake of simplicity and brevity of description, the parameter calibration device 500 can execute the functional modules and beneficial effects corresponding to the parameter calibration method provided in the embodiment of the present invention. For technical details not fully described in the embodiment of the parameter calibration device, reference can be made to the parameter calibration method provided in the embodiment of the present invention. The specific working process of the parameter calibration device 500 can also refer to the specific execution process corresponding to the parameter calibration method provided in the embodiment of the present invention, and will not be described one by one here.

[0170] An embodiment of the present invention provides a computer-readable storage medium, on which computer program instructions executable by a processor are stored. When the computer program instructions are called by the processor, the processor executes any parameter calibration method provided by the embodiment of the present invention, or executes steps in any possible implementation manner of any parameter calibration method provided by the embodiment of the present invention.

[0171] In some embodiments, the storage medium may be a flash memory, a hard disk, an optical disk, a register, a magnetic surface storage, a removable disk, a CD-ROM, a random access memory (RAM), a read-only memory (ROM), an electrically programmable ROM, and an electrically erasable programmable ROM, or any other form of storage medium known in the technical field, or various devices including one or any combination of the above storage media.

[0172] In some embodiments, computer program instructions may be in the form of a program, software, software module, script or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a stand-alone program or as a module, component, subroutine or other unit suitable for use in a computing environment.

[0173] As an example, computer program instructions may, but do not necessarily, correspond to a file in a file system, may be stored as part of a file that stores other programs or data, for example, in one or more scripts in a HyperText Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files storing one or more modules, subroutines, or code portions).

[0174] As an example, the computer program instructions may be deployed to be executed on a computing device (including devices such as smart terminals and servers), or on multiple computing devices located at one location, or on multiple computing devices distributed at multiple locations and interconnected by a communication network. It is easy to understand that all or part of the steps of the method described in the embodiments of the present invention may be directly implemented using electronic hardware or processor executable computer program instructions, or a combination of the two.

[0175] Those skilled in the art will appreciate that the embodiments provided by the present invention are only illustrative, and the order of writing the steps in the method of the embodiment does not mean a strict execution order and does not constitute any limitation on the implementation process. The order can be adjusted, merged and deleted according to actual needs, and the modules or sub-modules, units or sub-units in the device or system of the embodiment can be merged, divided and deleted according to actual needs. For example, the division of units is only a logical function division, and there can be other division methods in actual implementation. For another example, multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed.

[0176] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a general hardware platform, and of course can also be implemented by hardware. Those skilled in the art can understand that all or part of the processes in the above embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0177] It should be noted that the above embodiments are intended to illustrate the technical concept and features of the present invention, and their purpose is to enable people familiar with this technology to understand the content of the present invention and implement it accordingly, and cannot be used to limit the scope of protection of the present invention. Ordinary technicians in this field can understand that all or part of the processes of the above embodiments can be implemented according to the technical solutions recorded in the embodiments of the present invention, or some of the technical features can be equivalently replaced. It can be understood that these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and should be regarded as equal changes and modifications based on the embodiments of the present invention, and should all fall within the scope of the claims of the present invention.

Claims

1. A parameter calibration method, characterized in that: Applied to a controller, the method comprises: Acquiring target data of a radar vision system, wherein the radar vision system is communicatively connected to the controller and comprises a radar device and a camera device, and the target data comprises azimuth data and elevation data of the radar device, and azimuth data and elevation data of the camera device; Based on the target data, calculating a target azimuth difference and a target elevation difference between the radar device and the camera device; The target azimuth angle difference and the target elevation angle difference are sent to the radar vision system, so that the radar vision system replaces the current azimuth angle difference and the current elevation angle difference with the target azimuth angle difference and the target elevation angle difference.

2. The method according to claim 1, characterized in that The step of obtaining target data of the radar vision system includes: Acquire a first quantity of first azimuth angle data and first elevation angle data from the radar device, and acquire a second quantity of second azimuth angle data and second elevation angle data from the camera device; According to the timestamps of the azimuth data and the elevation data, the first azimuth data and the second azimuth data with the same timestamp are combined into a data pair, and the first pitch angle data and the second pitch angle data with the same timestamp are combined into a data pair to obtain the target data, which includes an azimuth data pair and an elevation data pair.

3. The method according to claim 2, characterized in that The step of calculating the target azimuth difference and the target elevation difference between the radar device and the camera device based on the target data includes: A linear regression algorithm is used to fit the azimuth angle data pair and the pitch angle data pair to obtain a target azimuth angle difference and a target pitch angle difference. The linear regression algorithm includes one or more of a least squares algorithm, a weighted least squares algorithm, a minimum absolute deviation regression algorithm, a ridge regression algorithm, an elastic net regression algorithm, and a lasso regression algorithm.

4. The method according to claim 2, characterized in that: The step of acquiring a first quantity of first azimuth angle data and first elevation angle data from the radar device comprises: Extracting first azimuth angle data and first elevation angle data based on a standard target object in each frame of radar data from a plurality of frames of radar data detected and acquired by the radar device, until a first number of first azimuth angle data and first elevation angle data are extracted; The step of acquiring a second amount of second azimuth angle data and second pitch angle data from the camera device comprises: From the plurality of frames of image data captured by the camera device, second azimuth angle data and second elevation angle data based on the standard target object are extracted from each frame of image data until a second number of second azimuth angle data and second elevation angle data are extracted.

5. The method according to claim 2, characterized in that: After acquiring a first quantity of first azimuth angle data and first elevation angle data from the radar device, and acquiring a second quantity of second azimuth angle data and second elevation angle data from the camera device, acquiring target data of the radar vision system further includes: Filtering target angle data that meets a data screening condition from the first azimuth angle data and the first elevation angle data of the radar device, the data screening condition comprising an azimuth angle being less than or equal to a first angle threshold and an elevation angle being less than or equal to a second angle threshold, the target angle data comprising azimuth angle data and elevation angle data; With the timestamp of the target angle data as the horizontal axis and the target angle data as the vertical axis, a target function corresponding to the target angle data is constructed, and the target angle data of the target function is smoothed by exponential weighted averaging to obtain the target azimuth angle data and target elevation angle data of the radar device.

6. The method according to claim 5, characterized in that The obtaining of target data of the radar vision system also includes: The target azimuth data and the target elevation data of the radar device are normalized to obtain first target data, wherein the first target data is the azimuth data and the elevation data of the radar device after screening, smoothing and normalization.

7. The method according to claim 6, characterized in that The obtaining of target data of the radar vision system also includes: The second azimuth angle data and the second elevation angle data of the camera device are normalized to obtain second target data, wherein the second target data is the normalized azimuth angle data and the elevation angle data of the camera device.

8. A controller, characterized in that: include: A processor and a memory communicatively connected to the processor; The memory stores computer program instructions executable by the processor, and when the computer program instructions are called by the processor, the processor executes the parameter calibration method according to any one of claims 1 to 7.

9. A drone detection system, characterized in that: include: The controller as claimed in claim 8; as well as, The radar vision system is connected to the controller in communication, and includes a radar device and a camera device. The radar device is used to detect the target object to obtain radar data, and the camera device is used to detect the target object to obtain image data.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer program instructions executable by a processor, and when the computer program instructions are called by the processor, the processor executes the parameter calibration method according to any one of claims 1 to 7.

Images