A Fast Target Calibration Method and System for Unmanned Aerial Vehicles

Abstract

The present invention provides a fast target calibration method and system for an unmanned aerial vehicle, relating to the technical field of image processing. The present invention includes a vision camera, a calibration prototype, a voiceprint collector, a prototype console, a software monitoring console, and a calibration console. It cooperates to verify whether the control amounts of the voiceprint ranging and the prototype console match in the front and rear distances. Then, the prototype console is used to control the unmanned aerial vehicle to fly horizontally in a plane, and it is judged whether the spatial angle data is accurately measured. Finally, based on the accurate relative distance data and spatial angle data, the corresponding prototype position information and prototype speed information are calculated, so as to provide an adjustment basis for the distance parameter, angle parameter, position parameter, and speed parameter. The present invention skillfully utilizes the advantages of voiceprint ranging in linear distance and the advantages of image comparison in plane angle measurement, and cooperates with the closely linked data verification logic line to achieve the standardized, fast, and automated target calibration task only by relying on simple equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and particularly to a fast target calibration method and system for unmanned aerial vehicles (UAVs). Background Art

[0002] With the progress of technology and the development of the times, UAVs have been applied in all aspects of our lives. While UAVs bring us convenience, we also need to manage them reasonably and effectively. In some places where low-altitude flight control is required, we are required to set up UAV identification systems to achieve the purpose of orderly management.

[0003] There are various types of existing UAV identification systems. Among them, the UAV system based on pure vision algorithms is the most widely used. It has the advantages of convenient deployment, low hardware cost, simple software algorithm, and minimal interference to the surrounding environment (especially no electromagnetic interference), which cannot be replaced by other UAV identification systems.

[0004] However, when the UAV system based on pure vision algorithms is deployed in different scenarios, it is necessary to calibrate parameters one by one according to different environmental conditions to achieve accurate identification and fast response tracking in different scenarios. However, the calibration process often involves parameter coordination among multiple parties.

[0005] The traditional calibration method relies on manual use of various instruments and equipment and combines the feedback of the picture for repeated calibration. The disadvantage of this is that there is calibration subjectivity, relying on the experience judgment of the calibrator, and the UAV system based on pure vision algorithms cannot directly give reliable data feedback. Therefore, it often takes 3 - 5 days to complete the calibration (previous engineering experience).

[0006] Therefore, it is necessary to provide a fast target calibration method and system for UAVs to solve the technical problems that the calibration of the existing UAV system based on pure vision algorithms depends on manual execution, and the calibration time is slow and the visual feedback data is not reliable. Summary of the Invention

[0007] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a fast target calibration method and system for UAVs, aiming to solve the technical problems that the calibration of the existing UAV system based on pure vision algorithms depends on manual execution, and the calibration time is slow and the visual feedback data is not reliable.

[0008] To achieve the above purpose, the present application proposes a fast target calibration system for UAVs, which is used to calibrate UAV targets for the UAV system based on pure vision algorithms, and includes:

[0009] A vision camera for collecting low-altitude image data; wherein, the low-altitude image data is used for the UAV system to perform pure vision algorithm recognition;

[0010] A calibration prototype for performing calibration flight missions; wherein, the calibration prototype continuously generates specific voiceprint signals when performing calibration flight missions;

[0011] A voiceprint collector, which is accompanied by a vision camera and is used to collect voiceprint signals around the vision camera;

[0012] A prototype console for controlling the calibration prototype to perform calibration flight missions;

[0013] A software monitoring console for monitoring parameter information of each software layer inside the UAV system and recording timestamps;

[0014] A calibration console, which is electrically connected to the vision camera, the voiceprint collector, the prototype console and the software monitoring console respectively, generates corresponding calibration flight missions according to the voiceprint signals and low-altitude image data, and when performing calibration flight missions, analyzes the prototype position information and prototype speed information of the calibration prototype through the voiceprint signals and low-altitude image data, and compares the information with the parameter information of the software layer to complete the corresponding target calibration mission;

[0015] Wherein, the calibration console analyzes spatial angle data through low-altitude image data, and analyzes relative distance data through specific voiceprint signals; converts the spatial angle data and relative distance data into prototype position information, combines the change situation of the prototype position information with the timestamp to obtain the prototype speed information.

[0016] As a further solution, the calibration flight missions include a linear away mission, a linear approach mission, an aerial hovering mission, a planar cross flight mission, a spatial oblique angle flight mission, a multi-type combination mission and a custom route mission; wherein, the multi-type combination mission is a combination of the linear away mission, the linear approach mission, the aerial hovering mission, the planar cross flight mission and the spatial oblique angle flight mission.

[0017] As a further solution, the target calibration missions include an initial verification mission, a parameter adjustment mission and a calibration verification mission; wherein, the initial verification mission is used to verify whether the prototype position information and prototype speed information of the analysis calibration prototype are credible; the parameter adjustment mission is used to compare the information of the prototype position information and prototype speed information with the parameter information of the software layer, and adjust the error parameters to obtain calibration parameter information; the calibration verification mission is used to verify whether the calibration parameter information is credible.

[0018] On the other hand, the present invention provides a fast target calibration method for UAVs, which is applied to a fast target calibration system for UAVs as described in any one of the above, and includes the following steps:

[0019] Step 1: Start the calibration prototype to hover at low altitude and emit a specific voiceprint signal, and continuously obtain the voiceprint signal and low-altitude image data through the visual camera and the voiceprint collector;

[0020] Step 2: The calibration console controls the calibration prototype to perform the calibration flight task through the prototype console, and reads the voiceprint signal of the voiceprint collector and the low-altitude image data of the visual camera;

[0021] Step 3: Execute the initial verification task to analyze whether the prototype position information and prototype speed information of the calibration prototype are credible; if credible, execute Step 4, otherwise, rapid target calibration cannot be performed;

[0022] Step 4: Execute the parameter adjustment task. The calibration console reads the parameter information of the software layer through the software monitoring console, compares it with the prototype position information and prototype speed information of the parsed calibration prototype, and adjusts the error parameters to obtain the calibration parameter information calibration task;

[0023] Step 5: Execute the calibration verification task. The drone system analyzes the prototype position information and prototype speed information of the calibration prototype through the calibration parameter information. If it matches the prototype position information and prototype speed information of the calibration prototype parsed by the calibration console, the calibration verification is successful and a rapid target calibration is completed; otherwise, re-execute Step 4, and if the verification times are exceeded, the rapid target calibration fails.

[0024] As a further solution, the initial verification task is executed through the following steps:

[0025] Execute voiceprint initial verification through the voiceprint collector, and execute the next step when the voiceprint initial verification is successful;

[0026] Execute visual initial verification through the visual camera, and execute the next step when the visual initial verification is successful;

[0027] Execute line initial verification through the voiceprint collector and the visual camera; if the verification is successful, the prototype position information and prototype speed information of the calibration prototype parsed by the calibration console are credible; otherwise, rapid target calibration cannot be performed.

[0028] As a further solution, the voiceprint initial verification is executed through the following steps:

[0029] Collect the voiceprint signal through the voiceprint collector; among them, if a specific voiceprint signal is collected and the voiceprint wavelength is the same as the preset value, execute the next step; otherwise, the voiceprint initial verification fails;

[0030] The prototype console controls the calibration prototype to perform a linear away task or a linear approach task with the voiceprint collector as the origin and at a specified flight speed;

[0031] The voiceprint collector continuously collects specific voiceprint signals when performing a linear away task or a linear approaching task, and performs wavelength velocity analysis to obtain the corresponding voiceprint analysis velocity;

[0032] Compare whether the specified flight speed matches the voiceprint analysis speed; if so, proceed to the next step, otherwise, the initial voiceprint verification fails;

[0033] The prototype console controls the calibrated prototype to perform an aerial hovering task with the voiceprint collector as the origin and a specified flight radius;

[0034] The calibration console combines the origin position coordinates, the voiceprint analysis speed, and the timestamp to analyze and obtain relative distance data and compare it with the flight radius; if they are the same, the initial voiceprint verification is successful, otherwise, the initial voiceprint verification fails.

[0035] As a further solution, the visual initial verification is performed through the following steps:

[0036] Collect low-altitude image data through a visual camera; among them, if the image of the calibrated prototype is collected and can be tracked normally, proceed to the next step; otherwise, the visual initial verification fails;

[0037] The prototype console controls the calibrated prototype to perform a planar side flight task or a spatial diagonal flight task with the visual camera as the origin and a specified flight distance;

[0038] The visual camera continuously collects images of the calibrated prototype when performing a planar side flight task or a spatial diagonal flight task, and analyzes and obtains spatial angle data through comparison of upper and lower frame images;

[0039] The calibration console combines the origin position coordinates, the voiceprint analysis speed, and the timestamp to analyze and obtain relative distance data;

[0040] The calibration console combines the origin position coordinates, the spatial angle data, the relative distance data, and the camera pitch angle data to calculate the prototype position information;

[0041] Compare whether the change amount of the prototype position information matches the flight distance; if so, the visual initial verification is successful, otherwise, the visual initial verification fails.

[0042] As a further solution, the line initial verification is performed through the following steps:

[0043] Set the custom line, custom speed, and custom angle used for the custom line task;

[0044] The prototype console controls the calibrated prototype to perform the custom line task;

[0045] Parse the prototype position information and spatial angle data of the calibrated prototype through the calibration console;

[0046] The prototype speed information is obtained by combining the change of the prototype position information with the time stamp;

[0047] Match each prototype position information with the custom route, the custom speed with the prototype speed information, and the custom angle with the spatial angle data;

[0048] Among them, if all are matched, the prototype position information and the prototype speed information of the calibrated prototype are credible; otherwise, the initial verification of the route fails.

[0049] As a further solution, the parameter adjustment task is executed through the following steps:

[0050] Start the UAV system and monitor the parameter information through the software monitoring console;

[0051] The calibration console controls the calibrated prototype to execute multi-type combination tasks through the prototype console; among them,

[0052] When executing the linear away task or the linear approaching task, the relative distance data is obtained by analysis, and the distance parameters of the UAV system are adjusted;

[0053] When executing the hovering task in the air or the horizontal flight task on the plane, the spatial angle data is obtained by analysis, and the angle parameters of the UAV system are adjusted;

[0054] When executing the spatial oblique angle flight task, the relative distance data and the spatial angle data are analyzed, and combined with the origin position coordinates, the spatial angle data, the relative distance data and the camera pitch angle data, the prototype position information is calculated;

[0055] The position parameters of the UAV system are adjusted through the prototype position information;

[0056] The prototype speed information is obtained by combining the change of the prototype position information with the time stamp, and the speed parameters of the UAV system are adjusted.

[0057] As a further solution, the calibration verification task is executed through the following steps:

[0058] Start the UAV system and monitor the parameter information through the software monitoring console;

[0059] Set the custom route, custom speed and custom angle used in the custom route task;

[0060] The prototype console controls the calibrated prototype to execute the custom route task;

[0061] Perform UAV target recognition and tracking on the calibrated prototype through the UAV system;

[0062] Obtain the distance parameter, angle parameter, position parameter and speed parameter of the UAV system and compare them;

[0063] Among them, if the distance parameter and the position parameter define a custom route, the angle parameter matches the custom angle, and the custom speed and the speed parameter all match, the calibration verification is successful.

[0064] Compared with the related technology, a fast target calibration method and system for UAVs provided by the present invention have the following advantages:

[0065] The present invention cleverly utilizes the advantages of acoustic fingerprint ranging in linear distance and the advantages of image comparison in planar angle measurement, and then cooperates with a tightly linked data verification logic line to achieve a standardized, fast and automated target calibration task only by relying on simple equipment (the prototype console, software monitoring console and calibration console can be set through one computer). Brief Description of the Drawings

[0066] The drawings here are incorporated into the specification and form a part of this specification, showing embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application.

[0067] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the following will briefly introduce the drawings required for use in the description of the embodiments or the related technologies. Obviously, for those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0068] Figure 1 It is a schematic structural diagram of a fast target calibration system for UAVs provided by the present invention;

[0069] Figure 2 It is a schematic diagram of the steps of a fast target calibration method for UAVs provided by the present invention.

[0070] The realization of the purpose of the present application, functional features and advantages will be further described in combination with the embodiments with reference to the drawings. Detailed Embodiments

[0071] To make the purposes, technical solutions and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Usually, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0072] Please refer to Figure 1, an embodiment of the present application provides a fast target calibration system for drones, which is used to calibrate drone targets for a drone system based on a pure vision algorithm, including:

[0073] A vision camera for collecting low-altitude image data; wherein, the low-altitude image data is used for the drone system to perform pure vision algorithm recognition;

[0074] A calibration prototype for performing calibration flight tasks; wherein, the calibration prototype continuously generates a specific voiceprint signal when performing calibration flight tasks;

[0075] A voiceprint collector, which is arranged adjacently to the vision camera and is used to collect the voiceprint signal around the vision camera;

[0076] A prototype console for controlling the calibration prototype to perform calibration flight tasks;

[0077] A software monitoring console for monitoring the parameter information of each software layer inside the drone system and recording timestamps;

[0078] A calibration console, which is electrically connected to the vision camera, the voiceprint collector, the prototype console and the software monitoring console respectively, and generates corresponding calibration flight tasks according to the voiceprint signal and the low-altitude image data, and when performing the calibration flight tasks, analyzes the prototype position information and prototype speed information of the calibration prototype through the voiceprint signal and the low-altitude image data, and completes the corresponding target calibration task by comparing the information with the parameter information of the software layer;

[0079] Among them, the calibration console analyzes the spatial angle data through the low-altitude image data, and analyzes the relative distance data through the specific voiceprint signal; converts the spatial angle data and the relative distance data into prototype position information, combines the change situation of the prototype position information with the timestamp, and obtains the prototype speed information.

[0080] It should be noted that: during the calibration process, mainly the distance parameters, angle parameters, position parameters and speed parameters identified by the drone system after discovering the drone are calibrated and adjusted so that the corresponding parameter data can be accurately obtained, so as to meet the drone monitoring and positioning requirements.

[0081] However, due to the lack of other reliable data sources, the drone system based on the pure vision algorithm can only obtain parameters through image differences. Therefore, when performing drone target calibration, it is often necessary for manual personnel to go to conduct one-by-one debugging of the parameters, and the debugging methods are different, and it is very difficult to perform target calibration quickly and accurately.

[0082] The present invention conducts an in-depth analysis on this issue and finds that during the calibration process, the parameter that is prone to problems is mainly the distance parameter. The reason is that the pure vision algorithm identifies the distance by combining the target size with other image features (especially the monocular system). When the target size feature is not clear, the distance parameter obtained is also inaccurate.

[0083] In addition, although the front-to-back distance between the upper and lower frame images is not obvious, the distance when the plane moves will directly reflect the position difference of the image target, and the spatial angle data (that is, the spatial angle value) can be obtained. In other words, the spatial angle data can be obtained through the image.

[0084] Combined with the known origin position coordinates (visual camera), relative distance data and camera pitch angle data, the prototype position information can be calculated. Then, the time-varying state of the prototype position information can be calculated to obtain the corresponding prototype speed information, and the distance parameters, angle parameters, position parameters and speed parameters can be calibrated and adjusted.

[0085] According to the above analysis, the core of the problem becomes how to accurately measure the relative distance data and verify whether the measured spatial angle data is accurate. We need to accurately measure the relative distance data, which means we need to use external means, such as radar ranging, laser ranging and other technologies; among them, radar ranging requires the use of large radar equipment, which has high deployment costs, large volume, and signal interference, and is not suitable for complex electromagnetic scenes in cities. Laser ranging requires accurate projection onto the drone and accurate reception of the reflected laser, which requires high sensitivity and accuracy of the equipment.

[0086] To this end, the present invention performs distance measurement based on specific soundprint signals. The specific soundprint signals can be the sound generated by the drone itself during flight and related to the rotation speed of the wings, or can be specific audio played by installed speakers. Combined with the Doppler effect, accurate speed measurement can be achieved. Based on the known origin position coordinates and timestamp, the corresponding relative distance data can be calculated.

[0087] The Doppler effect is: the received frequency of the wave becomes higher when the wave source moves towards the observer, and the received frequency becomes lower when the wave source moves away from the observer; where, assuming that the wavelength of the original wave source is λ, the wave speed is c, and the observer's moving speed is v: when the observer approaches the wave source, the observed wave source frequency is (c+v) / λ, if the observer moves away from the wave source, the observed wave source frequency is (cv) / λ, therefore, when the wavelength and source frequency are known, we can calculate the corresponding moving speed v.

[0088] The advantages of doing this are as follows: distance measurement can be carried out with simple devices, and the propagation characteristics of waves determine that they spread outwards. There is no need for target positioning. Only by setting up a voiceprint collector to identify and collect specific voiceprint signals can the wavelength and wave source frequency be obtained. Then, combined with the sound propagation speed of the specific voiceprint signal, the moving speed of the calibration prototype (the linear moving speed with respect to the origin position coordinates) can be calculated. However, when it comes to the planar moving speed, the change is not obvious, but fortunately, we can make up for this part through image recognition, thus forming a complete data verification logic line.

[0089] That is, the data verification logic line of this solution is: first, verify whether the control amounts of voiceprint ranging and the prototype console in the front-back distance match. If they match, it indicates that the voiceprint ranging is accurate and the prototype console can accurately control the calibration prototype. Based on this, then control the UAV to fly horizontally in a plane through the prototype console and judge whether the spatial angle data is accurately measured. Finally, based on the accurate relative distance data and spatial angle data, calculate the corresponding prototype position information and prototype speed information, and then provide an adjustment basis for the distance parameter, angle parameter, position parameter, and speed parameter.

[0090] Therefore, the calibration flight tasks include linear moving away tasks, linear moving closer tasks, hovering tasks in the air, horizontal flying tasks in a plane, spatial oblique flying tasks, multi-type combination tasks, and custom route tasks; among them,

[0091] The linear moving away tasks and linear moving closer tasks are linear motions relative to the origin, which can minimize the influence of planar movement on voiceprint ranging to the greatest extent. The hovering tasks in the air are to hover around the origin. Mainly, in the case of the same distance but different angles, judge whether the voiceprint ranging can accurately measure the hovering radius.

[0092] The horizontal flying tasks in a plane are to fly horizontally on a plane, which can maximize the influence of flight on the spatial included angle value in the low-altitude image data. The corresponding spatial oblique flying tasks are to combine linear flight and horizontal flight and fly at an oblique angle in the spatial position (executed when the relative distance data can be accurately measured). The multi-type combination tasks are for comprehensive verification to judge whether the set parameter values match the actually measured parameter values. The custom route tasks are to cover all aspects of parameters during parameter adjustment, so as to perform all-round parameter adjustment.

[0093] Specifically, the target calibration tasks include initial verification tasks, parameter adjustment tasks, and calibration verification tasks; among them, the initial verification tasks are used to verify whether the prototype position information and prototype speed information of the parsed calibration prototype are credible; the parameter adjustment tasks are used to compare the prototype position information and prototype speed information with the parameter information in the software layer and adjust the error parameters to obtain the calibration parameter information; the calibration verification tasks are used to verify whether the calibration parameter information is credible.

[0094] Please refer to Figure 2 , the present invention provides a fast target calibration method for an unmanned aerial vehicle (UAV). In a fast target calibration system for a UAV as described in any one of the above, the method includes the following steps:

[0095] Step 1: Start the calibration prototype to hover at low altitude and emit a specific voiceprint signal, and continuously obtain the voiceprint signal and low-altitude image data through a visual camera and a voiceprint collector;

[0096] Step 2: The calibration console controls the calibration prototype to perform a calibration flight mission through the prototype console, and reads the voiceprint signal of the voiceprint collector and the low-altitude image data of the visual camera;

[0097] Step 3: Execute an initial verification task to analyze whether the prototype position information and prototype speed information of the calibration prototype are credible; if credible, execute Step 4, otherwise, fast target calibration cannot be performed;

[0098] Step 4: Execute a parameter adjustment task. The calibration console reads the parameter information of the software layer through the software monitoring console, compares it with the prototype position information and prototype speed information of the parsed calibration prototype, and adjusts the error parameters to obtain the calibration parameter information calibration task;

[0099] Step 5: Execute a calibration verification task. The UAV system analyzes the prototype position information and prototype speed information of the calibration prototype through the calibration parameter information. If it matches the prototype position information and prototype speed information of the calibration prototype parsed by the calibration console, the calibration verification is successful, and a fast target calibration is completed; otherwise, re-execute Step 4. If the verification times are exceeded, the fast target calibration fails.

[0100] Specifically, the initial verification task is executed through the following steps:

[0101] Execute voiceprint initial verification through the voiceprint collector, and execute the next step when the voiceprint initial verification is successful;

[0102] Execute visual initial verification through the visual camera, and execute the next step when the visual initial verification is successful;

[0103] Execute line initial verification through the voiceprint collector and the visual camera; if the verification is successful, the prototype position information and prototype speed information of the calibration prototype parsed by the calibration console are credible; otherwise, fast target calibration cannot be performed.

[0104] More specifically, the voiceprint initial verification is executed through the following steps:

[0105] Collect the voiceprint signal through the voiceprint collector; wherein, if a specific voiceprint signal is collected and the voiceprint wavelength is the same as the preset value, execute the next step; otherwise, the voiceprint initial verification fails;

[0106] The prototype console controls the calibrated prototype to perform a linear away task or a linear approach task with the voiceprint collector as the origin and at a specified flight speed;

[0107] The voiceprint collector continuously collects specific voiceprint signals during the execution of the linear away task or the linear approach task, and performs wavelength speed analysis to obtain the corresponding voiceprint analysis speed;

[0108] Compare whether the specified flight speed and the voiceprint analysis speed match; if so, proceed to the next step, otherwise, the initial voiceprint verification fails;

[0109] The prototype console controls the calibrated prototype to perform an aerial hovering task with the voiceprint collector as the origin and at a specified flight radius;

[0110] The calibration console combines the origin position coordinates, the voiceprint analysis speed, and the timestamp to analyze and obtain relative distance data and compare it with the flight radius; if they are the same, the initial voiceprint verification is successful, otherwise, the initial voiceprint verification fails.

[0111] Furthermore, the visual initial verification is performed through the following steps:

[0112] Collect low-altitude image data through the visual camera; among them, if the calibrated prototype image is collected and can be tracked normally, proceed to the next step; otherwise, the visual initial verification fails;

[0113] The prototype console controls the calibrated prototype to perform a planar lateral flight task or a spatial oblique angle flight task with the visual camera as the origin and at a specified flight distance;

[0114] The visual camera continuously collects the calibrated prototype images during the execution of the planar lateral flight task or the spatial oblique angle flight task, and analyzes and obtains spatial angle data through comparison of the upper and lower frame images;

[0115] The calibration console combines the origin position coordinates, the voiceprint analysis speed, and the timestamp to analyze and obtain relative distance data;

[0116] The calibration console combines the origin position coordinates, the spatial angle data, the relative distance data, and the camera pitch angle data to calculate the prototype position information;

[0117] Compare whether the change amount of the prototype position information matches the flight distance; if so, the visual initial verification is successful, otherwise, the visual initial verification fails.

[0118] Even further, the line initial verification is performed through the following steps:

[0119] Set the custom line, custom speed, and custom angle used for the custom line task;

[0120] The prototype console controls the calibrated prototype to execute custom route tasks;

[0121] The calibration console analyzes the prototype position information and spatial angle data of the calibrated prototype;

[0122] By combining the change situation of the prototype position information with the timestamp, the prototype speed information is obtained;

[0123] Match each prototype position information with the custom route, the custom speed with the prototype speed information, and the custom angle with the spatial angle data;

[0124] Among them, if all are matched, the prototype position information and prototype speed information of the calibrated prototype are credible; otherwise, the initial verification of the route fails.

[0125] Specifically, the parameter adjustment task is executed through the following steps:

[0126] Start the UAV system and monitor the parameter information through the software monitoring console;

[0127] The calibration console controls the calibrated prototype to execute multi-type combination tasks through the prototype console; among them,

[0128] When executing the linear away task or the linear approaching task, the relative distance data is obtained through analysis, and the distance parameters of the UAV system are adjusted;

[0129] When executing the hovering task in the air or the horizontal flight task on the plane, the spatial angle data is obtained through analysis, and the angle parameters of the UAV system are adjusted;

[0130] When executing the spatial oblique angle flight task, analyze the relative distance data and the spatial angle data, and combine the origin position coordinates, the spatial angle data, the relative distance data and the camera pitch angle data to calculate the prototype position information;

[0131] Adjust the position parameters of the UAV system through the prototype position information;

[0132] By combining the change situation of the prototype position information with the timestamp, the prototype speed information is obtained, and the speed parameters of the UAV system are adjusted.

[0133] More specifically, the calibration verification task is executed through the following steps:

[0134] Start the UAV system and monitor the parameter information through the software monitoring console;

[0135] Set the custom route, custom speed and custom angle used in the custom route task;

[0136] The prototype console controls the calibrated prototype to execute the custom route task;

[0137] Perform UAV target recognition and tracking on the calibration prototype through the UAV system;

[0138] Obtain the distance parameter, angle parameter, position parameter and speed parameter of the UAV system and compare them;

[0139] Among them, if the distance parameter and the position parameter define a custom route, the angle parameter matches the custom angle, and the custom speed and the speed parameter all match, the calibration verification is successful.

[0140] The above are only partial embodiments of the present application, and do not limit the patent scope of the present application. Any equivalent structural transformation made under the technical concept of the present application by using the content of the specification and drawings of the present application, or any direct / indirect application in other related technical fields is included in the patent protection scope of the present application.

Claims

1. A fast target calibration system for unmanned aerial vehicles, which is used to calibrate the targets of unmanned aerial vehicles in an unmanned aerial vehicle system based on a pure vision algorithm, and is characterized in that, Including: A visual camera, which is set on the ground for collecting low-altitude image data; wherein, the low-altitude image data is used for the UAV system to perform pure vision algorithm recognition; A calibration prototype, which is used to execute the calibration flight mission; wherein, the calibration prototype continuously generates a specific voiceprint signal when executing the calibration flight mission; A voiceprint collector, which is accompanied with the visual camera and set on the ground, and is used for collecting the voiceprint signal around the visual camera; A prototype console, which is used to control the calibration prototype to execute the calibration flight mission; A software monitoring console, which is used to monitor the parameter information of each software layer inside the UAV system and record the timestamp; A calibration console, which is electrically connected to the visual camera, the voiceprint collector, the prototype console and the software monitoring console respectively, and generates a corresponding calibration flight mission according to the voiceprint signal and the low-altitude image data, and when executing the calibration flight mission, analyzes the prototype position information and the prototype speed information of the calibration prototype through the voiceprint signal and the low-altitude image data, and compares with the parameter information of the software layer to complete the corresponding target calibration task; Wherein, the calibration flight mission includes a linear away mission, a linear approach mission, a planar cross flight mission and a spatial oblique angle flight mission; the target calibration task includes an initial verification task, which is used to verify whether the prototype position information and the prototype speed information of the analyzed calibration prototype are credible, including voiceprint initial verification and visual initial verification; When performing the voiceprint initial verification, it is judged by matching the specified flight speed and the voiceprint analysis speed; wherein, the voiceprint collector continuously collects the specific voiceprint signal when executing the linear away mission or the linear approach mission, and performs wavelength speed analysis to obtain the corresponding voiceprint analysis speed; When performing the visual initial verification: it is judged by matching the change amount of the prototype position information and the flight distance; wherein, the visual camera continuously collects the calibration prototype images when executing the planar cross flight mission or the spatial oblique angle flight mission, and analyzes to obtain the spatial angle data through the comparison of the upper and lower frame images; the calibration console combines the origin position coordinates, the spatial angle data, the relative distance data and the camera pitch angle data, calculates to obtain the prototype position information, and then combines the change situation of the prototype position information with the timestamp to obtain the prototype speed information.

2. The rapid target calibration system for an unmanned aerial vehicle according to claim 1, wherein The calibration flight mission further includes an in-air hovering mission, a multi-type combination mission and a custom route mission; wherein, the multi-type combination mission is a combination of the linear away mission, the linear approach mission, the in-air hovering mission, the planar cross flight mission and the spatial oblique angle flight mission.

3. The rapid target calibration system for an unmanned aerial vehicle according to claim 1, characterized in that, The target calibration task further includes a parameter adjustment task and a calibration verification task; wherein, the parameter adjustment task is used to compare the prototype position information and the prototype speed information with the parameter information of the software layer, and adjust the error parameters to obtain the calibration parameter information; the calibration verification task is used to verify whether the calibration parameter information is credible.

4. A rapid target calibration method for an unmanned aerial vehicle, which is applied to a rapid target calibration system for an unmanned aerial vehicle according to any one of claims 1 to 3, characterized in that, Including the following steps: Step 1: Start the calibration prototype to hover low in the air and emit a specific voiceprint signal, and continuously obtain the voiceprint signal and the low-altitude image data through the visual camera and the voiceprint collector; Step 2: Calibration Console Control The calibration console controls the prototype through the prototype console to perform a calibration flight mission, and reads the voiceprint signal of the voiceprint collector and the low-altitude image data of the vision camera; Step 3: Execute the Initial Verification Task Analyze whether the prototype position information and prototype speed information of the calibration prototype are credible; if credible, execute Step 4, otherwise, fast target calibration cannot be performed; 5. A rapid target calibration method for an unmanned aerial vehicle according to claim 4, characterized in that Step 4: Execute the Parameter Adjustment Task The calibration console reads the parameter information of the software layer through the software monitoring console, compares it with the prototype position information and prototype speed information parsed from the calibration prototype, and adjusts the error parameters to obtain the calibration parameter information calibration task; Step 5: Execute the Calibration Verification Task The UAV system analyzes the prototype position information and prototype speed information of the calibration prototype through the calibration parameter information. If it matches the prototype position information and prototype speed information parsed by the calibration console, the calibration verification is successful, and a fast target calibration is completed; otherwise, re-execute Step 4. If the verification times are exceeded, the fast target calibration fails. The initial verification task is executed through the following steps:

6. A rapid target calibration method for an unmanned aerial vehicle according to claim 5, characterized in that, Execute voiceprint initial verification through the voiceprint collector, and execute the next step when the voiceprint initial verification is successful; Execute visual initial verification through the vision camera, and execute the next step when the visual initial verification is successful; Execute line initial verification through the voiceprint collector and the vision camera; If the verification is successful, the prototype position information and prototype speed information parsed by the calibration console are credible; otherwise, fast target calibration cannot be performed. The voiceprint initial verification is executed through the following steps: Collect the voiceprint signal through the voiceprint collector; among them, if a specific voiceprint signal is collected and the voiceprint wavelength is the same as the preset value, execute the next step; otherwise, the voiceprint initial verification fails; The prototype console controls the calibration prototype to perform a linear away task or a linear approach task with the voiceprint collector as the origin and a specified flight speed; 7. A rapid target calibration method for an unmanned aerial vehicle according to claim 6, characterized in that, The voiceprint collector continuously collects the specific voiceprint signal during the linear away task or the linear approach task, and performs wavelength speed analysis to obtain the corresponding voiceprint analysis speed; Compare whether the specified flight speed and the voiceprint analysis speed match; if so, execute the next step, otherwise, the voiceprint initial verification fails; The prototype console controls the calibration prototype to perform an aerial hovering task with the voiceprint collector as the origin and a specified flight radius; The calibration console combines the origin position coordinates, the voiceprint analysis speed, and the timestamp to analyze and obtain the relative distance data and compare it with the flight radius; if they are the same, the voiceprint initial verification is successful, otherwise, the voiceprint initial verification fails. The visual initial verification is executed through the following steps: Collect low-altitude image data through the vision camera; Among them, if the calibration prototype image is collected and can be tracked normally, execute the next step; otherwise, the visual initial verification fails; The prototype console controls the calibration prototype to perform a planar cross-flight mission or a spatial diagonal flight mission with the vision camera as the origin and a specified flight distance; The vision camera continuously collects the calibration prototype images during the planar cross-flight mission or the spatial diagonal flight mission, and analyzes and obtains the spatial angle data through the comparison of the upper and lower frame images; The calibration console combines the origin position coordinates, voiceprint analysis speed, and timestamp to parse and obtain relative distance data; The calibration console combines the origin position coordinates, spatial angle data, relative distance data, and camera pitch angle data to calculate the prototype position information; Compare whether the change in the prototype position information matches the flight distance; If so, the visual initial verification is successful; otherwise, the visual initial verification fails.

8. A rapid target calibration method for an unmanned aerial vehicle according to claim 7, characterized in that, The initial verification of the line is performed through the following steps: Set the custom line, custom speed, and custom angle used for the custom line task; The prototype console controls the calibration prototype to execute the custom line task; The calibration console parses the prototype position information and spatial angle data of the calibration prototype; Combining the change situation of the prototype position information with the timestamp to obtain the prototype speed information; Match each prototype position information with the custom line, the custom speed with the prototype speed information, and the custom angle with the spatial angle data; Among them, if all match, the prototype position information and prototype speed information of the calibration prototype are credible; otherwise, the initial verification of the line fails.

9. A rapid target calibration method for an unmanned aerial vehicle according to claim 8, characterized in that The parameter adjustment task is performed through the following steps: Start the UAV system and monitor the parameter information through the software monitoring console; The calibration console controls the calibration prototype to execute multi-type combination tasks through the prototype console; among them, When performing a linear away task or a linear approaching task, parse the relative distance data and adjust the distance parameter of the UAV system; When performing an aerial hovering task or a planar side flight task, parse the spatial angle data and adjust the angle parameter of the UAV system; When performing a spatial oblique angle flight task, parse the relative distance data and spatial angle data, and combine the origin position coordinates, spatial angle data, relative distance data, and camera pitch angle data to calculate the prototype position information; Adjust the position parameter of the UAV system through the prototype position information; Combining the change situation of the prototype position information with the timestamp to obtain the prototype speed information, and adjust the speed parameter of the UAV system.

10. A rapid target calibration method for an unmanned aerial vehicle according to claim 9, characterized in that The calibration verification task is performed through the following steps: Start the UAV system and monitor the parameter information through the software monitoring console; Set the custom line, custom speed, and custom angle used for the custom line task; The prototype console controls the calibration prototype to execute the custom line task; Perform UAV target recognition and tracking on the calibration prototype through the UAV system; Obtain and compare the distance parameter, angle parameter, position parameter, and speed parameter of the UAV system; Among them, if the distance parameter and position parameter match the custom line, the angle parameter matches the custom angle, and the custom speed and speed parameter match, the calibration verification is successful.

Images