Method and device for evaluating disposal efficiency of unmanned aerial vehicle, equipment and storage medium

Abstract

The invention is applied to the field of unmanned aerial vehicle disposal, and discloses a method and device for evaluating the disposal efficiency of an unmanned aerial vehicle, equipment and a storage medium. The method comprises the following steps: performing unmanned aerial vehicle detection by using a plurality of detection sensors; acquiring initial detection data corresponding to the target unmanned aerial vehicle detected by each detection sensor; disposing the target unmanned aerial vehicle; obtaining treatment detection data corresponding to the target unmanned aerial vehicle detected by each detection sensor after treatment; and inputting the initial detection data and the treatment detection data into a preset agent treatment efficiency evaluation model, and evaluating the treatment of the target unmanned aerial vehicle by using agent evaluation to obtain an evaluation result. Therefore, compared with the prior art in which the evaluation result is obtained through a manual discrimination mode, the evaluation result is obtained by inputting the detection data of each detection sensor before and after the target unmanned aerial vehicle is disposed into the preset agent disposal efficiency evaluation model for evaluation, and automatic evaluation of the disposal efficiency of the unmanned aerial vehicle is realized.

Description

Technical Field

[0001] This invention relates to the field of drone disposal, and more specifically to a method, apparatus, equipment, and storage medium for evaluating the effectiveness of drone disposal. Background Technology

[0002] With the rapid rise of the low-altitude economy, drones are increasingly used in logistics, emergency rescue, and urban governance. However, the risks to airspace safety, public privacy, and security for major events caused by their "black flights" and "reckless flights" are becoming increasingly prominent. Incidents of illegal drones intruding into airport airspace and key locations are frequent, posing a severe challenge to low-altitude order management and emergency response capabilities. To address this issue, various detection and handling equipment, including radar detection equipment, photoelectric detection equipment, radio detection equipment, RID (Remote Identification) ground receiving equipment, and radio jamming equipment, are widely used in the monitoring and handling of low-altitude drones. Radar detection equipment enables long-range, wide-area searches; photoelectric detection equipment provides visual and efficient identification; radio detection equipment captures signal trajectories; RID ground receiving equipment acquires drone identity information; and radio jamming equipment achieves precise interference with the telemetry and control link. The coordinated scheduling of these five types of detection equipment has become the core support for achieving full-process management of drones from "discovery to identification to location to handling."

[0003] In current technologies, the results of drone disposal are generally obtained through manual judgment. However, as the number of drone owners increases, the control of low-altitude environments becomes increasingly difficult. With the growing number of civilian consumer drones entering daily life, the workload for drone detection and countermeasures is also increasing. Manual judgment of disposal results is no longer sufficient to meet the automation needs of low-altitude drone disposal. Summary of the Invention

[0004] The technical problem to be solved by the present invention is how to realize the automatic evaluation of the handling efficiency of unmanned aerial vehicles (UAVs). The purpose is to provide a method, device, equipment, and storage medium for evaluating the handling efficiency of UAVs, so as to realize the automatic evaluation of the handling efficiency of UAVs.

[0005] This invention is achieved through the following technical solution:

[0006] In a first aspect, a method for evaluating the effectiveness of handling unmanned aerial vehicles (UAVs) includes: detecting UAVs using several types of detection sensors; acquiring initial detection data corresponding to the target UAV detected by each of the detection sensors; handling the target UAV; acquiring handling detection data corresponding to the target UAV detected by each of the detection sensors after handling; inputting the initial detection data and the handling detection data into a preset intelligent agent handling effectiveness evaluation model to evaluate the handling of the target UAV using intelligent agent evaluation, and obtaining an evaluation result.

[0007] In some embodiments, the detection sensor includes: radar detection equipment, photoelectric detection equipment, radio detection equipment, and UAV remote identification ground receiving equipment.

[0008] In some embodiments, the initial detection data includes an initial UAV state dataset and an initial detection device state set; wherein, the initial UAV state dataset includes initial size data, initial payload data, first initial UAV spatial coordinates, third initial UAV spatial coordinates, initial flight speed, and initial flight compliance parameters corresponding to the target UAV; the initial size data and the initial payload data are obtained by the photoelectric detection device; the first initial UAV spatial coordinates and the initial flight speed are obtained by the radar detection device; the initial detection device state set includes initial state data corresponding to each of the detection sensors; and the third initial UAV spatial coordinates are obtained by the radio detection device.

[0009] In some embodiments, the process of evaluating the handling of the target drone using an intelligent agent to obtain an evaluation result includes: using the intelligent agent to fuse the initial detection data according to a preset device fusion weight to obtain initial fused data; the device fusion weight is obtained by training the intelligent agent's handling efficiency evaluation model; using the intelligent agent to fuse the handling detection data according to the device fusion weight to obtain handling fused data; using the intelligent agent to obtain the initial threat level corresponding to the target drone according to a preset threat weight and the initial detection data; using the intelligent agent to obtain the handling threat level corresponding to the handling of the target drone according to the threat weight and the handling detection data; the threat weight is obtained by training the intelligent agent's handling efficiency evaluation model; using the intelligent agent to determine the target action in a preset action space according to the initial threat level and the handling threat level; and determining the target action as the evaluation result.

[0010] In some embodiments, the initial state data corresponding to the photoelectric detection device includes: first initial turntable data, initial device coordinates, and initial installation attitude; the initial state data corresponding to the radio detection device includes: second initial turntable data; the device fusion weight includes photoelectric fusion weight, radio fusion weight, and radar fusion weight; the initial fusion data includes the initial fusion spatial coordinates corresponding to the target UAV; the initial fusion spatial coordinates are obtained by: obtaining the initial coordinates corresponding to the target UAV; the initial coordinates include the initial guidance latitude and longitude and the initial altitude corresponding to the target UAV; obtaining the initial radian latitude and longitude corresponding to the initial guidance latitude and longitude; obtaining the initial distance between the photoelectric detection device and the initial coordinates based on the initial radian latitude and longitude and the initial device coordinates; fusing the first initial turntable data and the second initial turntable data using the photoelectric fusion weight and the radio fusion weight to obtain the initial fusion turntable data; obtaining the second initial UAV spatial coordinates based on the initial distance, the initial fusion turntable data, and the initial installation attitude; fusing the first initial UAV spatial coordinates, the third initial UAV spatial coordinates, and the second initial UAV spatial coordinates using the photoelectric fusion weight, the radio fusion weight, and the radar fusion weight to obtain the initial fusion spatial coordinates.

[0011] In some embodiments, the agent's handling performance evaluation model is obtained by: using several detection sensors to detect a preset reference drone; obtaining initial training detection data corresponding to the reference drone detected by each of the detection sensors; handling the reference drone and obtaining handling reference results; obtaining handling training detection data corresponding to the reference drone detected by each of the detection sensors after handling; inputting the initial training detection data and the handling training detection data into a preset initial agent handling performance evaluation model, so as to use the agent to train the initial agent handling performance evaluation model and obtain the agent handling performance evaluation model.

[0012] In some embodiments, the state space of the initial agent handling effectiveness evaluation model includes a pre-handling state vector and a post-handling state vector; the pre-handling state vector is obtained through the initial training probe data; the post-handling state vector is obtained through the handling training probe data; the action space of the agent handling effectiveness evaluation model represents the space of handling evaluation results; the reward function of the agent handling effectiveness evaluation model includes a threat change reward, an evaluation consistency reward, and a data reliability reward; wherein, the threat change reward is obtained through the initial training threat and the handling training threat; the initial training threat is obtained through a preset training threat weight and the initial training probe data; the handling training threat is obtained through the training threat weight and the handling training probe data; the evaluation consistency reward is obtained through the handling reference result and the training reference action; the training reference action is obtained by the agent in the action space based on the initial training threat and the handling training threat; the data reliability reward is obtained through the initial training threat and / or the handling training threat; the agent trains the initial agent handling effectiveness evaluation model by maximizing the reward function to obtain the agent handling effectiveness evaluation model.

[0013] Secondly, an apparatus for evaluating the effectiveness of handling unmanned aerial vehicles (UAVs) includes: a detection module configured to detect UAVs using a plurality of detection sensors; an initial detection data acquisition module configured to acquire initial detection data corresponding to the target UAV detected by each of the detection sensors; a handling module configured to handle the target UAV; a handling detection data acquisition module configured to acquire handling detection data corresponding to the target UAV detected by each of the detection sensors after handling; and an evaluation module configured to input the initial detection data and the handling detection data into a preset intelligent agent handling effectiveness evaluation model to evaluate the handling of the target UAV using intelligent agent evaluation and obtain an evaluation result.

[0014] Thirdly, an electronic device includes: a processor and a memory storing program instructions, the processor being configured to execute the above-described method for evaluating the handling effectiveness of a drone when the program instructions are executed.

[0015] Fourthly, a storage medium stores program instructions that, when executed, perform the aforementioned method for evaluating the handling effectiveness of a drone.

[0016] Compared with existing technologies, this invention utilizes several detection sensors to detect drones, acquiring initial detection data corresponding to the target drone detected by each sensor. Then, the target drone is dealt with, and the post-dealt detection data corresponding to the target drone detected by each sensor is acquired. The initial detection data and the dealt detection data are then input into a preset intelligent agent dealt effectiveness evaluation model to assess the dealt effectiveness of the target drone and obtain an evaluation result. Thus, compared to existing technologies that obtain evaluation results through manual judgment, this solution achieves automatic evaluation of drone dealt effectiveness by inputting the detection data from each sensor before and after dealing with the target drone into a preset intelligent agent dealt effectiveness evaluation model, thereby using intelligent agent evaluation to assess the dealt effectiveness and obtain an evaluation result. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be considered as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort. In the drawings:

[0018] Figure 1 This is a flowchart illustrating a method for evaluating the handling performance of a drone, provided in an embodiment of this disclosure.

[0019] Figure 2 This is a flowchart of an embodiment of the present disclosure for obtaining an evaluation model of the processing efficiency of an intelligent agent;

[0020] Figure 3 This is a schematic diagram of an apparatus for evaluating the handling performance of a drone, provided in an embodiment of this disclosure;

[0021] Figure 4 This is a schematic diagram of an electronic device provided in an embodiment of this disclosure. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of this invention are only for explaining this invention and are not intended to limit this invention.

[0023] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0024] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0025] The flowcharts shown in the accompanying drawings are merely illustrative and do not necessarily include all content and operations / steps, nor do they necessarily have to be performed in the described order. For example, some operations / steps can be broken down, while others can be combined or partially combined; therefore, the actual execution order may change depending on the specific circumstances.

[0026] In this application, "multiple" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0027] Please see Figure 1 , Figure 1 This is a flowchart illustrating a method for evaluating the handling performance of a drone, as shown in an exemplary embodiment of this application.

[0028] Combination Figure 1 As shown in the embodiments of this disclosure, a method for evaluating the handling effectiveness of unmanned aerial vehicles (UAVs) is provided, the method comprising:

[0029] Step S101: Use several types of detection sensors to conduct drone detection.

[0030] It should be noted that the detection sensors include: radar detection equipment, photoelectric detection equipment, radio detection equipment, and UAV remote identification ground receiving equipment. This allows for the acquisition of detection data obtained by radar detection equipment, photoelectric detection equipment, radio detection equipment, and UAV remote identification ground receiving equipment from the target UAV, facilitating the fusion of multi-source detection data and improving the accuracy and comprehensiveness of the assessment.

[0031] It should be noted that the ground receiving device for remote drone identification is a RID ground receiving device. It is used to obtain the digital license plate of the target drone.

[0032] Step S102: Obtain the initial detection data corresponding to the target UAV detected by each detection sensor.

[0033] It should be noted that the initial detection data refers to the detection data continuously acquired by each detection sensor before the target drone was dealt with.

[0034] Specifically, the initial detection data includes an initial UAV state dataset and an initial detection device state set. The initial UAV state dataset includes the initial size data, initial payload data, first initial UAV spatial coordinates, third initial UAV spatial coordinates, initial flight speed, and initial flight compliance parameters corresponding to the target UAV. The initial size data and initial payload data are obtained through photoelectric detection equipment. The first initial UAV spatial coordinates and initial flight speed are obtained through radar detection equipment. The initial detection device state set includes the initial state data corresponding to each detection sensor. The third initial UAV spatial coordinates are obtained through radio detection equipment.

[0035] It should be noted that the initial size data corresponding to the target drone is the pixel size of the target drone obtained by the photoelectric detection equipment.

[0036] The initial payload data characterizes whether the target UAV is carrying a hazardous payload. In some embodiments, the initial payload data is 10 if the target UAV is carrying a hazardous payload, and 2 if the target UAV is not carrying a hazardous payload.

[0037] It should be noted that the initial state data corresponding to each detection sensor includes: the initial state data corresponding to the photoelectric detection equipment, the initial state data corresponding to the radio detection equipment, and the initial state data corresponding to the RID ground receiving equipment.

[0038] The initial state data for the photoelectric detection equipment includes: first initial turntable data, initial equipment coordinates, and initial installation attitude. The initial state data for the radio detection equipment includes: second initial turntable data and third initial UAV spatial coordinates. The initial state data for the RID ground receiving equipment includes: identity information integrity and compliance verification results.

[0039] It should be noted that in some embodiments, due to the increasing number of drones, sometimes target drones may not be detected by the photoelectric detection equipment. In such cases, the initial detection data will not include data acquired by the photoelectric detection equipment. Consequently, the subsequent processing detection data will also lack data acquired by the photoelectric detection equipment.

[0040] Step S103: Take action against the target drone.

[0041] It should be noted that handling the target drone includes handling the target drone according to the preset handling method.

[0042] The preset handling methods include one or more of the following: radio interference, GPS (Global Positioning System) spoofing, laser interception, and physical capture.

[0043] It should be noted that radio interference refers to interfering with the target drone at a preset interference power. For example, the preset interference power includes preset low power, preset medium power, and / or preset high power. Low power is less than medium power; medium power is less than high power.

[0044] It should be noted that when the preset handling method includes radio interference, the detection function of the radio detection equipment needs to be turned off. After the handling is completed, the detection function of the radio detection equipment should be turned back on to obtain the handling detection data corresponding to the target UAV detected by each detection sensor after the handling. In this way, since radio interference can also interfere with the detection function of the radio detection equipment, which can easily lead to errors in the acquisition of handling detection data, turning off the detection function of the radio detection equipment during radio interference and turning it back on after the handling is completed can obtain more accurate handling detection data.

[0045] Step S104: Obtain the detection data of the target UAV detected by each detection sensor after the disposal.

[0046] It should be noted that the detection data refers to the detection data continuously acquired by each detection sensor after the target drone has been dealt with.

[0047] Specifically, the disposal detection data includes a disposal drone status dataset and a disposal detection equipment status set. The disposal drone status dataset includes disposal size data, disposal payload data, spatial coordinates of the first disposal drone, spatial coordinates of the third disposal drone, disposal flight speed, and disposal flight compliance parameters corresponding to the target drone. The disposal size data and disposal payload data are obtained through photoelectric detection equipment. The spatial coordinates of the first disposal drone and disposal flight speed are obtained through radar detection equipment. The disposal detection equipment status set includes disposal status data corresponding to each detection sensor. The spatial coordinates of the third disposal drone are obtained through radio detection equipment.

[0048] It should be noted that the handling status data corresponding to each detection sensor includes: handling status data corresponding to radar detection equipment, handling status data corresponding to photoelectric detection equipment, handling status data corresponding to radio detection equipment, and handling status data corresponding to RID ground receiving equipment.

[0049] The handling status data corresponding to the photoelectric detection equipment includes: first handling turntable data, handling equipment coordinates, handling installation attitude, and handling lock status corresponding to the photoelectric detection equipment. When the handling lock status corresponding to the photoelectric detection equipment is 1, it indicates that the target UAV is locked; when the handling lock status is 0, it indicates that the target UAV is not locked.

[0050] The status data corresponding to the radio detection equipment includes: data from the second handling turntable and the spatial coordinates of the third handling UAV.

[0051] The processing status data corresponding to the RID ground receiving equipment includes: identity information integrity and compliance verification results.

[0052] Identity information integrity refers to the integrity of the digital license plate of the target drone, which is greater than or equal to 0 and less than or equal to 1. The larger the value, the more complete the digital license plate.

[0053] The compliance verification result indicates whether the target drone has applied for a permit. A compliance verification result of 1 indicates that the target drone is flying legally and has applied for a permit. A compliance verification result of 0 indicates that the target drone is flying illegally and has not applied for a permit.

[0054] Step S105: Input the initial detection data and the disposal detection data into the preset intelligent agent disposal performance evaluation model, so as to evaluate the disposal of the target UAV using intelligent agent evaluation and obtain the evaluation result.

[0055] Furthermore, the action against the target drone is evaluated using an intelligent agent, and the evaluation results are obtained. This includes: using an intelligent agent to fuse initial detection data according to preset device fusion weights to obtain initial fused data; obtaining device fusion weights by training an intelligent agent action performance evaluation model; using an intelligent agent to fuse action detection data according to device fusion weights to obtain action fused data; obtaining the initial threat level of the target drone according to preset threat weights and initial detection data; obtaining the action threat level of the target drone after action by using an intelligent agent according to threat weights and action detection data; obtaining threat weights by training an intelligent agent action performance evaluation model; determining the target action in a preset action space based on the initial threat level and the action threat level; and defining the target action as the evaluation result. In this way, by using an agent to fuse initial detection data based on device fusion weights obtained through training an agent-based response efficiency evaluation model, initial fused data is obtained. Then, the agent fuses response detection data based on device fusion weights to obtain response fused data. Next, the agent obtains the initial threat level of the target drone based on preset threat weights and the initial detection data. Finally, the agent obtains the response threat level of the target drone after response based on the threat weights and response detection data. Finally, the agent determines the target action within a preset action space based on the initial and response threat levels, and sets the target action as the evaluation result. This achieves automatic evaluation of target drone response based on the initial and response threat levels before and after response, thereby improving evaluation efficiency.

[0056] It should be noted that the initial detection data and the action detection data are input into the preset agent action performance evaluation model, that is, the initial detection data and the action detection data are determined as the state space data required for the agent to select an action. Specifically, the initial detection data is the pre-action state vector in the state space data, and the action detection data is the post-action state vector in the state space data.

[0057] It should be noted that the intelligent agent fuses the initial detection data according to preset device fusion weights. That is, for initial detection data of the same type obtained from different detection sensors, the preset device fusion weights are used to weight each data to obtain fused data of that type. For initial detection data obtained from only one detection sensor, this initial detection data can be directly identified as the corresponding fused data.

[0058] Specifically, the initial fusion data includes: initial fusion spatial coordinates, identity information integrity and compliance verification results, initial size data, initial payload data, initial flight speed, and initial lock status.

[0059] Optionally, the initial fusion data includes the initial fusion spatial coordinates corresponding to the target UAV; if there is no data obtained by the photoelectric detection device in the initial detection data, the initial fusion spatial coordinates are obtained in the following way: by fusing the spatial coordinates of the first initial UAV and the spatial coordinates of the third initial UAV using radio fusion weight and radar fusion weight, the initial fusion spatial coordinates are obtained.

[0060] Optionally, the equipment fusion weights include optoelectronic fusion weights, radio fusion weights, and radar fusion weights; the initial fusion data includes the initial fusion spatial coordinates corresponding to the target UAV; the initial fusion spatial coordinates are obtained as follows: obtain the initial coordinates corresponding to the target UAV; the initial coordinates include the initial guidance latitude and longitude and the initial altitude corresponding to the target UAV; obtain the initial radian latitude and longitude corresponding to the initial guidance latitude and longitude; obtain the initial distance between the optoelectronic detection device and the initial coordinates based on the initial radian latitude and longitude and the initial equipment coordinates; fuse the first initial turntable data and the second initial turntable data using the optoelectronic fusion weights and radio fusion weights to obtain the initial fusion turntable data; obtain the second initial UAV spatial coordinates based on the initial distance, the initial fusion turntable data, and the initial installation attitude; fuse the first initial UAV spatial coordinates, the third initial UAV spatial coordinates, and the second initial UAV spatial coordinates using the optoelectronic fusion weights, radio fusion weights, and radar fusion weights to obtain the initial fusion spatial coordinates. In this way, by obtaining the initial coordinates of the target UAV, and then obtaining the initial radian latitude and longitude corresponding to the initial guidance latitude and longitude, the initial distance between the electro-optical detection device and the initial coordinates is obtained based on the initial radian latitude and longitude and the initial device coordinates. The first and second initial turntable data are then fused using electro-optical fusion weights and radio fusion weights to obtain the initial fused turntable data. The second initial UAV spatial coordinates are obtained based on the initial distance, the initial fused turntable data, and the initial installation attitude. Finally, the first, third, and second initial UAV spatial coordinates are fused using electro-optical fusion weights, radio fusion weights, and radar fusion weights to obtain the initial fused spatial coordinates. This method can fuse the initial UAV spatial coordinates obtained from three devices—electro-optical detection device, radar detection device, and radio detection device—to obtain the initial fused spatial coordinates, enabling multi-source fusion of the target UAV's spatial coordinates and improving the accuracy and comprehensiveness of the initial fused spatial coordinates.

[0061] It should be noted that the initial coordinates of the target UAV are the initial fused spatial coordinates of the target UAV at time t-1. These include the initial guidance latitude and longitude and the initial altitude of the target UAV at time t-1.

[0062] It should be noted that the altitude in each coordinate in this scheme refers to elevation.

[0063] It should be noted that the initial latitude and longitude are in degrees. After conversion, the initial latitude and longitude are in radians.

[0064] The initial equipment coordinates include the latitude, longitude, and altitude of the photoelectric detection equipment.

[0065] Furthermore, the initial distance between the photoelectric detection device and the initial coordinates is obtained based on the initial radian latitude and longitude and the initial device coordinates. This includes: determining the sum of the height of the photoelectric detection device in the initial device coordinates and a preset Earth radius as the first distance from the photoelectric detection device to the Earth's center; determining the sum of the initial height and the preset Earth radius as the second distance from the target UAV to the Earth's center; and calculating... The radian values ​​of the corresponding central angles of the photoelectric detection device and the target UAV are obtained; among them, This is the radian value of the central angle, and its unit is radians; The latitude of the photoelectric detection equipment; This is the initial guide for latitude in the latitude and longitude coordinates; The radius of the Earth; The longitude of the photoelectric detection equipment; For the initial guidance of longitude in latitude and longitude; through calculation The initial distance between the photoelectric detection device and the initial coordinates is obtained; where, This represents the initial distance between the photoelectric detection device and the initial coordinates. This is the first distance; This is the second distance.

[0066] It should be noted that the central angle The angle formed by the line connecting the photoelectric detection device to the Earth's center and the line connecting the target UAV to the Earth's center, with the Earth's center as the center.

[0067] It should be noted that the initial distance between the photoelectric detection device and the initial coordinates is the distance between the target UAV and the photoelectric detection device at time t-1.

[0068] It should be noted that the first initial turntable data includes: the first initial azimuth angle and the first initial pitch angle of the turntable; the second initial turntable data includes: the second initial azimuth angle and the second initial pitch angle of the turntable.

[0069] The initial fusion turntable data includes the initial fusion azimuth angle and the initial fusion pitch angle of the turntable.

[0070] Furthermore, by fusing the first initial turntable data and the second initial turntable data using optoelectronic fusion weights and radio fusion weights, initial fused turntable data is obtained, including: calculating the first initial azimuth angle and the first azimuth product of the optoelectronic fusion weights; calculating the second initial azimuth angle and the second azimuth product of the radio fusion weights; determining the sum of the first azimuth product and the second azimuth product as the initial fused azimuth angle; calculating the first initial pitch angle and the first pitch product of the optoelectronic fusion weights; calculating the second initial pitch angle and the second pitch product of the radio fusion weights; and determining the sum of the first pitch product and the second pitch product as the initial fused pitch angle.

[0071] Furthermore, the second initial UAV spatial coordinates are obtained based on the initial distance, initial fusion turntable data, and initial installation attitude, including: through calculation The initial distance between the target UAV and the photoelectric detection device is obtained at time t after the target UAV has moved; where, Let t be the initial distance between the target UAV and the photoelectric detection device after the target UAV moves; The initial size data of the target UAV at time t; The initial size data of the target UAV at time t-1 is given. Based on the initial fused pitch angle and the initial distance between the target UAV and the photoelectric detection device at time t after the target UAV moves, the spherical coordinates of the target UAV are determined; the spherical coordinates are converted into carrier coordinates with the photoelectric device as the origin; the carrier coordinate increment is obtained based on the carrier coordinate increment and the initial installation attitude to obtain the second initial UAV spatial coordinates.

[0072] Specifically, the spherical coordinates are (Az, El, R2). Where Az is the initial fused azimuth angle; El is the initial fused pitch angle.

[0073] Furthermore, the spherical coordinates are converted into carrier coordinates with the optoelectronic device as the origin, including: obtaining the horizontal axis coordinate of the carrier coordinates by calculating X1=R2×cos(El) ×sin(Az); where X1 is the horizontal axis coordinate of the rectangular coordinates; obtaining the vertical axis coordinate of the carrier coordinates by calculating Y1=R2×cos(El) ×cos(Az); where Y1 is the vertical axis coordinate of the rectangular coordinates; and obtaining the vertical axis coordinate of the carrier coordinates by calculating Z1=R2×sin(El); where Z1 is the vertical axis coordinate of the rectangular coordinates.

[0074] It should be noted that the target carrier coordinates at time t, with the optoelectronic device as the origin, are the carrier coordinates corresponding to the target UAV. Using the same method, the carrier coordinates at time t-1, with the optoelectronic device as the origin, can also be obtained, which will not be elaborated upon here.

[0075] The carrier coordinate increments include: horizontal axis increment, vertical axis increment, and vertical axis increment.

[0076] Specifically, the difference between the horizontal axis coordinate at time t and the horizontal axis coordinate at time t-1 can be used to obtain the horizontal axis increment, which is the amount of movement of the target UAV in the horizontal direction from time t-1 to time t; the difference between the vertical axis coordinate at time t and the vertical axis coordinate at time t-1 can be used to obtain the vertical axis increment, which is the amount of movement of the target UAV in the vertical direction from time t-1 to time t; and the difference between the vertical axis coordinate at time t and the vertical axis coordinate at time t-1 can be used to obtain the vertical axis increment, which is the amount of movement of the target UAV in the vertical direction from time t-1 to time t.

[0077] Furthermore, the second initial UAV spatial coordinates are obtained based on the carrier coordinate increment and the initial installation attitude, including: converting the carrier coordinate increment into a coordinate increment in the northeast-northeast coordinate system based on the initial installation attitude; and obtaining the second initial UAV spatial coordinates based on the coordinate increment in the northeast-northeast coordinate system and the initial equipment coordinates.

[0078] It should be noted that converting the carrier coordinates to the northeast-sky coordinate system essentially involves three consecutive rotations, in the following order: yaw rotation, pitch rotation, and roll rotation. The total rotation matrix after these three rotations is the product of the rotation matrices corresponding to the three rotations.

[0079] It should be noted that yaw rotation is rotation around the U-axis in the northeast-sky coordinate system; pitch rotation is rotation around the E-axis in the northeast-sky coordinate system; and roll rotation is rotation around the N-axis in the northeast-sky coordinate system.

[0080] In some embodiments, converting the carrier coordinate increments into coordinate increments in the northeast-northeast coordinate system based on the initial installation attitude includes: obtaining a first rotation matrix corresponding to the yaw angle rotation, a second rotation matrix corresponding to the pitch angle rotation, and a third rotation matrix corresponding to the roll angle rotation based on the initial installation attitude; determining the total rotation matrix by multiplying the first, second, and third rotation matrices; and obtaining the coordinate increments in the northeast-northeast coordinate system based on the total rotation matrix and the carrier coordinate increments.

[0081] It should be noted that the initial installation attitude includes: the roll angle, heading angle, and pitch angle of the photoelectric detection equipment.

[0082] Specifically, the first rotation matrix is ,in, Let be the first rotation matrix. The attitude heading angle for the photoelectric detection equipment. The first rotation matrix allows the horizontal axis of the rectangular coordinate system to be aligned with the north-northeast coordinate system. Angular direction.

[0083] Second rotation matrix ,in, This is the second rotation matrix; This represents the pitch angle. The vertical pitch attitude of the carrier can be adjusted using the second rotation matrix.

[0084] Third rotation matrix ,in, This is the third rotation matrix; This refers to the roll angle. The third rotation matrix can be used to adjust the left and right roll attitude of the carrier.

[0085] Total rotation matrix After expansion, the total rotation matrix is ​​as follows:

[0086] ;in, This is the total rotation matrix.

[0087] It should be noted that obtaining the coordinate increment in the northeast celestial coordinate system based on the total rotation matrix and the carrier coordinate increment includes: calculating... To obtain the coordinate increments in the northeast-central coordinate system, the following steps are performed. This represents the coordinate increment in the northeast-northeast coordinate system. This represents the increment of the E-axis in the northeast-northeast coordinate system; This represents the increment of the N-axis in the northeast-northeast coordinate system; This represents the increment of the U-axis in the northeast-northeast coordinate system; The increment is the horizontal axis. The increment is the vertical axis. This represents the vertical axis increment.

[0088] In some embodiments, the formula After expansion, the following formula is obtained:

[0089] This formula can be used to directly obtain the coordinate increment in the northeast-northeast coordinate system.

[0090] Furthermore, the second initial UAV spatial coordinates are obtained based on the coordinate increments in the Northeast Celestial Coordinate System and the initial equipment coordinates, including: obtaining the second longitude of the second initial UAV spatial coordinates by calculating L = lon1 + ΔE / (r + alt1) / cos(lat1) × (180 / π); where L is the second longitude of the second initial UAV spatial coordinates in degrees; obtaining the second latitude of the second initial UAV spatial coordinates by calculating B = lat1 + ΔN / (r + alt1) × (180 / π); where B is the second latitude of the second initial UAV spatial coordinates in degrees; and obtaining the second altitude of the second initial UAV spatial coordinates by calculating H = alt1 + ΔU; where H is the second altitude of the second initial UAV spatial coordinates.

[0091] The initial spatial coordinates of the UAV include the first longitude, first latitude, and first altitude of the target UAV, collected by radar detection equipment.

[0092] The third initial UAV spatial coordinates include the target UAV's third longitude, third latitude, and third altitude, collected by radio detection equipment.

[0093] The initial fusion spatial coordinates include: the initial fusion longitude, initial fusion latitude, and initial fusion altitude of the target UAV.

[0094] Furthermore, by fusing the first initial UAV spatial coordinates, the third initial UAV spatial coordinates, and the second initial UAV spatial coordinates using optoelectronic fusion weights, radio fusion weights, and radar fusion weights, initial fused spatial coordinates are obtained. This includes: calculating the first longitude product with the radar fusion weights; calculating the second longitude product with the optoelectronic fusion weights; calculating the third longitude product with the radio fusion weights; determining the initial fused longitude by summing the first longitude product, the third longitude product, and the second longitude product; calculating the first latitude product with the radar fusion weights; calculating the second latitude product with the optoelectronic fusion weights; calculating the third latitude product with the radio fusion weights; determining the initial fused latitude by summing the first latitude product, the third latitude product, and the second latitude product; calculating the first altitude product with the radar fusion weights; calculating the second altitude product with the optoelectronic fusion weights; calculating the third altitude product with the radio fusion weights; determining the initial fused altitude by summing the first altitude product, the third altitude product, and the second altitude product.

[0095] It should be noted that the method of using an agent to fuse processed detection data according to device fusion weights to obtain processed fused data is the same as the method of using an agent to fuse initial detection data according to preset device fusion weights to obtain initial fused data, and will not be repeated here.

[0096] Specifically, the data to be processed and merged includes the spatial coordinates of the processed data, the results of the integrity and compliance verification of the identity information, the size of the processed data, the payload data of the processed data, the flight speed of the processed data, and the lock status of the processed data.

[0097] The preset threat weights include: intrusion threat weight, risk threat weight, and mobile threat weight.

[0098] Furthermore, the intelligent agent is used to obtain the initial threat level of the target drone based on preset threat weights and initial detection data. This includes: weighting the identity information integrity and compliance verification results to obtain the target drone's flight compliance; obtaining the initial flight area deviation based on the initial fused spatial coordinates; obtaining the initial sensitive distance based on the initial fused spatial coordinates; weighting the flight compliance, initial flight area deviation, and initial sensitive distance to obtain the initial intrusion threat; weighting the initial payload data and initial size data to obtain the initial risk threat; obtaining the initial heading change rate based on the initial fused spatial coordinates; and weighting the initial flight speed, initial heading change rate, and initial lock-on status to obtain the initial maneuver threat. The intelligent agent calculates... This gives you an initial threat level. Initial threat level; Weighting of intrusion threats; This is the initial intrusion threat; Risk threat weighting; For initial risk threats; Weighting of mobile threats; This is an initial maneuvering threat.

[0099] It should be noted that the initial flight area deviation is obtained through multiple initial fused spatial coordinates. Specifically, the flight trend of the target UAV can be determined through multiple initial fused spatial coordinates, thereby determining the normalized value of the deviation angle of the target UAV toward the preset sensitive area or preset protected area.

[0100] The protected area is a no-fly zone, that is, an area where the target drone cannot fly; for example, military areas.

[0101] Sensitive areas are restricted flight zones or temporary control zones, which are areas where flying is permitted but restrictions must be strictly observed; for example: height-restricted areas around airports, urban centers, areas around high-rise buildings, and areas within 100 meters of high-voltage lines / communication base stations.

[0102] The initial sensitive distance is a normalized value of the distance between the latest acquired initial fused spatial coordinates and the boundary of the preset sensitive area. The greater the distance from the boundary of the preset sensitive area, the greater the corresponding threat, and the greater the initial sensitive distance.

[0103] The initial rate of change of heading is obtained through multiple initial fused spatial coordinates. Specifically, the flight direction of the target UAV can be determined using multiple initial fused spatial coordinates. The angular velocity of the change in the target UAV's flight direction is then defined as the initial rate of change of heading.

[0104] Furthermore, the intelligent agent is used to obtain the corresponding threat level of the target drone after handling based on threat weights and handling detection data. This includes: using the intelligent agent to weight the integrity and compliance verification results of identity information to obtain the flight compliance of the target drone; using the intelligent agent to obtain the deviation of the handling flight area based on the handling fused spatial coordinates; using the intelligent agent to obtain the handling sensitive distance based on the handling fused spatial coordinates; using the intelligent agent to weight the flight compliance, the deviation of the handling flight area, and the handling sensitive distance to obtain the handling intrusion threat; using the intelligent agent to weight the handling payload data and the handling size data to obtain the handling risk threat; using the intelligent agent to obtain the handling heading change rate of the target drone based on the handling fused spatial coordinates; and using the intelligent agent to weight the handling flight speed, the handling heading change rate, and the handling lock-on status to obtain the handling maneuver threat. The intelligent agent calculates... To obtain the threat level for handling. Among them, To address the level of threat; Weighting of intrusion threats; To deal with intrusion threats; Risk threat weighting; To address risks and threats; Weighting of mobile threats; To deal with mobile threats.

[0105] It should be noted that the threat weights are obtained by training an agent's response effectiveness evaluation model. In some embodiments, the intrusion threat weight is 0.4; the risk threat weight is 0.3; and the mobile threat weight is 0.3.

[0106] The preset action space represents the space for evaluating the handling results, specifically including whether the handling was successful or unsuccessful.

[0107] Furthermore, the intelligent agent determines the target action within a preset action space based on the initial threat level and the response threat level, including: through calculation The system obtains the threat reduction rate; if the threat reduction rate is greater than or equal to a preset first threshold, the successful handling is identified as the target action; if the threat reduction rate is less than the preset first threshold, the failed handling is identified as the target action.

[0108] It should be noted that if the threat reduction rate is greater than 0 and less than the preset first threshold, it means that the target drone has been interfered with, but the degree of interference is low and the threat reduction rate of the target drone is small, so it needs to be dealt with again. Therefore, it is still set as a failure to handle the situation, so as to facilitate the unified handling in the future.

[0109] The method for evaluating the handling efficiency of unmanned aerial vehicles (UAVs) provided in this disclosure involves using several detection sensors to detect the UAV, acquiring initial detection data corresponding to the target UAV detected by each sensor, then handling the target UAV, and acquiring the handling detection data corresponding to the target UAV detected by each sensor after handling. The initial detection data and the handling detection data are then input into a preset intelligent agent handling efficiency evaluation model to assess the handling of the target UAV using intelligent agent evaluation, thereby obtaining an evaluation result. Compared to the prior art's method of obtaining evaluation results through manual judgment, this solution achieves automatic evaluation of UAV handling efficiency by inputting the detection data from each sensor before and after handling the target UAV into a preset intelligent agent handling efficiency evaluation model, thereby using intelligent agent evaluation to assess the handling of the target UAV and obtain an evaluation result.

[0110] Furthermore, after obtaining the assessment results, the process also includes: if the assessment result indicates that the handling has failed, re-handling the target drone. The handling method can be the same as the previous handling method, or other handling methods can be used; there are no restrictions here.

[0111] Please see Figure 2 , Figure 2 A flowchart for obtaining an evaluation model of the intelligent agent's processing efficiency.

[0112] like Figure 2 As shown, the intelligent agent's processing efficiency evaluation model is obtained in the following way:

[0113] Step S201: Use several types of detection sensors to detect the preset reference UAV;

[0114] Step S202: Obtain the initial training detection data corresponding to the reference UAV detected by each detection sensor;

[0115] Step S203: Take action on the reference drone and obtain the action reference result;

[0116] Step S204: Obtain the processing training detection data corresponding to the reference UAV detected by each detection sensor after processing;

[0117] Step S205: Input the initial training probe data and the processing training probe data into the preset initial agent processing performance evaluation model, so as to use the agent to train the initial agent processing performance evaluation model and obtain the agent processing performance evaluation model.

[0118] In this embodiment, several detection sensors are used to detect a preset reference drone, acquiring initial training detection data corresponding to the reference drone detected by each sensor. Then, the reference drone is processed, and a processing reference result is obtained. Next, processing training detection data corresponding to the reference drone detected by each sensor after processing is acquired. The initial training detection data and processing training detection data are then input into a preset initial agent processing efficiency evaluation model. This model is then used to train the agent to obtain the agent processing efficiency evaluation model. This allows the trained agent processing efficiency evaluation model to accurately and automatically evaluate the target drone based on detection data before and after processing, improving the automation and accuracy of drone processing efficiency evaluation.

[0119] It should be noted that the initial training detection data is the detection data continuously acquired by each detection sensor before the reference drone was dealt with.

[0120] The initial training detection data includes an initial training UAV state dataset and an initial training detection device state set. The initial training UAV state dataset includes the initial training size data, initial training payload data, spatial coordinates of the first initial training UAV, spatial coordinates of the third initial training UAV, initial training flight speed, and initial training flight compliance parameters corresponding to the target UAV. The initial training size data and initial training payload data are acquired through photoelectric detection equipment. The spatial coordinates of the first initial training UAV and the initial training flight speed are acquired through radar detection equipment. The initial training detection device state set includes the initial training state data corresponding to each detection sensor. The spatial coordinates of the third initial training UAV are acquired through radio detection equipment.

[0121] The initial training state data corresponding to each detection sensor includes: the initial training state data corresponding to the radar detection equipment, the initial training state data corresponding to the photoelectric detection equipment, the initial training state data corresponding to the radio detection equipment, and the initial training state data corresponding to the RID ground receiving equipment.

[0122] The initial training state data corresponding to the radar detection equipment includes: radar detection confidence and radar signal-to-noise ratio.

[0123] The initial training status data corresponding to the photoelectric detection equipment includes: the first initial training turntable data, the initial training equipment coordinates, the initial training installation posture, the initial training sharpness score, the initial training photoelectric turntable positioning accuracy, and the photoelectric signal-to-noise ratio.

[0124] The initial training status data corresponding to the radio detection equipment includes: second initial training turntable data, demodulation success rate, and radio signal-to-noise ratio.

[0125] The initial training status data corresponding to the RID ground receiving equipment includes: identity information integrity, compliance verification results, decoding success rate, and RID signal-to-noise ratio.

[0126] It should be noted that the method for dealing with the reference drone is the same as the method for dealing with the target drone, and will not be repeated here.

[0127] The handling reference result refers to the actual handling result of the reference drone. In some embodiments, if the reference drone makes a forced landing, returns to home, or leaves the preset sensitive area, the handling is determined to be successful; otherwise, the handling is confirmed to be unsuccessful.

[0128] It should be noted that the detection data for the disposal training refers to the detection data continuously acquired by each detection sensor after the disposal of the reference drone.

[0129] The disposal training detection data includes a disposal training UAV status dataset and a disposal training detection device status set. The disposal training UAV status dataset includes the disposal training size data, disposal training payload data, spatial coordinates of the first disposal training UAV, spatial coordinates of the third disposal training UAV, disposal training flight speed, and disposal training flight compliance parameters corresponding to the target UAV. The disposal training size data and disposal training payload data are acquired through photoelectric detection equipment. The spatial coordinates and disposal training flight speed of the first disposal training UAV are acquired through radar detection equipment. The disposal training detection device status set includes the disposal training status data corresponding to each detection sensor. The spatial coordinates of the third disposal training UAV are acquired through radio detection equipment.

[0130] The corresponding response training data for each detection sensor includes: response training status data for radar detection equipment, response training status data for photoelectric detection equipment, response training status data for radio detection equipment, and response training status data for RID ground receiving equipment.

[0131] The training status data for handling radar detection equipment includes: radar detection confidence level and radar signal-to-noise ratio.

[0132] The corresponding data for the photoelectric detection equipment includes: data of the first treatment training turntable, coordinates of the treatment training equipment, installation posture of the treatment training, resolution score of the treatment training, positioning accuracy of the photoelectric turntable and the optical-to-electrical noise ratio.

[0133] The data on the response training status of the radio detection equipment includes: data from the second response training turntable, demodulation success rate, and radio signal-to-noise ratio.

[0134] The processing training status data corresponding to the RID ground receiving equipment includes: identity information integrity, compliance verification results, decoding success rate, and RID signal-to-noise ratio.

[0135] Furthermore, the state space of the initial agent's handling effectiveness evaluation model includes a pre-handling state vector and a post-handling state vector; the pre-handling state vector is obtained through initial training probe data; the post-handling state vector is obtained through handling training probe data; the action space of the agent's handling effectiveness evaluation model represents the space of handling evaluation results; the reward function of the agent's handling effectiveness evaluation model includes a threat change reward, an evaluation consistency reward, and a data reliability reward; wherein, the threat change reward is obtained through initial training threat and handling training threat; the initial training threat is obtained through preset training threat weights and initial training probe data; the handling training threat is obtained through training threat weights and handling training probe data; the evaluation consistency reward is obtained through handling reference results and training reference actions; the training reference actions are obtained by utilizing the agent in the action space based on the initial training threat and handling training threat; the data reliability reward is obtained through the initial training threat and / or handling training threat; the agent trains the initial agent's handling effectiveness evaluation model by maximizing the reward function to obtain the agent's handling effectiveness evaluation model. This allows the agent to train an initial agent response effectiveness evaluation model while maximizing rewards for threat level change, assessment consistency, and data reliability. Consequently, the agent can automatically evaluate the drone response effectiveness based on the trained model, ensuring high threat level change, high assessment accuracy, and high assessment reliability.

[0136] It should be noted that during training, the state vector before the action is the initial training probe data, and the state vector after the action is the action training probe data.

[0137] It should be noted that by using state space and action space, we can simulate the dynamic scenario of the entire drone handling process, ensuring the generalization of the trained model.

[0138] Furthermore, the reward function of the intelligent agent's performance evaluation model is: .in, The reward value of the reward function; This is the reward for changes in threat level, and it is the core term of the reward function; To evaluate consistency reward items; This is a reward item for data reliability. The threat reward weight is a preset value; in some embodiments, the threat reward weight is 0.6. Rewards are given for changes in threat level; The default consistency weight is 0.2 in some embodiments. To evaluate consistency rewards; The reliability weight is a preset value; in some embodiments, the reliability weight is 0.1. Rewards are given for data reliability.

[0139] It should be noted that the threat level change reward corresponds to the training threat level reduction rate. The method for obtaining the training threat level reduction rate through the initial training threat level and the handling training threat level can refer to the aforementioned method for obtaining the threat level reduction rate through the initial threat level and the handling threat level; the method for obtaining the initial training threat level through preset training threat weights and initial training probe data can refer to the aforementioned method for obtaining the initial threat level through threat weights and initial probe data; and the method for obtaining the handling training threat level through training threat weights and handling training probe data can refer to the aforementioned method for obtaining the handling threat level through threat weights and handling probe data, and will not be repeated here.

[0140] In some embodiments, if the training threat reduction rate is greater than or equal to a preset first threshold, a first preset value is determined as the threat change reward; if the training threat reduction rate is greater than 0 and less than the preset first threshold, a second preset value is determined as the threat change reward; if the training threat reduction rate is less than or equal to 0, a third preset value is determined as the threat change reward. The first preset value is greater than the second preset value; the second preset value is greater than the third preset value. For example, the first preset value can be 10; the second preset value can be 3; and the third preset value can be -10.

[0141] It should be noted that the consistency reward is obtained as follows: when the action reference result is the same as the training reference action, the fourth preset value is determined as the consistency reward; when the action reference result is different from the training reference action, the fifth preset value is determined as the consistency reward. In some embodiments, the fourth preset value is greater than the fifth preset value. For example, the fourth preset value is 2, and the fifth preset value is -3. In this way, by evaluating the consistency reward, the probability of the agent selecting the wrong target action can be reduced.

[0142] The method for determining the training reference movement can refer to the aforementioned method for determining the target movement, and will not be repeated here.

[0143] Furthermore, the data reliability reward is obtained as follows: Based on the initial training threat level and / or the handling training threat level, the detection confidence level and signal-to-noise ratio standardized values ​​corresponding to each detection sensor are obtained; through calculation... Obtain the reliability score corresponding to the i-th detection sensor; where, The reliability score is given for the i-th detection sensor. Let be the detection confidence level corresponding to the i-th detection sensor; Let i be the normalized signal-to-noise ratio value corresponding to the i-th detection sensor; calculate the sum of the reliability scores corresponding to each detection sensor; and calculate... To obtain the device credibility of the i-th detection sensor; where, Let represent the device reliability corresponding to the i-th detection sensor; The reliability scores corresponding to each detection sensor are summed; the mean reliability of the device corresponding to each detection sensor is calculated to obtain the mean reliability value; if the mean reliability value is greater than or equal to a preset reliability threshold, a sixth preset value is determined as the data reliability reward; if the mean reliability value is less than the preset reliability threshold, a seventh preset value is determined as the data reliability reward; the sixth preset value is greater than the seventh preset value. In some embodiments, the sixth preset value is 1; the seventh preset value is 0. This improves the reliability of the intelligent agent's processing efficiency evaluation model.

[0144] It should be noted that for radar detection equipment, the corresponding detection confidence level is the radar detection confidence level; and the corresponding signal-to-noise ratio (SNR) normalized value is the normalized value of the radar SNR.

[0145] For photoelectric detection equipment, the corresponding detection confidence level is the weighted sum of the initial training sharpness score and the initial training photoelectric turntable positioning accuracy; the corresponding signal-to-noise ratio standardized value is the photoelectric signal-to-noise ratio.

[0146] For radio detection equipment, the corresponding detection confidence level is the demodulation success rate; the corresponding signal-to-noise ratio normalized value is the radio signal-to-noise ratio.

[0147] For RID ground receiving equipment, the corresponding detection confidence level is the decoding success rate; the corresponding signal-to-noise ratio normalized value is the RID signal-to-noise ratio.

[0148] It should be noted that the agent trains an initial agent action performance evaluation model by maximizing the reward function to obtain the agent action performance evaluation model. That is, during training, the agent continuously adjusts various weights, such as threat weights and other weights, such as device fusion weights, to iteratively train and maximize the reward function. Training is considered complete when a preset number of training iterations is reached, or when the change in the reward function value is less than a preset reward threshold. The initial agent action performance evaluation model after training is then determined as the agent action performance evaluation model.

[0149] It should be noted that the intelligent agent is the decision-maker for evaluating the outcome of the action. It is responsible for calculating the threat level based on the state space data of each detection sensor, judging the success or failure of the action, and iteratively optimizing the initial intelligent agent action effectiveness evaluation model to obtain the intelligent agent action effectiveness evaluation model.

[0150] This solution provides an automated method for evaluating the effectiveness of low-altitude UAV disposal results. It improves the automatic process loop of "discovery-identification-location-disposal-evaluation", reduces the labor cost of the low-altitude UAV automatic disposal system, provides users with high-confidence evaluation results, and improves the disposal efficiency of the low-altitude UAV automatic disposal system so that more means can be immediately called upon to carry out disposal when the disposal result is judged to be a failure.

[0151] Combination Figure 3 As shown in the figure, this disclosure provides an apparatus 30 for evaluating the handling performance of unmanned aerial vehicles (UAVs). The apparatus includes: a detection module 31, an initial detection data acquisition module 32, a handling module 33, a handling detection data acquisition module 34, and an evaluation module 35.

[0152] Among them, the detection module 31 is configured to use several types of detection sensors to conduct UAV detection;

[0153] The initial detection data acquisition module 32 is configured to acquire the initial detection data corresponding to the target UAV detected by each detection sensor;

[0154] The disposal module 33 is configured to dispose of the target drone;

[0155] The disposal detection data acquisition module 34 is configured to acquire disposal detection data corresponding to the target UAV detected by each detection sensor after disposal.

[0156] The evaluation module 35 is configured to input the initial detection data and the disposal detection data into a preset intelligent agent disposal performance evaluation model, so as to evaluate the disposal of the target UAV using intelligent agent evaluation and obtain evaluation results.

[0157] The apparatus for evaluating the handling efficiency of unmanned aerial vehicles (UAVs) provided in this disclosure utilizes several detection sensors to detect UAVs, acquires initial detection data corresponding to the target UAV detected by each sensor, then handles the target UAV, and acquires the handling detection data corresponding to the target UAV detected by each sensor after handling. The initial detection data and the handling detection data are then input into a preset intelligent agent handling efficiency evaluation model to evaluate the handling of the target UAV using intelligent agent evaluation, thereby obtaining an evaluation result. In this way, compared to the prior art that obtains evaluation results through manual judgment, this solution achieves automatic evaluation of UAV handling efficiency by inputting the detection data of each sensor before and after handling the target UAV into a preset intelligent agent handling efficiency evaluation model to evaluate the handling of the target UAV using intelligent agent evaluation, thereby obtaining an evaluation result.

[0158] Furthermore, the detection sensors include: radar detection equipment, photoelectric detection equipment, radio detection equipment, and UAV remote identification ground receiving equipment.

[0159] Furthermore, the initial detection data includes an initial UAV state dataset and an initial detection device state set; wherein, the initial UAV state dataset includes the initial size data, initial payload data, first initial UAV spatial coordinates, third initial UAV spatial coordinates, initial flight speed, and initial flight compliance parameters corresponding to the target UAV; the initial size data and initial payload data are obtained through photoelectric detection equipment; the first initial UAV spatial coordinates and initial flight speed are obtained through radar detection equipment; the initial detection device state set includes the initial state data corresponding to each detection sensor; the third initial UAV spatial coordinates are obtained through radio detection equipment.

[0160] Furthermore, the evaluation module is configured to evaluate the handling of the target drone using an agent evaluation mechanism in the following ways to obtain evaluation results: The agent fuses initial detection data according to preset device fusion weights to obtain initial fused data; the device fusion weights are obtained by training an agent handling efficiency evaluation model; the agent fuses handling detection data according to the device fusion weights to obtain handling fused data; the agent obtains the initial threat level of the target drone based on preset threat weights and initial detection data; the agent obtains the handling threat level of the target drone after handling based on threat weights and handling detection data; the threat weights are obtained by training an agent handling efficiency evaluation model; the agent determines the target action in a preset action space based on the initial threat level and handling threat level; and the target action is determined as the evaluation result.

[0161] Furthermore, the initial state data corresponding to the photoelectric detection equipment includes: first initial turntable data, initial equipment coordinates, and initial installation attitude; the initial state data corresponding to the radio detection equipment includes: second initial turntable data; the equipment fusion weights include photoelectric fusion weights, radio fusion weights, and radar fusion weights; the initial fusion data includes the initial fusion spatial coordinates corresponding to the target UAV; the evaluation module is configured to obtain the initial fusion spatial coordinates in the following ways: obtain the initial coordinates corresponding to the target UAV; the initial coordinates include the initial guidance latitude and longitude and the initial altitude corresponding to the target UAV; obtain the initial radian latitude and longitude corresponding to the initial guidance latitude and longitude; obtain the initial distance between the photoelectric detection equipment and the initial coordinates based on the initial radian latitude and longitude and the initial equipment coordinates; fuse the first initial turntable data and the second initial turntable data using the photoelectric fusion weights and the radio fusion weights to obtain the initial fusion turntable data; obtain the second initial UAV spatial coordinates based on the initial distance, the initial fusion turntable data, and the initial installation attitude; fuse the first initial UAV spatial coordinates, the third initial UAV spatial coordinates, and the second initial UAV spatial coordinates using the photoelectric fusion weights, the radio fusion weights, and the radar fusion weights to obtain the initial fusion spatial coordinates.

[0162] Furthermore, the device for evaluating the handling efficiency of unmanned aerial vehicles (UAVs) also includes a model acquisition module. The model acquisition is configured to acquire an agent handling efficiency evaluation model by: detecting a preset reference UAV using several types of detection sensors; acquiring initial training detection data corresponding to the reference UAV detected by each detection sensor; handling the reference UAV and acquiring the handling reference result; acquiring handling training detection data corresponding to the reference UAV detected by each detection sensor after handling; inputting the initial training detection data and the handling training detection data into a preset initial agent handling efficiency evaluation model to train the initial agent handling efficiency evaluation model using an agent, thereby obtaining an agent handling efficiency evaluation model.

[0163] Furthermore, the state space of the initial agent's handling effectiveness evaluation model includes a pre-handling state vector and a post-handling state vector; the pre-handling state vector is obtained through initial training probe data; the post-handling state vector is obtained through handling training probe data; the action space of the agent's handling effectiveness evaluation model represents the space of handling evaluation results; the reward function of the agent's handling effectiveness evaluation model includes a threat change reward, an evaluation consistency reward, and a data reliability reward; wherein, the threat change reward is obtained through initial training threat and handling training threat; the initial training threat is obtained through preset training threat weights and initial training probe data; the handling training threat is obtained through training threat weights and handling training probe data; the evaluation consistency reward is obtained through handling reference results and training reference actions; the training reference actions are obtained by utilizing the agent in the action space based on the initial training threat and handling training threat; the data reliability reward is obtained through the initial training threat and / or handling training threat; the agent trains the initial agent's handling effectiveness evaluation model by maximizing the reward function to obtain the agent's handling effectiveness evaluation model.

[0164] It should be noted that the apparatus for evaluating the handling efficiency of unmanned aerial vehicles (UAVs) provided in the above embodiments and the method for obtaining carbon emission inversion models provided in the above embodiments belong to the same concept. The specific operation methods of each module and unit have been described in detail in the method embodiments and will not be repeated here. In practical applications, the apparatus for evaluating the handling efficiency of UAVs provided in the above embodiments can be assigned to different functional modules as needed, that is, the internal structure of the apparatus can be divided into different functional modules to complete all or part of the functions described above. This is not a limitation here.

[0165] Combination Figure 4 As shown, this disclosure provides an electronic device including a processor 41 and a memory 42. Optionally, the device may further include a communication interface 43 and a bus 44. The processor 41, communication interface 43, and memory 42 can communicate with each other via the bus 44. The communication interface 43 can be used for information transmission. The processor 41 can call logical instructions in the memory 42 to execute the method for evaluating the handling performance of a drone as described in the above embodiment.

[0166] Furthermore, the logical instructions in the aforementioned memory 42 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium.

[0167] The memory 42, as a storage medium, can be used to store software programs and computer-executable programs, such as program instructions / modules corresponding to the methods in the embodiments of this disclosure. The processor 41 executes functional applications and data processing by running the program instructions / modules stored in the memory 42, that is, it implements the method for evaluating the handling effectiveness of the UAV in the above embodiments.

[0168] The memory 42 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the terminal device. Furthermore, the memory 42 may include high-speed random access memory and may also include non-volatile memory.

[0169] The electronic device provided in this disclosure uses several detection sensors to detect drones, acquires initial detection data corresponding to the target drone detected by each sensor, then handles the target drone, and acquires the handling detection data corresponding to the target drone detected by each sensor after handling. The initial detection data and the handling detection data are then input into a preset intelligent agent handling efficiency evaluation model to evaluate the handling of the target drone using intelligent agent evaluation, thus obtaining an evaluation result. In this way, compared to the prior art that obtains evaluation results through manual judgment, this solution achieves automatic evaluation of drone handling efficiency by inputting the detection data of each sensor before and after handling the target drone into a preset intelligent agent handling efficiency evaluation model to evaluate the handling of the target drone using intelligent agent evaluation, thereby obtaining an evaluation result.

[0170] This disclosure provides a storage medium storing computer-executable instructions configured to perform the method described above for evaluating the handling effectiveness of a drone.

[0171] The aforementioned storage media can be either transient computer-readable storage media or non-transitory computer-readable storage media. Non-transitory storage media include various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks, and can also be transient storage media.

[0172] The foregoing description and accompanying drawings fully illustrate embodiments of this disclosure to enable those skilled in the art to practice them. Other embodiments may include structural, logical, electrical, procedural, and other changes. The embodiments represent only possible variations. Individual components and functions are optional unless explicitly required, and the order of operation may vary. Parts and features of some embodiments may be included in or replace parts and features of other embodiments. Moreover, the terminology used in this application is for describing embodiments only and is not intended to limit the claims. As used in the description of embodiments and claims, the singular forms “a,” “an,” and “the” are intended to equally include the plural forms unless the context clearly indicates otherwise. Similarly, the term “and / or” as used in this application means including one or more of the associated listed items and all possible combinations thereof. Additionally, when used in this application, the term "comprise" and its variations "comprises" and / or "comprising" refer to the presence of stated features, integrals, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or groups thereof. Without further limitations, an element defined by the phrase "comprises a..." does not exclude the presence of other identical elements in the process, method, or apparatus that includes said element. In this document, each embodiment may focus on the differences from other embodiments, and similar or identical parts between embodiments can be referred to mutually. For methods, products, etc., disclosed in the embodiments, if they correspond to the method section disclosed in the embodiments, the relevant parts can be referred to the description of the method section.

[0173] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the embodiments of this disclosure. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0174] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than that shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. In the descriptions corresponding to the flowcharts and block diagrams in the accompanying drawings, the operations or steps corresponding to different blocks may also occur in a different order than disclosed in the description, and sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. Each block in a block diagram and / or flowchart, and combinations of blocks in a block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

Claims

1. A method for evaluating the operational effectiveness of unmanned aerial vehicles (UAVs), characterized in that, include: Using several types of detection sensors for drone detection; Acquire the initial detection data corresponding to the target UAV detected by each of the aforementioned detection sensors; The target drone was dealt with. After processing, acquire the processing detection data corresponding to the target UAV detected by each of the aforementioned detection sensors; The initial detection data and the disposal detection data are input into a preset intelligent agent disposal efficiency evaluation model to evaluate the disposal of the target UAV using intelligent agent evaluation and obtain evaluation results.

2. The method according to claim 1, characterized in that, The detection sensors include: radar detection equipment, photoelectric detection equipment, radio detection equipment, and UAV remote identification ground receiving equipment.

3. The method according to claim 2, characterized in that, The initial detection data includes an initial UAV status dataset and an initial detection device status dataset; The initial UAV state dataset includes the initial size data, initial payload data, first initial UAV spatial coordinates, third initial UAV spatial coordinates, initial flight speed, and initial flight compliance parameters corresponding to the target UAV; the initial size data and the initial payload data are obtained through the photoelectric detection device; the first initial UAV spatial coordinates and the initial flight speed are obtained through the radar detection device; the initial detection device state set includes the initial state data corresponding to each of the detection sensors; and the third initial UAV spatial coordinates are obtained through the radio detection device.

4. The method according to claim 3, characterized in that, The assessment of the handling of the target drone using intelligent agent evaluation, and the resulting assessment, includes: The agent fuses the initial detection data according to preset device fusion weights to obtain initial fused data; the device fusion weights are obtained by training the agent's processing efficiency evaluation model; the agent fuses the processing detection data according to the device fusion weights to obtain processing fused data. The agent obtains the initial threat level of the target drone based on a preset threat weight and the initial detection data; the agent also obtains the response threat level of the target drone after response based on the threat weight and the response detection data; the threat weight is obtained by training the agent's response effectiveness evaluation model. The intelligent agent determines the target action within a preset action space based on the initial threat level and the handling threat level. The target action is determined as the evaluation result.

5. The method according to claim 4, characterized in that, The initial state data corresponding to the photoelectric detection device includes: first initial turntable data, initial device coordinates, and initial installation attitude; the initial state data corresponding to the radio detection device includes: second initial turntable data; the device fusion weights include photoelectric fusion weights, radio fusion weights, and radar fusion weights; the initial fusion data includes the initial fusion spatial coordinates corresponding to the target UAV; the initial fusion spatial coordinates are obtained through the following method: Obtain the initial coordinates corresponding to the target UAV; the initial coordinates include the initial guiding latitude and longitude and the initial altitude of the target UAV; Obtain the initial radian latitude and longitude corresponding to the initial guiding latitude and longitude; The initial distance between the photoelectric detection device and the initial coordinates is obtained based on the initial radian latitude and longitude and the initial device coordinates; The first initial turntable data and the second initial turntable data are fused using the optoelectronic fusion weight and the radio fusion weight to obtain initial fused turntable data; The second initial UAV spatial coordinates are obtained based on the initial distance, the initial fusion turntable data, and the initial installation attitude. The initial fused spatial coordinates are obtained by fusing the first initial UAV spatial coordinates, the third initial UAV spatial coordinates, and the second initial UAV spatial coordinates using the optoelectronic fusion weight, the radio fusion weight, and the radar fusion weight.

6. The method according to any one of claims 1 to 5, characterized in that, The intelligent agent's processing efficiency evaluation model is obtained through the following method: Several types of detection sensors are used to detect a pre-set reference UAV; Acquire the initial training detection data corresponding to the reference UAV detected by each of the aforementioned detection sensors; The reference drone is processed, and the processing reference result is obtained; After processing, acquire the processing training detection data corresponding to the reference UAV detected by each of the aforementioned detection sensors; The initial training probe data and the disposal training probe data are input into a preset initial agent disposal performance evaluation model, so as to use the agent to train the initial agent disposal performance evaluation model and obtain the agent disposal performance evaluation model.

7. The method according to claim 6, characterized in that, The state space of the initial agent handling efficiency evaluation model includes a pre-handling state vector and a post-handling state vector; the pre-handling state vector is obtained through the initial training probe data; the post-handling state vector is obtained through the handling training probe data. The action space of the intelligent agent's processing efficiency evaluation model represents the space of the processing evaluation results. The reward function of the agent's handling effectiveness evaluation model includes a threat level change reward, an evaluation consistency reward, and a data reliability reward. Specifically, the threat level change reward is obtained through the initial training threat level and the handling training threat level; the initial training threat level is obtained through preset training threat weights and the initial training probe data; the handling training threat level is obtained through the training threat weights and the handling training probe data; the evaluation consistency reward is obtained through the handling reference result and the training reference action; the training reference action is obtained by the agent in the action space based on the initial training threat level and the handling training threat level; and the data reliability reward is obtained through the initial training threat level and / or the handling training threat level. The agent trains the initial agent performance evaluation model by maximizing the reward function to obtain the agent performance evaluation model.

8. An apparatus for evaluating the performance of unmanned aerial vehicles (UAVs), characterized in that, include: The detection module is configured to use several types of detection sensors for drone detection; The initial detection data acquisition module is configured to acquire the initial detection data corresponding to the target UAV detected by each of the detection sensors; The disposal module is configured to dispose of the target drone; The disposal detection data acquisition module is configured to acquire disposal detection data corresponding to the target UAV detected by each of the detection sensors after disposal. The evaluation module is configured to input the initial detection data and the disposal detection data into a preset intelligent agent disposal efficiency evaluation model, so as to evaluate the disposal of the target UAV using intelligent agent evaluation and obtain evaluation results.

9. An electronic device comprising a processor and a memory storing program instructions, characterized in that, The processor is configured to, when running the program instructions, execute the method for evaluating the handling effectiveness of a drone as described in any one of claims 1 to 7.

10. A storage medium storing program instructions, characterized in that, When the program instructions are executed, they perform the method for evaluating the handling effectiveness of a drone as described in any one of claims 1 to 7.

Images