Positioning result determination method and device, equipment, medium and product

Abstract

The invention discloses a positioning result determination method and device, equipment, a medium and a product. The method comprises the following steps: determining position information of a target unmanned aerial vehicle based on at least two positioning modes; based on the environmental parameters of the current surrounding environment of the target unmanned aerial vehicle, determining a first confidence coefficient corresponding to each positioning mode; determining index attributes of a plurality of perception quality evaluation indexes associated with the target unmanned aerial vehicle in the positioning mode, and determining a second confidence coefficient corresponding to the positioning mode based on the plurality of index attributes; and determining a target positioning result corresponding to the target unmanned aerial vehicle based on the first confidence and the second confidence corresponding to each positioning mode and all the position information. According to the technical scheme provided by the invention, the effect of improving the positioning accuracy and reliability of the unmanned aerial vehicle is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer processing, and in particular to a positioning result determination method, device, equipment, medium and product. BACKGROUND

[0002] With the development of unmanned aerial vehicle technology, low-altitude unmanned aerial vehicles are widely used in logistics, agriculture, security and emergency rescue and other fields. The application of unmanned aerial vehicles not only improves work efficiency, but also provides a new method and perspective for solving complex problems. In order to ensure the work efficiency and accuracy of the unmanned aerial vehicle, and to ensure the controllability of the unmanned aerial vehicle, the unmanned aerial vehicle usually needs to be positioned.

[0003] Current unmanned aerial vehicle detection technology mostly relies on a single type of sensor. However, these single detection methods are prone to high false alarm rates in bad weather conditions or complex environments. For example, visual sensors are extremely sensitive to light conditions and are easily lost in low light conditions such as night or fog; radar sensors will have reduced detection efficiency when encountering rain and fog. SUMMARY

[0004] The present application provides a positioning result determination method, device, equipment, medium and product to improve the accuracy and reliability of unmanned aerial vehicle positioning.

[0005] According to an aspect of the present application, a positioning result determination method is provided, which comprises:

[0006] determining the position information of a target unmanned aerial vehicle based on at least two positioning methods;

[0007] determining a first confidence corresponding to each positioning method based on the environmental parameters of the current surrounding environment of the target unmanned aerial vehicle;

[0008] For each positioning method, determine the index attributes of a plurality of perception quality evaluation indicators associated with the target unmanned aerial vehicle in the positioning method, and determine the second confidence corresponding to the positioning method based on a plurality of index attributes; wherein the plurality of perception quality evaluation indicators at least include data quality indicators, environmental impact indicators and historical positioning reliability indicators; the index attributes include positioning data quality attributes corresponding to the data quality indicators, environmental impact attributes corresponding to the environmental impact indicators, and positioning reliability attributes corresponding to the historical positioning reliability indicators;

[0009] determining a target positioning result corresponding to the target unmanned aerial vehicle based on the first confidence and the second confidence corresponding to each positioning method, and all the position information.

[0010] According to another aspect of the present application, there is provided a positioning result determination apparatus, comprising:

[0011] a position information determination module configured to determine position information of a target UAV based on at least two positioning modes;

[0012] a first confidence determination module configured to determine a first confidence corresponding to each of the positioning modes based on environmental parameters of a current surrounding environment of the target UAV;

[0013] a second confidence determination module configured to determine, for each of the positioning modes, index attributes of a plurality of perception quality evaluation indexes associated with the target UAV under the positioning mode, and determine a second confidence corresponding to the positioning mode based on the index attributes; wherein the plurality of perception quality evaluation indexes at least include a data quality index, an environmental influence index and a historical positioning reliability index; and the index attributes include a positioning data quality attribute corresponding to the data quality index, an environmental influence attribute corresponding to the environmental influence index and a positioning reliability attribute corresponding to the historical positioning reliability index;

[0014] a target positioning result determination module configured to determine a target positioning result corresponding to the target UAV based on the first confidence and the second confidence corresponding to each of the positioning modes, and all of the position information.

[0015] According to another aspect of the present application, there is provided an electronic device, comprising:

[0016] at least one processor; and a memory connected to the at least one processor in communication; wherein

[0017] the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to perform the positioning result determination method according to any one of the embodiments of the present application.

[0018] According to another aspect of the present application, there is provided a computer readable storage medium storing computer instructions for enabling a processor to perform the positioning result determination method according to any one of the embodiments of the present application when executed by the processor.

[0019] According to another aspect of the present application, there is provided a computer program product comprising a computer program for enabling a processor to perform the positioning result determination method according to any one of the embodiments of the present application when executed by the processor.

[0020] The technical scheme of the embodiment of the present application determines the position information of the target unmanned aerial vehicle based on at least two positioning modes; obtains the environment parameters of the current surrounding environment of the target unmanned aerial vehicle, determines the first confidence degree corresponding to each positioning mode based on the environment parameters; for each positioning mode, determines the index attributes of a plurality of perception quality evaluation indexes associated with the target unmanned aerial vehicle under the positioning mode, and determines the second confidence degree corresponding to the positioning mode based on the plurality of index attributes; wherein the plurality of perception quality evaluation indexes include a data quality index, an environment influence index and a historical positioning reliability index; based on the first confidence degree and the second confidence degree corresponding to each positioning mode and all the position information, a target positioning result corresponding to the target unmanned aerial vehicle is determined, which solves the problem of poor unmanned aerial vehicle positioning accuracy caused by the detection mode based on a single sensor in the prior art, realizes the acquisition of the position information of the target unmanned aerial vehicle by fusing at least two positioning modes, and dynamically evaluates the first confidence degree of each positioning mode under the current scene in combination with the environment parameters; at the same time, a plurality of perception quality evaluation indexes (including a data quality index, an environment influence index and a historical positioning reliability index) associated with the target unmanned aerial vehicle are further introduced, the second confidence degree of the positioning mode is determined by analyzing a plurality of index attributes under the positioning mode; finally, the first confidence degree and the second confidence degree of a plurality of positioning modes and the position information provided by each positioning mode are comprehensively weighted and fused to generate a target positioning result with high precision and high robustness, thereby improving the accuracy and reliability of unmanned aerial vehicle positioning in a complex and variable environment.

[0021] It should be understood that the content described in this part is not intended to identify key or important features of the embodiments of the present application, nor is it used to limit the scope of the present application. Other features of the present application will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0023] Figure 1 is a flowchart of a positioning result determination method according to an embodiment of the present application;

[0024] Figure 2 is a flowchart of a positioning result determination method according to an embodiment of the present application;

[0025] Figure 3 is a flowchart of a positioning result determination method according to an embodiment of the present application;

[0026] Figure 4 is a flow chart of a positioning result determination method according to an embodiment of the present application;

[0027] Figure 5 is a structural schematic diagram of a positioning result determination apparatus according to an embodiment of the present application;

[0028] Figure 6 is a structural schematic diagram of an electronic device implementing a positioning result determination method according to an embodiment of the present application. DETAILED DESCRIPTION

[0029] In order to make the personnel in the art better understand the present application scheme, the technical scheme in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should belong to the scope of protection of the present application.

[0030] It should be noted that the terms "first", "second" and the like in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or a chronological sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device that includes a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0031] It should be noted that in the technical scheme of the present disclosure, the collection, collection, updating, analysis, processing, use, transmission, storage and the like of user personal information are in accordance with the relevant legal regulations, are used for legal purposes, and do not violate public order and good customs. Necessary measures are taken to prevent illegal access to user personal information data, to maintain the security of user personal information and network security. It should also be noted that in the technical scheme of the present disclosure, the collection, collection, updating, analysis, processing, use, transmission, storage and the like of user personal information are carried out with the knowledge and permission of the user, in compliance with relevant privacy protection regulations.

[0032] Before introducing the technical solution, the application scenario can be introduced first. The technical solution provided by the application can be applied in any scene where the unmanned aerial vehicle needs to be positioned. For example, in the urban low-altitude logistics distribution scene, the unmanned aerial vehicle needs to accurately navigate to the designated receiving point in the complex environment with high-rise buildings and GNSS signal easily blocked. At this time, the technical solution provided in this embodiment can be used to fuse multi-source sensor information to realize real-time positioning with high robustness. Or, in the emergency rescue task, such as in the disaster area after an earthquake or landslide, the ground infrastructure is damaged, and the communication and positioning signal is limited. The position of the unmanned aerial vehicle needs to be accurately obtained to perform search and material delivery tasks. At this time, the technical solution provided in this embodiment can be used to complete reliable positioning under the condition of no GNSS or weak signal. Or, in the indoor inspection or warehouse management scene, the unmanned aerial vehicle needs to fly autonomously in a closed space without satellite signal and accurately position the shelf or equipment position. At this time, the technical solution provided in this embodiment can also be used to realize indoor positioning with centimeter-level accuracy by combining environmental features and inertial measurement data. Or, in the airport clearance protection area, when an unauthorized "black flight" unmanned aerial vehicle appears, it may seriously interfere with the safety of flight takeoff and landing. At this time, the technical solution provided in this embodiment can be used to fuse radar, radio frequency detection, and visual positioning information, and combine real-time weather, electromagnetic environment, and sensor data quality parameters to dynamically evaluate the credibility of each positioning source, thereby accurately determining the position of the invading unmanned aerial vehicle. Or, when performing security tasks for major activities in urban dense areas, complex building groups can cause multipath effects and signal shielding. A single sensor is difficult to stably track a low-altitude slow target. At this time, the technical solution provided in this embodiment can rely on adaptive fusion of multi-source sensing data to improve the positioning robustness in strong interference, weak light, or rainy weather.

[0033] Figure 1 is a flowchart of a positioning result determination method according to an embodiment of the application. The embodiment can be applied to any situation where the unmanned aerial vehicle needs to be positioned. The method can be executed by a positioning result determination device, which can be realized in the form of hardware and / or software, and can be configured in a computing device. As shown in Figure 1 , the method comprises:

[0034] S110, determining the position information of the target unmanned aerial vehicle based on at least two positioning modes.

[0035] The target unmanned aerial vehicle refers to an unmanned aerial vehicle whose spatial position needs to be determined. The positioning mode refers to a technical mode for obtaining position information of the unmanned aerial vehicle. For example, the positioning mode can be different positioning modes such as global navigation satellite system (GNSS) positioning, visual positioning, inertial navigation system (INS), ultra-wideband (UWB) positioning, Wi-Fi / Bluetooth beacon positioning, and laser radar point cloud matching. The position information refers to three-dimensional position coordinates of the target unmanned aerial vehicle in a preset coordinate system (such as a world geographical coordinate system or a local Cartesian coordinate system), which can include longitude, latitude and height, or x, y and z components.

[0036] It should be noted that the position information of one target unmanned aerial vehicle can be determined based on different positioning modes.

[0037] In the case where the positioning mode includes a visual recognition-based positioning mode, the specific implementation mode for determining the position information of the target unmanned aerial vehicle can be: acquiring an image sequence of the environment around the target unmanned aerial vehicle in real time through a camera device, comparing the current image with a pre-constructed environment visual map by using a feature extraction and matching algorithm (such as ORB, SIFT or deep learning feature), and calculating the three-dimensional pose of the target unmanned aerial vehicle in the map coordinate system by combining a PnP (Perspective-n-Point) or bundle adjustment (Bundle Adjustment) geometric optimization method, so as to obtain the position information; or in the case where there is no prior map, a visual-inertial odometer (VIO) technology is used, image data collected by the camera device and inertial measurement unit (IMU) data reported by the target unmanned aerial vehicle are fused, the relative pose of the target unmanned aerial vehicle relative to the initial position is estimated in real time by tracking the motion of feature points between consecutive image frames and combining the acceleration and angular velocity integrals in the IMU data, and the position information is output in the local coordinate system; or a deep learning model is used to regress the global position coordinates of the target unmanned aerial vehicle from the input image, the deep learning model has learned a large number of image-position pairs with labels in the training stage, and the position information of the target unmanned aerial vehicle in the input image can be output in the inference stage.

[0038] For example, the camera device can be a visible light camera. The target unmanned aerial vehicle can be detected by the visible light camera, the target distance between the visible light camera and the target unmanned aerial vehicle can be calculated by monocular or binocular vision geometry, the three-dimensional coordinates can be calculated by combining the pixel coordinates , the focal length f and the attitude angle of the target unmanned aerial vehicle, and the position information of the target unmanned aerial vehicle is obtained. , , .

[0039] The position information of the target unmanned aerial vehicle can be determined based on formula (1) .

[0040] Equation (1) is expressed as: ; wherein, represents the x-coordinate component in the position information . represents the coordinate component in the position information . represents the coordinate component in the position information . represents the target distance between the visible light camera and the target UAV; represents the focal length of the visible light camera; represents the attitude angle of the target UAV. The pixel coordinate is the two-dimensional projection position of the target UAV on the image plane after the visible light camera collects the image, which represents the specific position of the target UAV in the camera image.

[0041] Through the pixel coordinate, the mathematical mapping relationship from the two-dimensional image plane to the three-dimensional physical space can be established in combination with the intrinsic parameters (such as the focal length ) and extrinsic parameters (such as the relative pose relationship with the target UAV) of the camera.

[0042] Secondly, in equation (1), the pixel coordinate is used to calculate the x and y coordinate components (i.e. and ) of the target UAV in the three-dimensional space. Among them, ( ) and ( ) respectively represent the horizontal and vertical offset amounts of the target UAV in the image relative to the image center (principal point), and these offset amounts together with the focal length f determine the spatial proportional relationship of the target UAV in the horizontal and vertical directions, thereby realizing the conversion from the two-dimensional pixel position to the three-dimensional space coordinate.

[0043] Finally, the pixel coordinate ( ) is used in combination with the target distance , the focal length , and the attitude angle to constitute a complete triangulation or perspective projection model, so that the position information of the target UAV can be calculated only through monocular or binocular vision information.

[0044] When the positioning method includes infrared imaging-based positioning, the specific implementation of determining the location information of the target UAV can be as follows: First, capture the thermal radiation distribution image of the environment surrounding the target UAV using an infrared camera. Second, use a feature extraction algorithm to identify landmarks or structures with unique thermal characteristics. Third, match these features with a pre-established infrared thermal image database to calculate the relative position of the target UAV with respect to these landmarks, thereby determining its location information. Alternatively, in dynamic environments, use a particle filtering-based method to process the infrared image sequence between consecutive frames. Estimate the target UAV's trajectory through prediction and update steps, and combine the tracking results of infrared feature points to correct the position estimate, achieving real-time positioning. Finally, fuse infrared imaging and inertial measurement unit (IMU) data, and use algorithms such as extended Kalman filter (EKF) or unscented Kalman filter (UKF) to perform optimal estimation of both data. By compensating for errors that may occur in the infrared sensor during rapid movement and the cumulative drift of the IMU, positioning accuracy is improved, especially maintaining stable position output in low-light or smoke environments.

[0045] For example, an infrared camera can be used to detect a target drone based on its thermal radiation characteristics, compensating for the failure of visible light cameras in conditions such as nighttime, fog, and smoke. The location information of the target drone can be obtained by collecting thermal radiation data and calculating it. ={ , , }

[0046] Given the target dimensions (e.g., rotor diameter L) and f, the focal length of the infrared camera, the distance between the infrared camera and the target drone can be calculated using the pixel width w of the image captured by the infrared camera. ;

[0047] in, ;

[0048] Combined with the target UAV's pitch angle and azimuth (Image center offset), calculate the target UAV's position information: the target UAV's position information can be determined based on formula (2). .

[0049] Formula (2) can be expressed as: .in, Represents location information The x-coordinate component in the equation; Represents location information In Coordinate components; Represents location information In Coordinate components.

[0050] In another implementation, the localization method can combine visual localization with LiDAR point cloud matching. In this approach, visible light cameras and LiDAR are used to acquire images of the target drone's surrounding environment and 3D point cloud data, respectively. Visual localization extracts feature points from the images and matches them with a known map to estimate the target drone's position. LiDAR scans the surrounding environment and registers its point cloud with a pre-built high-precision 3D map (using algorithms such as ICP or NDT) to obtain the target drone's position. The position information obtained from both methods is then fused to obtain an updated position. Because visual localization is sensitive to changes in lighting while LiDAR provides stable geometric perception, their complementary nature can improve the reliability of drone localization in complex scenarios.

[0051] By integrating at least two positioning methods, the advantages of different sensors or positioning sources can be fully utilized, making up for the limitations of a single method in a specific environment. This not only improves the accuracy and stability of positioning results, but also enhances robustness and adaptability under complex conditions such as signal blockage, dynamic interference, or lack of infrastructure, thereby improving the accuracy of UAV positioning.

[0052] S120. Based on the environmental parameters of the target UAV's current surrounding environment, determine the first confidence level for each positioning method.

[0053] The current surrounding environment refers to the physical space in which the target UAV is located during operation and its dynamic or static characteristics, including but not limited to building density, vegetation cover, light intensity, electromagnetic interference level, and weather conditions. Environmental parameters refer to a series of quantifiable or perceptible environmental characteristics used to describe the physical space in which the target UAV is currently located and its external conditions. For example, environmental parameters include, but are not limited to, light intensity, visibility, weather conditions (such as rain, fog, and snow), electromagnetic field disturbance values, signal obstruction rate, building or obstacle density, terrain undulation, temperature, humidity, air pressure, wind speed, and wind direction. Environmental parameters may also include parameters related to sensor performance, such as the signal-to-noise ratio (SNR) of GNSS signals, the number of visible satellites, the intensity of multipath effects, and the signal strength and distribution of Wi-Fi or UWB base stations. The first confidence level refers to the degree of credibility or reliability assessment of location information obtained using a particular positioning method under current environmental conditions, and can be expressed in numerical form. For example, the higher the first confidence level of a positioning method, the higher the reliability of the positioning result (i.e., location information) obtained by that positioning method; the lower the first confidence level of a positioning method, the lower the reliability of the positioning result obtained by that positioning method.

[0054] Environmental parameters of the target drone's current surroundings can be obtained in several ways. One method is to use various sensors onboard the drone, such as temperature sensors, humidity sensors, barometers, and GPS modules, to measure environmental parameters. Another method is to use high-resolution cameras combined with computer vision technology to analyze the surrounding scene and infer environmental features, such as terrain, scene, brightness, and the location and type of obstacles, and then determine environmental parameters based on these features. Additionally, environmental parameters of the target drone's current surroundings can be received from weather stations.

[0055] Furthermore, by combining environmental parameters with a confidence mapping table, the initial confidence level of different positioning methods in the current environmental scenario can be determined. The confidence mapping table includes the potential confidence levels for different positioning methods in different environmental scenarios.

[0056] For example, the confidence mapping table can be found in Table 1 below.

[0057] Table 1

[0058]

[0059] Table 1 shows the confidence levels of various positioning methods, including LiDAR, visual recognition, infrared imaging, 5G-A base stations, and acoustic recognition, in different environmental scenarios. The sum of the first confidence levels of all positioning methods in the same environmental scenario is 1. Specifically, the allocation of the first confidence levels for the various positioning methods is shown in Table 2 below.

[0060] Table 2

[0061]

[0062] The advantage of determining the first confidence level for each positioning method based on environmental parameters is that it not only avoids positioning deviations caused by blindly relying on a single positioning source in adverse environments, but also provides a second confidence level for different positioning methods for subsequent multi-source fusion, thereby improving the overall positioning's environmental adaptability, robustness, and accuracy stability.

[0063] S130. For each positioning method, determine the indicator attributes of multiple perception quality assessment indicators associated with the target UAV under the positioning method, and determine the second confidence level corresponding to the positioning method based on multiple indicator attributes.

[0064] Among them, perception quality assessment indicators refer to quantitative or qualitative evaluation parameters used to measure the performance of a positioning method under the current positioning state. Multiple perception quality assessment indicators include at least data quality indicators, environmental impact indicators, and historical positioning reliability indicators. Data quality indicators reflect the integrity, noise level, and update frequency of the original observation data, such as image clarity, point cloud density, and IMU sampling rate. Environmental impact indicators reflect the degree of interference from the external environment on the positioning process, such as illumination changes, multipath effects, magnetic field disturbances, and occlusion rate. Historical positioning reliability indicators reflect the stability and consistency of the positioning results themselves, such as position variance, reprojection error, convergence speed, and historical trajectory smoothness. Indicator attributes refer to the specific numerical or state characteristics of each perception quality assessment indicator at the current moment. Indicator attributes include positioning data quality attributes corresponding to data quality indicators, environmental impact attributes corresponding to environmental impact indicators, and positioning reliability attributes corresponding to historical positioning reliability indicators. Positioning data quality attributes characterize the inherent reliability of the observation data used under the corresponding positioning method. Environmental impact attributes are indicator attributes used to measure the degree of interference of the environment on the positioning of the corresponding positioning method. The location reliability attribute is an indicator that reflects the long-term location stability under a given location method. The second confidence level refers to the overall reliability assessment of the current output location information of the location method after considering multiple indicator attributes.

[0065] In lidar-based positioning, the specific implementation method for determining the second confidence level is as follows: Multiple perception quality assessment indicators associated with the target UAV are acquired, including data quality indicators such as point cloud density, effective echo rate, and ranging noise level; environmental impact indicators cover the degree of laser attenuation caused by weather conditions like fog, rain, and snow, the proportion of highly reflective or low-reflective materials in the environment, and the density of dynamic obstacles; historical positioning reliability indicators include the convergence residual of point cloud registration (such as NDT or ICP), the smoothness of pose changes between historical frames, and the consistency with map matching. By combining these indicators, if the point cloud is dense and stable, the environmental interference with laser propagation is small, and the registration results converge well, a higher second confidence level is assigned to the lidar positioning; otherwise, it is reduced, thus achieving dynamic quantification of the positioning reliability of the current positioning method.

[0066] In the vision-based localization method, the specific implementation of determining its second confidence level is as follows: Corresponding indicator attributes are extracted from three types of perception quality assessment indicators. Data quality indicators include image clarity, number of feature points, and matching success rate. Environmental impact indicators include light intensity, scene texture richness, dynamic object occlusion ratio, and atmospheric visibility. Historical localization reliability indicators include reprojection error, keyframe tracking continuity, and recent localization success rate (number of successful frames / total number of frames). When image quality is high, ambient lighting is suitable, texture is sufficient, and pose estimation is stable, the system determines the visual localization status is good and assigns a high second confidence level. If blurring, overexposure, feature loss, or frequent tracking loss occurs, the confidence level is lowered accordingly to reflect the actual reliability of visual perception under the current conditions.

[0067] In infrared imaging-based positioning, the specific implementation method for determining the second confidence level is as follows: Analyze the corresponding index attributes. The positioning data quality attributes corresponding to the data quality index include the thermal image signal-to-noise ratio, the distinguishability of thermal feature points, and the stability of inter-frame thermal images. The environmental impact attributes corresponding to the environmental impact index mainly consider the difference between ambient temperature and target temperature, the degree of absorption of infrared radiation by smoke / water vapor, and the intensity of background thermal clutter interference. The positioning reliability attributes corresponding to the historical positioning reliability index are the confidence score of thermal feature matching, the magnitude of the covariance of pose calculation, and the rationality of continuous positioning trajectories. When the thermal contrast between the target UAV and the background in the infrared image is high, atmospheric transmission is good, and thermal feature matching is stable, the second confidence level of infrared positioning is set to a higher level. However, in scenarios with small temperature differences, dense fog, or severe thermal interference, the confidence level is reduced to adapt to the high sensitivity of infrared perception to the thermal environment.

[0068] In the positioning method based on 5G-A (5G-Advanced) base stations, the specific implementation method for determining its second confidence level is as follows: the positioning data quality attributes corresponding to its data quality indicators include Received Signal Reference Power (RSRP), Channel State Information (CSI) accuracy, time / angle measurement resolution, and data update frequency; the environmental impact attributes corresponding to the environmental impact indicators cover multipath effect intensity, non-line-of-sight (NLOS) propagation probability, base station geometric distribution (such as GDOP value), and electromagnetic interference level; the positioning reliability attributes corresponding to the historical positioning reliability indicators are the standard deviation of multiple positioning results, the position deviation trend with auxiliary sensors (such as IMU), and the continuity of historical positioning trajectories. When the signal is strong, the base station layout is reasonable, the NLOS impact is small, and the measurement is stable, a higher second confidence level is assigned to 5G-A positioning; in urban canyons, indoor obstructions, or sparsely populated base station areas, the second confidence level is dynamically lowered according to the degree of indicator degradation.

[0069] In acoustic wave-based positioning, the specific implementation method for determining its second confidence level is as follows: the positioning data quality attributes corresponding to its data quality indicators include the signal-to-noise ratio of the sound source signal received by the microphone array, and the measurement accuracy of the time difference of arrival (TDoA) or direction of arrival (DoA); the environmental impact attributes corresponding to the environmental impact indicators involve the environmental noise level, reverberation time, the influence of air temperature and humidity on the speed of sound, and the degree of obstruction or reflection of sound waves by obstacles; the positioning reliability attributes corresponding to the historical positioning reliability indicators are reflected in the geometric consistency of the sound source positioning solution, the consistency of the multi-band fusion results, and the rationality of the sound source position movement at continuous moments. In quiet, open, and acoustically stable environments, if the sound signal is clear, the direction finding is accurate, and the trajectory is smooth, the second confidence level of acoustic wave positioning is high; while in noisy, high-reverberation, or complex acoustic scenarios with strong reflections, its confidence level is reduced accordingly to truly reflect the effectiveness of acoustic positioning in the current environment.

[0070] By systematically analyzing multiple perception quality assessment indicators (covering three dimensions: data quality, environmental impact, and positioning reliability) associated with each positioning method, and comprehensively determining the second confidence level based on the corresponding indicator attributes, a multi-faceted and dynamic evaluation of positioning performance is achieved. Compared to relying solely on a single environmental parameter or coarse signal strength judgment, this improves the accuracy and granularity of confidence level assessment, thereby ensuring the reliability and accuracy of subsequent multi-source positioning.

[0071] In this embodiment, the environmental parameters include light intensity; determining the index attributes of various perception quality assessment indicators associated with the target UAV under the positioning method includes: when the positioning method includes a visual recognition-based positioning method, determining the third confidence level of the target UAV's location information based on the positioning method, and determining the positioning data quality attribute under the data quality index; determining the environmental impact attribute under the environmental impact index based on the light intensity in the environmental parameters; determining the total number of frames used to continuously locate the target UAV based on the positioning method in the first historical time period before the current positioning time, and the number of successful frames that have continuously and successfully located the historical location information of the target UAV, and determining the positioning reliability attribute under the historical positioning reliability index based on the number of successful frames and the total number of frames.

[0072] The third confidence score refers to the original confidence score attached to the location information directly output by the visual recognition positioning method. For example, it can be generated by the internal evaluation mechanism of the visual algorithm (such as feature matching score, optimized residual, covariance matrix, etc.). Illumination intensity characterizes the brightness of the surrounding scene, and this parameter affects image quality. The first historical time period refers to a preset time window (e.g., the most recent 5 seconds or 100 frames) prior to the current positioning time. The total number of frames refers to the total number of image frames processed by the visual positioning method within the first historical time period. The number of successful frames refers to the number of frames from the total number of image frames from which valid location information was successfully calculated.

[0073] In this embodiment, the third confidence score of the location information output by the visual positioning algorithm used in the visual recognition-based positioning method can be mapped to a positioning data quality attribute. In this method, after completing pose estimation in each frame, the visual positioning algorithm generates a confidence score, i.e., the third confidence score, for the output location information based on factors such as the number of feature points, reprojection error, and optimization convergence. For example, the visual positioning algorithm can be a YOLO model. For instance, when features are abundant, matching is stable, and optimization residuals are small, a high third confidence score indicates excellent image data quality, thus assigning a higher positioning data quality attribute value.

[0074] Environmental impact attributes can be determined based on real-time acquired illumination intensity. In this approach, if the illumination intensity is within a moderate range (e.g., 300–1000 lux) and the image is neither overexposed nor underexposed, the environment is considered favorable for visual positioning, and a higher environmental impact attribute value is assigned. Conversely, if the illumination intensity is too strong (leading to highlight clipping) or too weak (resulting in high image noise and loss of detail), the environmental interference is considered severe, and the environmental impact attribute is correspondingly reduced. Alternatively, the illumination intensity can be normalized or standardized to obtain the environmental impact attribute.

[0075] For example, the environmental impact attribute under the environmental impact indicator can be determined based on formula (3). Formula (3) is expressed as: Lux is the unit of light intensity. 2000 Lux is a reference value or threshold, which can be considered a normal or ideal light level. By comparing the current light intensity with the reference value to achieve dimensionlessness, and taking the square root of the comparison value, the numerical range can be compressed and the sensitivity under high light conditions can be reduced.

[0076] The reliability of a vision-based localization system can be determined by statistically analyzing its historical success rate. This method continuously records the visual localization process within a first historical time period, calculating the total number of frames (i.e., the total number of image frames processed) and the number of successful frames (i.e., the number of frames that successfully output a valid position). The ratio of successful frames to total frames is the localization success rate, which characterizes the stability and continuity of visual localization in recent operations. The localization success rate can be used as a localization reliability attribute. Alternatively, a high success rate (e.g., exceeding 90%) indicates robust operation of the vision system, thus assigning a high localization reliability attribute; frequent tracking loss or failure to converge reduces this attribute.

[0077] For example, the total number of frames within each first historical period can be preset. Total Frames The system was successfully and stably positioned at 50 frames per second. Then locate reliable attributes =0.5. Update frame by frame. If the target drone is successfully detected in the current frame and is the same as in the previous frame, then the target has been successfully located to frame 51. The latest min( )=51, Location Reliability Attribute =0.51.

[0078] The technical solution provided in this embodiment transforms the third confidence level within visual positioning into a positioning data quality attribute, constructs an environmental impact attribute based on illumination intensity, and defines a positioning reliability attribute using historical positioning success rates. These three attributes are then combined to determine the second confidence level corresponding to the visual positioning method. This achieves a multi-dimensional, dynamic, and refined evaluation of visual positioning performance. It not only considers the intrinsic quality of the data itself and the real-time impact of the external environment but also introduces stability considerations over time, thereby improving the comprehensiveness and accuracy of the confidence level assessment. The resulting second confidence level can more reasonably guide the allocation of multi-source fusion weights or the switching of positioning strategies, enhancing the robustness and accuracy of UAV positioning in scenarios with drastic changes in illumination, complex textures, or frequent dynamic interference.

[0079] In this embodiment, the environmental parameters include ambient temperature and the angle between the target luminescent object and the reference plane. The system determines the attribute of various perception quality assessment indicators associated with the target UAV under the positioning method, including: when the positioning method includes infrared imaging-based positioning, determining the target temperature difference based on the target UAV's current total heat generation power, effective heat dissipation area, and a predetermined heat transfer coefficient; and determining the positioning data quality attribute under the data quality indicator based on the target temperature difference and the ambient temperature in the environmental parameters; determining the environmental impact attribute under the environmental impact indicator based on the angle between the target luminescent object and the reference plane in the environmental parameters; and determining the positioning reliability attribute under the historical positioning reliability indicator based on the matching degree between the hot spot morphology features detected by the positioning method and the hot spot template during the second historical time period before the current positioning time.

[0080] The current total heat generation power refers to the total heat generated by the target UAV per unit time under its current operating state, which can originate from the power consumption of the motor, battery, and electronic components. Effective heat dissipation area refers to the effective surface area of ​​the target UAV that can exchange heat with the environment. The heat transfer coefficient is a physical parameter used to characterize the heat conduction and convection capabilities of the target UAV. Target temperature difference refers to the difference between the average surface temperature of the target UAV and the temperature of its surrounding environment. Ambient temperature refers to the air temperature of the current ambient space. Positioning data quality attributes refer to attributes reflecting the thermal contrast quality of infrared images, determined based on the target temperature difference and ambient temperature, under the data quality index dimension. The target emitting object can be the sun. The reference plane can be the ground or a fixed reference surface. The included angle refers to the spatial angle between the target emitting object and the normal to the reference plane, affecting the infrared imaging viewing angle and thermal radiation reception efficiency. The environmental impact attribute is an index attribute determined based on this included angle, measuring the degree of interference of environmental geometric factors on infrared positioning. The second historical time period refers to a preset time window (such as the most recent few seconds or frames) before the current positioning time. Hot spot morphology characteristics refer to the shape, size, temperature distribution, and other features of the hot area presented by the target UAV in the infrared image. The hotspot template is a pre-established standard model representing the hotspot of a target UAV under its current positioning state. The matching degree refers to the morphological similarity between the current hotspot morphology and the hotspot template. The positioning reliability attribute is an indicator reflecting the long-term consistency of infrared positioning, determined based on this matching degree.

[0081] In this embodiment, the target temperature difference can be calculated based on a thermodynamic model, and the positioning data quality attribute can be determined accordingly. In this method, the average surface temperature rise of the target UAV can be estimated using a steady-state thermal balance formula based on its current total heat generation power, effective heat dissipation area, and heat transfer coefficient, thus obtaining the target temperature difference. This difference is then combined with the real-time difference between the ambient temperature and the target temperature to determine the positioning data quality attribute. For example, if the target temperature difference is large (e.g., above 5°C), it indicates that the target hotspot is clearly identifiable in the infrared image, assigning a high positioning data quality attribute; if the target temperature difference is too small (e.g., close to the ambient noise level), the hotspot is easily obscured, correspondingly reducing the positioning data quality attribute to reflect the inherent quality of the data.

[0082] For example, the current total heat output of the target UAV can be determined based on formula (4). Formula (4) is expressed as: ;in, It is the motor power of the target drone. It is the battery power of the target drone; This represents the current total heat generation power. The steady-state heat balance formula can be expressed as: ;in, The target temperature difference is represented by A; the effective heat dissipation area is represented by h; and the heat transfer coefficient is represented by h, which is related to wind speed, altitude, and temperature. The heat transfer coefficient can be determined based on formula (5). Formula (5) is expressed as: T is the temperature correction value, and P is the altitude correction value. ; The base natural convection coefficient (a fixed value) is used, and v is the real-time wind speed in the environmental parameters. , These are the fitting parameters, which can be obtained by querying the characteristic parameters of the UAV model. The positioning data quality attribute under the data quality index can be determined based on formula (6). Formula (6) is expressed as: Positioning data quality attribute = 1 - |ambient temperature - target temperature difference| / 50℃. Where, || represents the absolute value. ℃ represents the unit of temperature (degrees Celsius).

[0083] In this embodiment, the attitude angle (such as pitch or tilt) of the target emitting object relative to the ground (reference plane) can be obtained as the angle between the target emitting object and the reference plane. The environmental impact attribute can then be determined based on the size of this angle. For example, a smaller angle indicates weaker solar radiation, potentially leading to hot spots being blocked, deformed, or experiencing severe radiation attenuation, resulting in decreased image quality; thus, a lower environmental impact attribute can be assigned. Conversely, a larger angle can be assigned a higher environmental impact attribute. Alternatively, the cosine of the angle can be used as the environmental impact attribute.

[0084] For example, the environmental impact attribute can be determined based on formula (7); formula (7) is expressed as: Environmental impact attribute = Cos(solar angle). Cos represents the cosine function.

[0085] In this embodiment, the positioning reliability attribute can be evaluated through the stability of historical hotspot matching. In this method, hotspot morphological features (such as isotherm contours, centroid shift, and temperature gradient distribution) are continuously extracted from infrared images during the second historical period and compared with pre-stored hotspot templates to calculate the matching degree for each frame of infrared image. If the matching degree of most frames of infrared images is high and fluctuates little, it indicates that the hotspot morphology is stable and the target recognition is reliable, thus assigning a high positioning reliability attribute; if the matching degree is low, and the hotspot frequently deforms or disappears, it indicates that the positioning result is unreliable, and the positioning reliability attribute is correspondingly lowered. Alternatively, the matching degree corresponding to each frame of infrared image during the second historical period can be averaged to obtain the mean value. This mean value is then used as the positioning reliability attribute.

[0086] The technical solution provided in this embodiment determines the positioning data quality attribute by combining the target temperature difference and the ambient temperature in the environmental parameters, constructs the environmental impact attribute based on the angle between the target luminescent object and the reference plane, and defines the positioning reliability attribute by using the matching degree between historical hot spots and templates. Then, it integrates these three attributes to jointly determine the second confidence level corresponding to the infrared imaging positioning method. This realizes a multi-dimensional, physical mechanism-driven confidence level assessment of infrared positioning performance. It not only considers the physical characteristics of the heat source itself, but also integrates the consistency between the solar angle and historical infrared imaging positioning, improving the accuracy of the confidence level assessment of the infrared imaging positioning method and enhancing the positioning accuracy of UAVs in complex scenarios such as night, smoke, and low light.

[0087] It should be noted that the method for determining the second confidence level for a given positioning method can be the same, based on multiple indicator attributes for different positioning methods. Taking the determination of the second confidence level for any given positioning method as an example, the specific implementation can be as follows: based on the indicator weights corresponding to multiple perception quality assessment indicators, a weighted sum is calculated of the positioning data quality attribute, environmental impact attribute, and positioning reliability attribute corresponding to that positioning method to obtain the second confidence level for that positioning method. The sum of the indicator weights corresponding to all perception quality assessment indicators is 1. The indicator weights are used to characterize the importance of the perception quality assessment indicators for UAV positioning.

[0088] For example, the second confidence level corresponding to the positioning method can be determined based on formula (8). Formula (8) is expressed as: ;in, Indicates the first The second confidence level of this positioning method. The weights of the data quality indicators. The weights of environmental impact indicators. The weights of historical positioning reliability indicators.

[0089] It should also be noted that the weights of the various perceived quality assessment indicators may differ depending on the positioning method.

[0090] S140. Based on the first and second confidence levels corresponding to each positioning method, and all location information, determine the target positioning result corresponding to the target UAV.

[0091] Among them, the target positioning result refers to the final positioning result of the target UAV.

[0092] In this embodiment, the location information determined by the positioning method corresponding to the highest confidence level among all confidence levels in the first and second confidence levels can be used as the target positioning result for the target UAV. Alternatively, a portion of the confidence levels can be selected from all confidence levels, and the location information corresponding to the selected portion of confidence levels can be fused to obtain the fused location information, which can then be used as the target positioning result for the target UAV. Alternatively, the location information corresponding to the positioning method with the highest priority among all or a portion of the confidence levels can be used as the target positioning result. Alternatively, the first and second confidence levels of the same positioning method can be fused to determine the comprehensive confidence level of that positioning method. The location information and the corresponding comprehensive confidence level of each positioning method are multiplied to obtain the product location information corresponding to each positioning method. All product location information is summed to obtain the target positioning result for the target UAV.

[0093] For example, the overall confidence level of the positioning method can be determined based on formula (8). Formula (8) is expressed as: ;in, Indicates the first The overall confidence level of the positioning methods. This indicates the weight of the second confidence level; This represents the weight of the second confidence level. Indicates the first The first confidence level of this positioning method; Indicates the first The second confidence level of this positioning method.

[0094] The target location result of the UAV can be determined based on formula (9). Formula (9) is expressed as: ; Indicates the first Location information determined by this positioning method. This indicates the target location result of the target drone.

[0095] To improve the reliability of UAV positioning and the safety of UAV operation, after determining the target positioning result corresponding to the target UAV, the positioning data reported by the target UAV can be received; a consistency check can be performed based on the target positioning result and the positioning data to obtain positioning deviation data; based on the signal-to-noise ratio, the threshold can be dynamically adjusted and the positioning deviation data can be used to determine whether to trigger an early warning prompt; if so, an early warning prompt message can be generated based on the positioning deviation data.

[0096] The positioning data reported by the target UAV refers to the location information acquired and actively transmitted by the target UAV through its onboard positioning module (such as GNSS, IMU, VIO, etc.). Positioning deviation data refers to the spatial distance difference or vector deviation between the target positioning result and the reported positioning data, used to quantify the degree of inconsistency. The dynamic signal-to-noise ratio (SNR) adjustment threshold is determined based on the second SNR associated with the target UAV, used to determine how large a positioning deviation constitutes an anomaly. The second SNR is the ratio of the signal power to the noise power associated with the target UAV, used to measure the strength of the useful signal generated by the target UAV relative to the background noise. Warning information refers to the alarm information generated when the positioning deviation data exceeds the dynamic threshold.

[0097] It should be noted that the positioning data reported by the target UAV relies on its onboard sensors, which may exhibit significant deviations or even be maliciously tampered with in complex electromagnetic environments, GNSS denied areas, or sensor malfunctions. Relying on only one source creates a security blind spot. By introducing consistency verification and dynamically adjusting the signal-to-noise ratio (SNR) threshold based on the target UAV's own generated second SNR, a warning alert is determined to avoid false alarms caused by noise. This enables intelligent identification and timely alerts for positioning anomalies (such as sensor drift, signal spoofing, communication interference, or flight control malfunctions).

[0098] Specifically, spatial consistency can be checked between the positioning data reported by the target UAV and the target positioning result to obtain positioning deviation data. The signal-to-noise ratio dynamic adjustment threshold can be determined based on formula (10). Formula (10) means: ;in, This indicates the threshold for dynamic adjustment of the signal-to-noise ratio; This is represented as the second signal-to-noise ratio. If the positioning deviation data exceeds this dynamic adjustment threshold of the signal-to-noise ratio, an alarm is triggered, and a warning message containing the magnitude and direction of the deviation and possible causes of the anomaly can be generated.

[0099] Furthermore, during the consistency verification process, if multiple positioning deviation data continuously exceed the corresponding dynamic signal-to-noise ratio adjustment threshold (e.g., exceeding the limit for two consecutive frames), it is determined that there is a positioning anomaly, triggering an early warning prompt, and noting in the prompt message that it may be caused by airborne positioning failure, communication interference, or external deception attacks.

[0100] The advantage of this setup is that by receiving the positioning data reported by the target UAV after obtaining the target positioning result, and performing consistency verification based on the two, and then combining it with a dynamic threshold mechanism driven by the second signal-to-noise ratio to determine whether to trigger an early warning, a closed-loop monitoring and anomaly detection of the UAV's positioning status is achieved. This not only reduces the false alarm rate and the missed alarm rate, but also enables timely issuance of UAV anomalies and early warnings, thereby enhancing the operational safety of the UAV.

[0101] The technical solution provided in this embodiment determines the location information of a target drone based on at least two positioning methods; acquires environmental parameters of the target drone's current surrounding environment, and determines a first confidence level corresponding to each positioning method based on the environmental parameters; for each positioning method, determines the indicator attributes of multiple perception quality assessment indicators associated with the target drone under the positioning method, and determines a second confidence level corresponding to the positioning method based on the multiple indicator attributes; wherein, the multiple perception quality assessment indicators include data quality indicators, environmental impact indicators, and historical positioning reliability indicators; based on the first and second confidence levels corresponding to each positioning method, and all location information, determines the target positioning result corresponding to the target drone, thus solving the problems of existing technologies. The single-sensor-based detection method in China suffers from poor accuracy in UAV positioning. This paper addresses this issue by fusing at least two positioning methods to obtain the target UAV's location information and dynamically evaluating the first confidence level of each method in the current scenario, incorporating environmental parameters. Furthermore, it introduces multiple perception quality assessment indicators associated with the target UAV (including data quality indicators, environmental impact indicators, and historical positioning reliability indicators), and analyzes the attributes of these indicators to determine the second confidence level of each positioning method. Finally, it combines the first confidence level, the second confidence level, and the location information provided by each positioning method, weighted and fused to generate a high-precision, highly robust target positioning result, thus improving the accuracy and reliability of UAV positioning in complex and ever-changing environments.

[0102] Figure 2 This is a flowchart of a positioning result determination method according to an embodiment of the present invention. Based on the foregoing embodiments, when the positioning method includes a lidar-based positioning method, the location information of the target UAV can be determined using a lidar device, and further, the index attributes of various perception quality assessment indicators associated with the target UAV under the positioning method can be determined. Specific implementation methods can be found in the technical solution of this embodiment. Technical terms that are the same as or corresponding to those in the above embodiments will not be repeated here.

[0103] like Figure 2 As shown, the method specifically includes the following steps:

[0104] S210. When the positioning method includes a lidar-based positioning method, the airspace containing the target UAV is scanned by a lidar device to obtain reflection point cloud data.

[0105] In this embodiment, the lidar device refers to an active optical remote sensing device installed on a ground station, base station, or other fixed / mobile platform. It senses the environment by emitting laser pulses and receiving the echoes reflected from the surface of objects. The airspace containing the target UAV refers to the three-dimensional spatial area covered by the lidar scan, in which the target UAV is in flight or hovering. The reflection point cloud data refers to the collection of a large number of discrete point clouds acquired by the lidar after scanning the airspace. Each point cloud represents the measurement result of a valid laser echo. The reflection point cloud data contains multiple point clouds with three-dimensional coordinates and reflection intensities. The three-dimensional coordinates refer to the spatial position of each point cloud in the global or local coordinate system (which can be represented as x, y, z). The reflection intensity refers to the amplitude or energy intensity of the laser echo signal, which is related to factors such as the target surface material, incident angle, and distance.

[0106] In a specific implementation, a pre-deployed lidar device can periodically scan a preset monitoring airspace to capture target drones. The lidar device can be deployed in a fixed location, covering a specific low-altitude area with its field of view and ranging range. It continuously emits laser beams and processes the returned signals to generate reflective point cloud data that includes background (such as buildings and trees) and dynamic targets (such as target drones).

[0107] Alternatively, the lidar device can be mounted on a rotatable gimbal, vehicle-mounted platform, or another drone. It can actively adjust its scanning direction based on the target drone's approximate position information, concentrating the laser beam to perform high-density, high-frequency scanning of the area where the target drone is located, thereby acquiring more refined reflection point cloud data. Since the target drone is in motion, predictive filtering (such as Kalman filtering) can be used to guide the lidar's pointing in real time, and clusters of points belonging to the same target drone can be correlated in consecutive frame point clouds to improve point cloud quality and positioning stability.

[0108] Alternatively, multiple LiDAR devices can be integrated to perform a coordinated scan of the airspace containing the target UAV, thereby enhancing the completeness of the detection. For example, multiple LiDAR devices can be deployed at different locations, each scanning overlapping or adjacent airspace, and their respective reflected point cloud data can be synchronized in time and then fused into a unified coordinate system. By overlaying point clouds from multiple perspectives, the problem of sparse point clouds caused by occlusion, angle limitations, or low reflectivity of a single device can be effectively compensated for, resulting in a more complete target geometric contour and more reliable three-dimensional point cloud coordinates.

[0109] S220. Perform spatial clustering on the valid point clouds in the reflection point cloud data to obtain at least one point cloud cluster.

[0110] Valid point clouds satisfy preset validity conditions. Specifically, valid point clouds refer to points selected from reflective point cloud data that meet these preset validity conditions. These points are more likely to originate from real targets (such as target drones) rather than noise or irrelevant background. The preset validity conditions are a set of rules used to determine the validity of point clouds, including: distance range constraints (limiting the distance from the point to the lidar to be within a reasonable detection range, excluding invalid echoes that are too close or too far), reflection intensity threshold conditions (requiring the reflection intensity of the point to be higher than a set threshold to filter low-energy noise or weak clutter), spatial neighborhood density conditions (requiring the point to have a certain number of neighboring points in its local neighborhood to exclude isolated noise points), and geographical altitude restrictions (limiting the z-coordinate of the point to be within the possible flight altitude range of the target drone, such as 30 meters to 120 meters). A point cloud cluster refers to each group of points formed after clustering valid point clouds, corresponding to a potential physical entity (such as a drone, a bird, or a static obstacle).

[0111] In this embodiment, the original reflection point cloud data can be filtered point by point according to preset validity conditions, retaining points that simultaneously meet the limitations of distance, reflection intensity, neighborhood density, and height as valid point clouds. Using Euclidean distance as a metric, a clustering algorithm is used to automatically identify high-density regions and divide them into independent point cloud clusters, while outliers in low-density regions are removed. This method does not require pre-specifying the number of clusters, can effectively handle irregularly shaped targets, and suppress residual noise. Optionally, clustering algorithms include K-Means clustering, Hierarchical Clustering, Density-Based Noise Applied Spatial Clustering (DBSCAN), Gaussian Mixture Model (GMM), Spectral Clustering, Mean Shift Clustering, Agglomerative Clustering, BIRCH Clustering, OPTICS Clustering, and Fuzzy C-Means Clustering (FCM), etc. In this embodiment, no specific limitation is made on the clustering algorithm.

[0112] Considering the characteristic that the density of LiDAR point clouds becomes sparser with increasing distance, the clustering neighborhood radius of each effective point cloud can be dynamically adjusted according to the distance from the LiDAR. For example, a smaller radius is used for nearby points to avoid excessive merging, while a larger radius is used for distant points to compensate for the sparsity of the point cloud. Based on satisfying all preset validity conditions, this ensures that targets at different distances can be completely clustered into independent point cloud clusters, thereby improving the detection completeness and positioning stability of long-range small UAVs.

[0113] S230. Determine the location information of the target UAV based on at least one point cloud cluster.

[0114] In this embodiment, candidate clusters can be selected from at least one point cloud cluster based on preset UAV geometric features, and the reflection intensity of multiple point clouds in the candidate clusters can be used as weights; alternatively, the reflection intensity of the point clouds and their weights are positively correlated. The three-dimensional coordinates of the point clouds are multiplied by their weights to obtain the weighted three-dimensional coordinates. The weighted three-dimensional coordinates of all point clouds in the candidate clusters are summed to obtain a coordinate sum; the coordinate sum is then divided by the sum of the weights of all point clouds in the candidate cluster to obtain the spatial center position of the candidate cluster. This spatial center position is determined as the position information of the target UAV.

[0115] For example, a lidar device can be used to emit electromagnetic laser waves and receive reflected signals. The data it collects includes, but is not limited to, distance, azimuth, pitch, altitude, and velocity, to further detect, locate, and track target drones. It should be noted that the lidar device can also include both fiber lidar and single-photon lidar operating modes. Single-photon lidar is used to enhance detection sensitivity in low-light environments to adapt to different environmental needs. It automatically selects between fiber lidar and single-photon lidar based on the ambient light intensity. For instance, when the ambient light intensity is <10 Lux, it forcibly switches to single-photon mode, calculating the target drone's position information from the reflected point cloud data. .

[0116] After measuring multiple valid point clouds, a point cloud cluster is obtained; the point cloud cluster contains the first point cloud. The three-dimensional coordinates of an effective point cloud can be represented as: ;in, ; c is the speed of light. Time of flight, which is the time elapsed from the emission of the laser to its reflection and reception by the target. The pitch angle of the target drone. The azimuth angle of the target UAV is given. The position information of the target UAV is obtained by calculating the weighted centroid of the point cloud cluster. = . . It refers to the weights of the point cloud, which are related to the reflection intensity of the point cloud. Positive correlation.

[0117] S240. Based on the number of invalid point clouds in the reflected point cloud data and the total number of point clouds in the reflected point cloud data, determine the proportion of invalid point clouds, and based on the proportion of invalid point clouds, determine the positioning data quality attributes under the data quality indicators.

[0118] Invalid point clouds refer to points in the reflected point cloud data that do not meet the preset validity conditions. These points may originate from environmental noise, multipath interference, non-target clutter (such as flying insects, raindrops, dust), or abnormal echoes exceeding reasonable physical limits. The total number of point clouds refers to the total number of points in the reflected point cloud data. The invalid point cloud ratio is the ratio of the number of invalid point clouds to the total number of point clouds, used to quantify the proportion of invalid components in the current point cloud data. The positioning data quality attribute refers to the specific attribute value determined based on the invalid point cloud ratio under the data quality indicators, used to characterize the overall cleanliness and reliability of the current LiDAR positioning input data.

[0119] In this embodiment, each point in the reflection point cloud data can be judged according to preset validity conditions, the number of invalid point clouds can be counted, and the ratio between the number of invalid point clouds and the total number of point clouds can be used as the invalid point cloud ratio. The invalid point cloud ratio can be used as a positioning data quality attribute. Alternatively, the invalid point cloud ratio can be used as an input variable and substituted into a monotonically decreasing continuous function (such as an exponential decay function or a reciprocal function), so that when the invalid point cloud ratio approaches 0, the positioning data quality attribute is close to its maximum value; and as the invalid point cloud ratio increases, the attribute value decreases rapidly, especially after the ratio exceeds a certain critical value.

[0120] Alternatively, the mean and variance of the proportion of invalid point clouds in the first historical time period prior to the current positioning time can be referenced to determine whether the current proportion of invalid point clouds is due to sudden interference or continuous degradation. If the current proportion of invalid point clouds is higher than the historical mean and exceeds the standard deviation range, it is considered an abnormal condition, and the positioning data quality attributes can be appropriately reduced; otherwise, relatively stable positioning data quality attributes are maintained. This setting can enhance robustness to transient interference and avoid over-evaluation of data quality due to brief environmental disturbances (such as birds flying overhead or brief rainfall).

[0121] Alternatively, the proportion of invalid point clouds can be input into formula (11) to obtain the location data quality attribute; formula (11) is expressed as: 1 - proportion of invalid point clouds / 0.3. Based on formula (11), it can be seen that the location data quality attribute decreases linearly with the increase of the proportion of invalid point clouds. When the proportion of invalid point clouds is 0, the location data quality attribute reaches its maximum value of 1, indicating that the data quality is optimal; when the proportion of invalid point clouds increases to 0.3 (i.e. 30%), the attribute value drops to 0, indicating that the data quality has deteriorated to an unacceptable level; if the proportion of invalid point clouds exceeds 0.3, the attribute value will become negative, which can be truncated to 0 or regarded as invalid in practical applications. Therefore, linear decay modeling of data quality is realized in the effective range of 0 to 0.3, reflecting the reasonable logic that the higher the proportion of invalid point clouds, the lower the reliability of location data.

[0122] S250. Based on meteorological visibility in environmental parameters, determine the visibility attenuation factor, and based on the visibility attenuation factor, determine the environmental impact attributes under environmental impact indicators.

[0123] Meteorological visibility represents the maximum horizontal distance at which a target can be clearly identified by the human eye or optical sensors in the atmosphere. The unit can be meters or kilometers, and it is affected by weather phenomena such as fog, haze, rain, snow, and dust storms. The "visibility attenuation factor" is a dimensionless value calculated based on meteorological visibility, used to characterize the degree of atmospheric attenuation of light waves (including lasers). Its value increases as visibility decreases. Environmental impact attributes refer to the specific attribute values ​​determined based on the visibility attenuation factor under environmental impact indicators, used to quantify the adverse impact of current meteorological conditions on lidar-based positioning methods.

[0124] In this embodiment, the meteorological visibility V and the visibility attenuation factor can be pre-established. The mapping relationship between them. For example, the mapping relationship can be represented as: Once the meteorological visibility is obtained from the environmental parameters, it can be substituted into the mapping relationship to obtain the visibility attenuation factor. Furthermore, the visibility attenuation factor can be substituted into formula (12) to obtain the environmental impact attribute. Formula (12) can be expressed as: e represents the natural exponential function.

[0125] According to formula (12), the first environmental impact attribute decreases monotonically with the increase of the visibility attenuation factor, and exhibits a non-linear exponential decay trend. When the visibility attenuation factor is 0 (i.e., meteorological visibility is extremely high and there is no atmospheric attenuation), the first environmental impact attribute reaches its maximum value of 1, indicating that the environment has almost no interference with the positioning method; as the visibility attenuation factor increases (e.g., due to fog, haze, rain, etc., the atmospheric transmittance decreases), the attribute value decreases rapidly, but the attenuation rate gradually slows down; for example, when the attenuation factor increases to 4, the attribute value further decreases to about 0.14. Therefore, the environmental impact attribute reflects the negative impact of deteriorating meteorological conditions on optical or radio frequency sensing performance, demonstrating the physical law that the worse the environment, the more severe the positioning interference, and the lower the environmental impact attribute.

[0126] S260. Based on the historical accuracy attributes of the target UAV located by the positioning method before the current positioning time, determine the positioning reliability attributes under the historical positioning reliability index.

[0127] In this context, "current positioning time" refers to the point in time when the positioning operation is being performed. "First historical period" refers to a preset time window preceding the current positioning time (e.g., the most recent 5 seconds, 10 seconds, or several consecutive positioning cycles), used for retrospective analysis of historical positioning performance. "Accuracy attribute" refers to the error measure between each positioning result from the LiDAR method and the actual location (or high-confidence reference location) within this first historical period. It can be expressed as the magnitude of the positioning deviation, directional consistency, or standard deviation of the error. "Positioning reliability attribute" refers to the specific attribute value determined based on the positioning accuracy attribute within the first historical period, under the historical positioning reliability index. It characterizes the recent positioning reliability of the LiDAR-based positioning method. Historical accuracy attributes can describe the accuracy (i.e., accuracy rate) of the location information obtained from positioning the target UAV using this positioning method before the current positioning time.

[0128] In this embodiment, the historical accuracy attribute can be determined by the error value between the target UAV's location information determined by the current positioning method and the final target positioning result. A higher error value results in a lower historical accuracy attribute, and vice versa. Alternatively, the accuracy attribute can be calculated as the ratio of the number of times the positioning method successfully outputs valid location information (i.e., no tracking loss, feature matching failure, or solution divergence occurred) to the total number of attempts within a preset historical period. The accuracy attribute of positioning the target UAV based on the positioning method within a preset historical period before the current positioning time can be calculated; or, the historical accuracy attribute of positioning the target UAV based on the positioning method for a preset number of times before the current positioning time. The average or root mean square error of all calculated historical accuracy attributes is then applied to obtain the positioning reliability attribute under this positioning method.

[0129] The technical solution provided in this embodiment, when the positioning method includes a lidar-based positioning method, uses data quality indicators to reflect the availability of point cloud data in real time, environmental impact indicators to dynamically perceive the attenuation effect of meteorological conditions on laser propagation, and historical positioning reliability indicators to comprehensively evaluate the long-term performance stability of the system. This achieves a second confidence level for the positioning method through multi-indicator collaborative evaluation, enabling intelligent perception of lidar performance changes in complex weather (such as fog, haze, rain, and snow) and dynamic target scenarios. This ensures the accuracy of target UAV positioning, effectively suppresses false positioning and false alarms in harsh environments, and improves the positioning accuracy, robustness, and safety of UAVs in low-altitude complex environments.

[0130] Figure 3This is a flowchart of a positioning result determination method according to an embodiment of the present invention. Based on the foregoing embodiments, when the positioning method includes a base station-based positioning method, the location information of the target UAV can be determined through the base station, and further, the indicator attributes of various perception quality assessment indicators associated with the target UAV under the positioning method can be determined. Specific implementation methods can be found in the technical solution of this embodiment. Technical terms that are the same as or corresponding to those in the above embodiments will not be repeated here.

[0131] like Figure 3 As shown, the method specifically includes the following steps:

[0132] S310. When the positioning method includes base station-based positioning, the base station transmits radio waves into the airspace containing the target UAV to obtain the reflected signal parameters.

[0133] Base station-based positioning refers to using ground-deployed wireless communication or sensing base stations (such as 5G-A sensing) to detect and locate target drones. A base station is a fixed or mobile infrastructure capable of transmitting and receiving wireless signals, such as a 5G-A (5G-Advanced) base station, millimeter-wave radar base station, or radio frequency sensing node. The airspace containing the target drone refers to the three-dimensional spatial area covered by the base station antenna beam, where the target drone is in flight and may reflect or scatter incident wireless signals. Wireless waves refer to electromagnetic wave signals actively transmitted by the base station, with frequencies covering Sub-6 GHz, millimeter waves (such as 24 GHz, 28 GHz, 39 GHz), or sensing bands. Reflected signal parameters refer to the physical quantity information contained in the wireless signal received by the base station and reflected back by the target drone.

[0134] In this embodiment, a base station can transmit directional millimeter-wave signals into the airspace and receive the backscattered echoes from the target UAV. In this method, the base station can operate in multi-station or single-station mode, using an antenna array to form a narrow beam to scan a preset low-altitude area. When the radio wave illuminates the target UAV, its metal structure or fuselage surface will generate effective reflection, and the reflected signal parameters are recorded. Optionally, the reflected signal parameters include, but are not limited to, time of arrival (ToA), time difference of arrival (TDoA), round-trip time (RTT), angle of arrival (AoA), angle of departure (AoD), Doppler shift, channel state information (CSI), received signal strength (RSSI), signal phase difference, signal-to-noise ratio (SNR), delay, angle of arrival, and signal strength, etc.

[0135] Alternatively, based on the Integrated Sensing and Communication (ISAC) architecture, 5G-A base stations can simultaneously perform communication and sensing functions. In this approach, the base station broadcasts downlink reference signals for communication into the airspace. If the target drone is a target (such as a drone connected to the network), it can receive and send back uplink signals normally. Simultaneously, the base station extracts AoA, TDoA, and Doppler reflection signal parameters by analyzing the reflected or multipath components returning from the drone's direction (even if no communication link is established). Because 5G-A base stations have large-scale MIMO antenna arrays and high time / frequency resolution, they can achieve high-precision sensing simultaneously during communication, thereby obtaining the target drone's location information, reducing deployment costs, and improving spectrum efficiency.

[0136] S320: Based on the reflected signal parameters and the preset speed of light, determine the current distance between the target drone and the base station.

[0137] The preset speed of light refers to the speed at which wireless waves (such as electromagnetic waves or laser pulses) propagate in space. The current distance refers to the spatial distance between the target drone and the base station at the current positioning moment.

[0138] In this embodiment, the reflected signal parameters include the reflection signal delay. The reflection signal delay refers to the propagation time of a wireless wave from the transmitting source (i.e., the base station), after being reflected by the target drone, etc., and then returning to the receiving end (i.e., the base station). The reflection signal delay directly reflects the path length of the signal between the transmitting end and the target drone. Since wireless waves propagate at the speed of light, the reflection signal delay and the current distance from the target drone to the base station satisfy formula (13). Formula (13) is expressed as: ;in, Represented as the speed of light; Represented as the speed of light; This represents the current distance between the target drone and the base station. The division by 2 is because the reflected signal delay corresponds to the total time of the "outbound + return journey".

[0139] S330: Based on the current distance and the current angle information of the target drone relative to the base station, determine the position information of the target drone.

[0140] The current angle information of the target UAV relative to the base station refers to the directional parameters of the target UAV as observed from the base station's perspective, which may include azimuth and elevation. The current angle information can be obtained by the base station's antenna array through the direction of arrival (AoA) estimation method.

[0141] Specifically, the current distance and the current angle information of the target UAV relative to the base station can be input into formula (14) to obtain the position information of the target UAV.

[0142] Formula (14) can be expressed as: ;

[0143] Among them, the location information of the target drone ; Indicates the current distance. This indicates the pitch angle in the current angle information; This indicates the azimuth angle in the current angle information.

[0144] S340. Based on the first signal-to-noise ratio of the base station, determine the location data quality attributes under the data quality index.

[0145] The first signal-to-noise ratio (SNR) refers to the ratio (in dB) of the useful signal power to the background noise power in the reflected or uplink signal related to the target UAV received by the base station. It measures signal clarity and detectability. The positioning data quality attribute refers to the specific attribute value determined based on the base station's first SNR under this data quality index. It is used to quantify the quality level of the current wireless observation data, thereby affecting the overall confidence assessment of this positioning method.

[0146] In this embodiment, the first signal-to-noise ratio (SNR) of the base station can be used as the positioning data quality attribute under the data quality index; alternatively, the first SNR of the base station can be normalized to obtain the positioning data quality attribute under the data quality index. Alternatively, a continuous nonlinear function can be used to map the first SNR to the positioning data quality attribute. For example, the continuous nonlinear function can be a sigmoid function, an exponential decay function, or a logarithmic compression function.

[0147] For example, the first signal-to-noise ratio can be normalized based on formula (15) to obtain the location data quality attributes under the data quality index. Formula (15) can be expressed as: (SNR2 / 30dB) 2 .

[0148] The signal-to-noise ratio (SNR) assessment based on base stations, under the data quality index, reflects the clarity and reliability of the current wireless observation signal, and thus characterizes whether the raw data upon which this positioning method relies has sufficient effective information for accurately calculating the target location. Specifically, a high SNR means that the useful signal is much stronger than the background noise, key parameters such as ranging and direction finding are extracted accurately, and the positioning results are highly reliable; while a low SNR indicates that the signal may be severely contaminated by noise, which can easily lead to distance estimation errors, blurred angle resolution, or even missed target detection, thereby reducing positioning performance.

[0149] S350. From the total signal energy of the UAV sensing target based on the base station, determine the proportion of multipath interference caused by the multipath effect, and based on the proportion of multipath interference, determine the environmental impact attribute under the environmental impact index.

[0150] The total signal energy refers to the sum of the energy of all wireless reflected or scattered signals originating from the target UAV received by the base station, including direct path (line-of-sight, LoS) signals and non-direct path (non-line-of-sight, NLoS) signals reflected from buildings, the ground, or other obstacles. Multipath effect refers to the phenomenon where wireless signals are reflected, refracted, or scattered by obstacles during propagation, resulting in the receiver receiving multiple copies of the signal with different delays and phases simultaneously. Interference signal refers to non-direct path signals generated by multipath effect, which, when superimposed on the direct signal, may cause amplitude fading, phase distortion, or ranging / direction finding errors. The multipath interference ratio is the ratio of interference signal energy to the total signal energy, used to quantify the degree of contamination of the observed signal by multipath effect. Environmental impact attributes characterize the complexity of the current airspace electromagnetic propagation environment and its potential interference with the base station's positioning accuracy.

[0151] In this embodiment, the base station can use channel estimation techniques (such as matched filtering or compressed sensing) to extract the channel impulse response from the received signal, identifying the strongest main path (considered a direct path) and other secondary paths (considered multipath interference). The sum of the energies of all paths other than the main path is defined as the interference signal energy. Dividing the interference signal energy by the total signal energy yields the multipath interference ratio. Furthermore, the multipath interference ratio can be mapped to an environmental impact attribute. For example, a lower multipath interference ratio corresponds to a higher environmental impact attribute value (indicating less environmental interference); a higher multipath interference ratio corresponds to a lower environmental impact attribute value.

[0152] For example, environmental impact attributes can be determined based on formula (16). Formula (16) can be expressed as: 1 - multipath interference ratio / 0.5.

[0153] Formula (16) reflects the linear negative impact of multipath interference on the positioning environment quality. Specifically, when the multipath interference ratio is 0 (i.e., there is no multipath component in the received signal, and ideal line-of-sight propagation), the environmental impact attribute reaches its maximum value of 1, indicating that the environment has almost no interference with the positioning process; as the multipath interference ratio increases, the environmental impact attribute decreases linearly, indicating that the environmental complexity increases and the positioning reliability decreases; when the multipath interference ratio reaches 0.5 (i.e., 50% of the received signal energy comes from multipath interference), the environmental impact attribute drops to 0, meaning that the environmental interference has become unacceptably severe; if the multipath interference ratio exceeds 0.5, the attribute value will become negative, and in practical applications it is usually truncated to 0, which is regarded as an invalid or high-risk state.

[0154] S360. Determine the number of clock anomalies when locating the target UAV based on the positioning method within the third historical time period before the current positioning time. Based on the number of clock anomalies, determine the positioning reliability attribute under the historical positioning reliability index.

[0155] The third historical period refers to a preset time window (e.g., the most recent 10 seconds, 30 seconds, or several consecutive positioning cycles) preceding the current positioning time, used for retrospective analysis of the base station status during historical positioning processes. Clock anomalies refer to time reference errors caused by local clock drift, synchronization failure, timestamp errors, or external time synchronization interruptions during the positioning process, which may lead to serious deviations in distance or location calculations. The number of clock anomalies refers to the total number of clock-related anomaly events detected or recorded by the base station within the third historical period. Positioning reliability attributes characterize the degree to which the positioning method has been affected by clock problems recently, thus reflecting its historical operational reliability.

[0156] In this embodiment, the cumulative number of clock anomalies during the third historical time period prior to the current positioning time when locating the target UAV based on the positioning method can be obtained. Furthermore, an exponentially decaying weighted model can be employed, assigning higher weights to recent anomalies to dynamically evaluate positioning reliability attributes. For example, considering the proximity of each anomaly in the clock anomaly count, anomalies closer to the current positioning time have a greater impact on current reliability. That is, each anomaly is weighted according to its occurrence time, and all weighted anomaly values ​​are summed to obtain a comprehensive anomaly attribute; this comprehensive anomaly attribute is then converted into a positioning reliability attribute using a nonlinear function.

[0157] Alternatively, the reciprocal of the number of clock anomalies can be used as a positioning reliability attribute. This setting reflects the impact of historical clock stability on the reliability of base station positioning methods: the fewer clock anomalies, the more reliable the base station time synchronization, and the higher the credibility of the location information. Specifically, when the number of clock anomalies is 0, its reciprocal mathematically approaches infinity, but in practical applications, an upper limit can be set or smoothing can be used to avoid undefined situations, at which point the positioning reliability attribute reaches its highest level. As the number of anomalies increases, the reciprocal decreases rapidly, indicating that the time reference is frequently inaccurate, leading to unreliable timestamps in ranging, direction finding, or cooperative positioning, which in turn causes base station location calculation errors or even failure.

[0158] The technical solution of this embodiment improves the accuracy of the positioning reliability assessment of the base station positioning method by using the first signal-to-noise ratio to reflect the positioning data quality attribute, the multipath interference ratio to characterize the degree of influence of the environment on the positioning method, and the number of clock anomalies to assess the historical positioning reliability of the positioning method when the positioning method includes a base station-based positioning method. Based on these multi-source index attributes, the second confidence of the positioning method is comprehensively evaluated, thereby improving the positioning accuracy of the UAV in complex electromagnetic environments, areas with severe multipath interference, and under conditions of unstable clock synchronization.

[0159] Figure 4 This is a flowchart of a positioning result determination method according to an embodiment of the present invention. Based on the foregoing embodiments, when the positioning method includes a sound wave recognition-based positioning method, the location information of the target UAV can be determined through noise data generated by the target UAV, and further, the index attributes of various perception quality assessment indicators associated with the target UAV under the positioning method can be determined. Specific implementation methods can be found in the technical solution of this embodiment. Technical terms that are the same as or corresponding to those in the above embodiments will not be repeated here.

[0160] like Figure 4 As shown, the method specifically includes the following steps:

[0161] S410. When the positioning method includes a positioning method based on acoustic wave recognition, acquire noise data generated by the target UAV.

[0162] Among them, the acoustic wave recognition-based positioning method refers to the detection and positioning of the target UAV by utilizing the sound signals generated during its flight. The rotors, motors, and aerodynamic structures of the target UAV emit characteristic noises during operation. Noise data refers to audio signals containing sound source information of the target UAV, collected by microphone arrays or other acoustic sensors deployed on the ground or in the air. It can exist in the form of time-domain waveforms or frequency-domain spectrograms, including physical characteristics such as sound intensity, frequency components, arrival time, and phase difference.

[0163] In this embodiment, noise data from the target UAV can be passively received using a fixedly deployed ground-based microphone array. For example, when the target UAV enters the effective detection range, its rotor noise is synchronously captured by the microphone array, and noise data from each channel is recorded. Alternatively, a mobile or reconfigurable acoustic sensing platform can be used to dynamically collect noise data. In this case, the microphone array can be mounted on a mobile carrier (such as a vehicle, robot, or another UAV) to collect noise data in real time during movement.

[0164] S420: Determine the location information of the target UAV based on noise data.

[0165] In this embodiment, after the microphones synchronously record noise data, the time difference between the arrival of the target UAV's sound source at each microphone can be accurately calculated through cross-correlation analysis or a generalized cross-correlation algorithm. A virtual sphere is constructed with each microphone as the center and the speed of sound multiplied by the time difference as the radius. The intersection of multiple spheres represents the position information of the target UAV.

[0166] For example, noise data generated by the target drone can be collected by deploying 4-8 acoustic sensors, and the location information of the target drone can be generated by using the drone rotor noise characteristics to assist in blind zone detection.

[0167] S430. Based on the second signal-to-noise ratio associated with the target UAV, determine the positioning data quality attributes under the data quality index.

[0168] The second signal-to-noise ratio (SNR) refers to the ratio of the acoustic signal power received by the microphone array from the target UAV to the background noise power in the acoustic wave recognition-based positioning method, expressed in decibels (dB). Positioning data quality attributes quantify the quality of the current acoustic observation data and affect the accuracy of the positioning method.

[0169] In this embodiment, a continuous nonlinear function can be used to map the second signal-to-noise ratio (SNR) to a positioning data quality attribute. For example, the sigmoid function, exponential decay function, or logarithmic compression function can be used to convert the second SNR into a smooth attribute value in the interval [0,1], which serves as the positioning data quality attribute. When the second SNR is extremely low, the positioning data quality attribute approaches 0, indicating that the acoustic data is almost unusable; as the second SNR increases, the positioning data quality attribute rises rapidly and tends to saturate in the high SNR region.

[0170] Alternatively, the location data quality attributes can be determined based on formula (17). Formula (17) is expressed as: SNR1 / 20dB. SNR1 represents the second signal-to-noise ratio.

[0171] Formula (17) reflects the linear positive correlation between the acoustic signal-to-noise ratio (SNR) and the positioning data quality attribute. Specifically, when the second SNR1 is 20dB, the positioning data quality attribute reaches 1.0, indicating the best data quality; when SNR1 is 10dB, the positioning data quality attribute is 0.5, indicating medium quality; when SNR1 is 0dB, the positioning data quality attribute is 0, indicating that the signal and noise are of equal strength, and the data quality is extremely poor; if SNR1 is negative (signal power is lower than noise), the positioning data quality attribute is also negative, which can be truncated to 0 in practical applications and regarded as an invalid state.

[0172] S440. Based on wind speed in the environmental parameters, determine the environmental impact attributes under the environmental impact indicators.

[0173] Wind speed refers to the speed at which air flows per unit time, and can be measured in meters per second (m / s). Environmental impact attributes characterize the potential interference of current wind conditions on the positioning method's performance, thus reflecting the comprehensive impact of environmental factors on positioning accuracy and reliability.

[0174] In this embodiment, environmental impact attributes can be determined through wind speed classification mapping rules. For example, wind speed can be divided into several intervals according to meteorological wind force level standards (e.g., 0–2 m / s is "light breeze", corresponding to environmental impact attribute 1.0; 3–5 m / s is "light wind", corresponding to 0.8; 6–8 m / s is "strong wind", corresponding to 0.5; >8 m / s is "strong wind", corresponding to 0.2), and a corresponding environmental impact attribute value can be assigned to each interval. When the current wind speed is obtained, its corresponding interval is matched, and the corresponding environmental impact attribute is output. Alternatively, exponential decay, logarithmic compression, or piecewise linear functions can be used to convert the wind speed value into a smooth attribute value within the interval [0,1] as the environmental impact attribute. For example, when the wind speed is 0, the environmental impact attribute is 1 (indicating no wind interference). As the wind speed increases, the environmental impact attribute decreases monotonically according to a preset function, tending to saturate (close to 0) in the high wind speed region.

[0175] Alternatively, the environmental impact attribute under the environmental impact index can be determined based on formula (18). Formula (18) is expressed as: 1 - wind speed / 10 m / s. Set 10 m / s as the critical threshold for wind speed impact.

[0176] Formula (18) reflects the linear negative correlation between wind speed and environmental impact attributes. Specifically, when the wind speed is 0 m / s (no wind), the environmental impact attribute reaches its maximum value of 1, indicating that the environment has almost no interference with the positioning process; as the wind speed increases, the environmental impact attribute decreases linearly, indicating that the negative impact of wind on positioning performance gradually increases; when the wind speed reaches 10 m / s, the environmental impact attribute drops to 0, meaning that the wind force has become unacceptably strong; if the wind speed exceeds 10 m / s, the environmental impact attribute will become negative, which can be truncated to 0 in practical applications and regarded as a high-risk state.

[0177] S450. Based on the UAV voiceprint features detected by the positioning method in the fourth historical period before the current positioning time and at least one pre-stored voiceprint feature in the voiceprint database, determine the positioning reliability attribute under the historical positioning reliability index.

[0178] The fourth historical time period refers to a preset time window prior to the current positioning time, used for retrospective analysis of acoustic recognition performance during historical positioning processes. UAV voiceprint features are acoustic fingerprints extracted from target UAV noise data, possessing individual identifiability, including spectral envelope, harmonic structure, and modulation mode. The voiceprint database is a pre-stored database of acoustic features for known UAV models or individuals, used for voiceprint matching and recognition. Pre-stored voiceprint features are reference voiceprint templates stored in the voiceprint database. The positioning reliability attribute characterizes the recent acoustic recognition reliability level of this positioning method.

[0179] In this embodiment, the voiceprint matching results of the target UAV based on acoustic wave recognition can be continuously recorded during the fourth historical time period, and the ratio of the number of successful matches with pre-stored voiceprint features to the total number of detections is calculated. If the matching success rate is high (e.g., >80%), it indicates that the acoustic environment is stable and the recognition algorithm is effective, and a high positioning reliability attribute value (e.g., 0.9) is assigned; if the success rate is low (e.g., <50%), it is considered that the acoustic recognition reliability has decreased, and a low positioning reliability attribute value (e.g., 0.3) is assigned.

[0180] Alternatively, the stability of voiceprint recognition can be assessed by analyzing the mean, variance, and trend changes of the voiceprint matching results (such as cosine similarity, Euclidean distance, etc.) of the target UAV continuously recorded in the fourth historical period using sound wave recognition-based positioning methods. This yields the positioning reliability attribute. For example, if the similarity remains consistently high with minimal fluctuations, the positioning reliability attribute is high; conversely, if the similarity fluctuates drastically or continuously decreases, the positioning reliability attribute is correspondingly lower.

[0181] The technical solution in this embodiment constructs a multi-dimensional quality assessment framework for acoustic positioning, organically integrating the instantaneous quality of acoustic signals, the influence of environmental wind force, and historical recognition stability to form a comprehensive quantitative assessment of the reliability of acoustic wave recognition-based positioning methods in complex acoustic environments. By using a second signal-to-noise ratio to evaluate the detectability of acoustic signals in real time, wind speed parameters to dynamically reflect atmospheric propagation conditions, and historical statistical evaluation of voiceprint matching to assess the long-term stability of the recognition algorithm, the accuracy of the reliability assessment for acoustic recognition positioning is improved, enhancing the robustness and accuracy of UAV positioning in complex acoustic environments.

[0182] Figure 5 This is a structural schematic diagram of a positioning result determination device provided according to an embodiment of the present invention. Figure 5 As shown, the device includes: a location information determination module 510, a first confidence level determination module 520, a second confidence level determination module 530, and a target positioning result determination module 540.

[0183] The system includes a location information determination module 510, used to determine the location information of the target UAV based on at least two positioning methods; a first confidence level determination module 520, used to determine a first confidence level corresponding to each positioning method based on environmental parameters of the target UAV's current surrounding environment; a second confidence level determination module 530, used to determine the indicator attributes of multiple perception quality assessment indicators associated with the target UAV under each positioning method, and to determine a second confidence level corresponding to the positioning method based on multiple indicator attributes; wherein the multiple perception quality assessment indicators include at least data quality indicators, environmental impact indicators, and historical positioning reliability indicators; the indicator attributes include positioning data quality attributes corresponding to the data quality indicators, environmental impact attributes corresponding to the environmental impact indicators, and positioning reliability attributes corresponding to the historical positioning reliability indicators; and a target positioning result determination module 540, used to determine the target positioning result corresponding to the target UAV based on the first confidence level and the second confidence level corresponding to each positioning method, and all the location information.

[0184] The technical solution of this embodiment determines the location information of a target drone based on at least two positioning methods; obtains environmental parameters of the target drone's current surrounding environment, and determines a first confidence level corresponding to each positioning method based on the environmental parameters; for each positioning method, determines the indicator attributes of multiple perception quality assessment indicators associated with the target drone under the positioning method, and determines a second confidence level corresponding to the positioning method based on the multiple indicator attributes; wherein, the multiple perception quality assessment indicators include data quality indicators, environmental impact indicators, and historical positioning reliability indicators; based on the first and second confidence levels corresponding to each positioning method, and all location information, determines the target positioning result corresponding to the target drone, thus solving the problem in the prior art. The single-sensor-based detection method leads to poor accuracy in UAV positioning. This paper addresses this issue by fusing at least two positioning methods to obtain the target UAV's location information and dynamically evaluating the first confidence level of each positioning method in the current scenario based on environmental parameters. Furthermore, it introduces multiple perception quality assessment indicators associated with the target UAV (including data quality indicators, environmental impact indicators, and historical positioning reliability indicators), and determines the second confidence level of each positioning method by analyzing the attributes of these indicators. Finally, it combines the first confidence level, the second confidence level, and the location information provided by each positioning method, and weightedly fuses them to generate a high-precision, highly robust target positioning result, improving the accuracy and reliability of UAV positioning in complex and ever-changing environments.

[0185] Optionally, based on the above-described device, the location information determination module 510 includes:

[0186] A reflection point cloud data determination unit is used to scan the airspace containing the target UAV using a lidar device when the positioning method includes a lidar-based positioning method to obtain reflection point cloud data; the reflection point cloud data includes multiple point clouds with three-dimensional coordinates and reflection intensity; a point cloud cluster determination unit is used to spatially cluster the valid point clouds in the reflection point cloud data to obtain at least one point cloud cluster; wherein the valid point clouds satisfy preset validity conditions, the validity conditions include distance range constraints, reflection intensity threshold conditions, spatial neighborhood density conditions, and geographical altitude restrictions; a first position determination unit is used to determine the position information of the target UAV based on at least one of the point cloud clusters.

[0187] Based on the above-mentioned device, optionally, the environmental parameters include meteorological visibility; the second confidence level determination module 530 includes:

[0188] The first quality attribute determination unit is used to determine the proportion of invalid point clouds based on the number of invalid point clouds in the reflected point cloud data and the total number of point clouds in the reflected point cloud data when the positioning method includes a lidar-based positioning method, and to determine the positioning data quality attribute under the data quality index based on the proportion of invalid point clouds; the first impact attribute determination unit is used to determine the visibility attenuation factor based on the meteorological visibility in the environmental parameters, and to determine the environmental impact attribute under the environmental impact index based on the visibility attenuation factor; the first reliability attribute determination unit is used to determine the positioning reliability attribute under the historical positioning reliability index based on the historical accuracy attributes of positioning the target UAV using the positioning method before the current positioning time.

[0189] Based on the above-mentioned device, optionally, the environmental parameters include light intensity; the second confidence determination module 530 includes:

[0190] The second quality attribute determination unit is used to determine the third confidence level of the target UAV's location information based on the positioning method when the positioning method includes a vision-based positioning method, and to determine the positioning data quality attribute under the data quality index; the second impact attribute determination unit is used to determine the environmental impact attribute under the environmental impact index based on the light intensity in the environmental parameters; the second reliability attribute determination unit is used to determine the total number of frames used to continuously locate the target UAV based on the positioning method in the first historical time period before the current positioning time, and the number of successful frames that have continuously and successfully located the historical location information of the target UAV, and to determine the positioning reliability attribute under the historical positioning reliability index based on the number of successful frames and the total number of frames.

[0191] Based on the above-mentioned device, optionally, the environmental parameters include ambient temperature and the angle between the target luminescent object and the reference plane; the second confidence determination module 530 includes:

[0192] The third quality attribute determination unit is used to determine the target temperature difference based on the target UAV's current total heat generation power, effective heat dissipation area, and predetermined heat transfer coefficient when the positioning method includes an infrared imaging-based positioning method, and to determine the positioning data quality attribute under the data quality index based on the target temperature difference and the ambient temperature in the environmental parameters; the third influence attribute determination unit is used to determine the environmental influence attribute under the environmental influence index based on the angle between the target luminescent object and the reference plane in the environmental parameters; the third reliability attribute determination unit is used to determine the positioning reliability attribute under the historical positioning reliability index based on the matching degree between the hot spot morphology features detected by the positioning method and the hot spot template in the second historical period before the current positioning time.

[0193] Optionally, based on the above-described device, the location information determination module 510 includes:

[0194] The reflected signal parameter determination unit is used to obtain reflected signal parameters by transmitting radio waves to the airspace containing the target UAV through the base station when the positioning method includes a base station-based positioning method; the current distance determination unit is used to determine the current distance between the target UAV and the base station based on the reflected signal parameters and a preset speed of light; the second position determination unit is used to determine the position information of the target UAV based on the current distance and the current angle information of the target UAV relative to the base station.

[0195] Optionally, based on the above-described device, the second confidence level determination module 530 includes:

[0196] The fourth quality attribute determination unit is used to determine the positioning data quality attribute under the data quality index based on the first signal-to-noise ratio of the base station when the positioning method includes a base station-based positioning method; the fourth impact attribute determination unit is used to determine the proportion of multipath interference caused by multipath effect from the total signal energy of the target UAV sensed based on the base station, and to determine the environmental impact attribute under the environmental impact index based on the multipath interference proportion; the fourth reliability attribute determination unit is used to determine the number of clock anomalies when positioning the target UAV based on the positioning method in the third historical time period before the current positioning time, and to determine the positioning reliability attribute under the historical positioning reliability index based on the number of clock anomalies.

[0197] Optionally, based on the above-described device, the location information determination module 510 includes:

[0198] The noise data determination unit is used to acquire noise data generated by the target drone when the positioning method includes a positioning method based on sound wave recognition; the third position determination unit is used to determine the position information of the target drone based on the noise data.

[0199] Based on the above-mentioned device, optionally, the environmental parameters include wind speed; the second confidence determination module 530 includes:

[0200] The fifth quality attribute determination unit is used to determine the positioning data quality attribute under the data quality index based on the second signal-to-noise ratio associated with the target UAV when the positioning method includes a positioning method based on acoustic wave recognition; the fifth impact attribute determination unit is used to determine the environmental impact attribute under the environmental impact index based on the wind speed among the environmental parameters; the sixth reliability attribute determination unit is used to determine the positioning reliability attribute under the historical positioning reliability index based on the UAV voiceprint features detected based on the positioning method within the fourth historical time period before the current positioning time and at least one pre-stored voiceprint feature in the voiceprint database.

[0201] Optionally, based on the above-described apparatus, the apparatus may further include:

[0202] A positioning data receiving unit is used to receive positioning data reported by the target UAV; a positioning deviation data determining unit is used to perform consistency verification based on the target positioning result and the positioning data to obtain positioning deviation data; a warning prompt judging unit is used to determine whether to trigger a warning prompt based on a dynamic signal-to-noise ratio adjustment threshold and the positioning deviation data; wherein, the dynamic signal-to-noise ratio adjustment threshold is determined based on a second signal-to-noise ratio associated with the target UAV; and a warning prompt unit is used to generate a warning prompt message based on the positioning deviation data if the signal-to-noise ratio is triggered.

[0203] The positioning result determination device provided in the embodiments of the present invention can execute the positioning result determination method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the method execution.

[0204] Figure 6This is a schematic diagram of the structure of an electronic device implementing the positioning result determination method of the embodiments of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0205] like Figure 6 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory 12 or a random access memory 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the read-only memory 12 or a computer program loaded from storage unit 18 into the random access memory 13. The random access memory 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, read-only memory 12, and random access memory 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0206] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0207] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as the location result determination method.

[0208] In some embodiments, the location result determination method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via read-only memory 12 and / or communication unit 19. When the computer program is loaded into random access memory 13 and executed by processor 11, one or more steps of the location result determination method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the location result determination method by any other suitable means (e.g., by means of firmware).

[0209] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0210] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0211] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory, read-only memory, erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0212] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0213] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0214] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0215] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication unit 19, or installed from storage unit 18, or installed from read-only memory 12. When the computer program is executed by processor 11, it performs the functions defined in the methods of the embodiments of the present invention.

[0216] This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the positioning result determination method provided in any embodiment of this invention.

[0217] In implementing the computer program product, computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0218] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0219] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for determining positioning results, characterized in that, include: The location information of the target drone is determined based on at least two positioning methods; Based on the environmental parameters of the target UAV’s current surrounding environment, a first confidence level is determined for each of the positioning methods. For each of the aforementioned positioning methods, the indicator attributes of multiple perception quality assessment indicators associated with the target UAV under the aforementioned positioning method are determined. Based on the multiple indicator attributes, a second confidence level corresponding to the positioning method is determined. The multiple perception quality assessment indicators include at least data quality indicators, environmental impact indicators, and historical positioning reliability indicators. The indicator attributes include positioning data quality attributes corresponding to the data quality indicators, environmental impact attributes corresponding to the environmental impact indicators, and positioning reliability attributes corresponding to the historical positioning reliability indicators. Based on the first confidence level and the second confidence level corresponding to each of the positioning methods, and all the location information, the target positioning result corresponding to the target UAV is determined.

2. The method according to claim 1, characterized in that, The determination of the target UAV's location information based on at least two positioning methods includes: When the positioning method includes a lidar-based positioning method, the airspace containing the target UAV is scanned by a lidar device to obtain reflection point cloud data; the reflection point cloud data includes multiple point clouds with three-dimensional coordinates and reflection intensity. Spatial clustering is performed on the effective point clouds in the reflected point cloud data to obtain at least one point cloud cluster; wherein, the effective point clouds meet preset validity conditions, the validity conditions include distance range constraints, reflection intensity threshold conditions, spatial neighborhood density conditions, and geographical height restrictions. The location information of the target UAV is determined based on at least one of the point cloud clusters.

3. The method according to claim 1, characterized in that, The environmental parameters include meteorological visibility; the indicator attributes for determining the various perception quality assessment indicators associated with the target UAV under the positioning mode include: When the positioning method includes a lidar-based positioning method, the proportion of invalid point clouds is determined based on the number of invalid point clouds in the reflected point cloud data and the total number of point clouds in the reflected point cloud data. Based on the proportion of invalid point clouds, the positioning data quality attribute under the data quality index is determined. Based on the meteorological visibility in the environmental parameters, a visibility attenuation factor is determined, and based on the visibility attenuation factor, the environmental impact attribute under the environmental impact index is determined. Based on the historical accuracy attributes of the target UAV located using the positioning method prior to the current positioning time, the positioning reliability attribute under the historical positioning reliability index is determined.

4. The method according to claim 1, characterized in that, The environmental parameters include light intensity; the indicator attributes for determining the various perception quality assessment indicators associated with the target UAV under the positioning mode include: When the positioning method includes a vision-based positioning method, a third confidence level of the target UAV's location information is determined based on the positioning method, and the positioning data quality attribute under the data quality index is determined. Based on the light intensity in the environmental parameters, determine the environmental impact attributes under the environmental impact indicators; The total number of frames used to continuously locate the target UAV based on the positioning method during the first historical time period before the current positioning time is determined, as well as the number of successful frames that successfully located the historical location information of the target UAV. Based on the number of successful frames and the total number of frames, the positioning reliability attribute under the historical positioning reliability index is determined.

5. The method according to claim 1, characterized in that, The environmental parameters include ambient temperature and the angle between the target luminescent object and the reference plane; the indicator attributes for determining the various perception quality assessment indicators associated with the target UAV under the positioning method include: When the positioning method includes an infrared imaging-based positioning method, the target temperature difference is determined based on the target UAV's current total heat generation power, effective heat dissipation area, and a predetermined heat transfer coefficient. Based on the target temperature difference and the ambient temperature in the environmental parameters, the positioning data quality attribute under the data quality index is determined. Based on the angle between the target luminescent object and the reference plane in the environmental parameters, the environmental impact attributes under the environmental impact index are determined; Based on the matching degree between the hot spot morphology features detected by the positioning method and the hot spot template in the second historical period before the current positioning time, the positioning reliability attribute under the historical positioning reliability index is determined.

6. The method according to claim 1, characterized in that, The determination of the target UAV's location information based on at least two positioning methods includes: When the positioning method includes a base station-based positioning method, the base station transmits wireless waves to the airspace containing the target UAV and obtains the reflected signal parameters. Based on the reflected signal parameters and the preset speed of light, the current distance between the target drone and the base station is determined; Based on the current distance and the current angle information of the target drone relative to the base station, the position information of the target drone is determined.

7. The method according to claim 1, characterized in that, The indicator attributes for determining the multiple perception quality assessment indicators associated with the target UAV under the positioning mode include: When the positioning method includes a base station-based positioning method, the positioning data quality attribute under the data quality index is determined based on the first signal-to-noise ratio of the base station. From the total signal energy of the target UAV sensed by the base station, determine the proportion of multipath interference caused by multipath effect, and based on the multipath interference proportion, determine the environmental impact attribute under the environmental impact index. Determine the number of clock anomalies when locating the target UAV using the aforementioned positioning method within the third historical time period prior to the current positioning time, and determine the positioning reliability attribute under the aforementioned historical positioning reliability index based on the number of clock anomalies.

8. The method according to claim 1, characterized in that, The determination of the target UAV's location information based on at least two positioning methods includes: When the positioning method includes a positioning method based on sound wave recognition, noise data generated by the target UAV is acquired; Based on the noise data, the location information of the target UAV is determined.

9. The method according to claim 1, characterized in that, The environmental parameters include wind speed; the indicator attributes for determining the multiple perception quality assessment indicators associated with the target UAV under the positioning mode include: When the positioning method includes a positioning method based on acoustic wave recognition, the positioning data quality attribute under the data quality index is determined based on the second signal-to-noise ratio associated with the target UAV. Based on the wind speed among the environmental parameters, determine the environmental impact attributes under the environmental impact indicators; Based on the UAV voiceprint features detected using the positioning method during the fourth historical period prior to the current positioning time, and at least one pre-stored voiceprint feature in the voiceprint database, the positioning reliability attribute under the historical positioning reliability index is determined.

10. The method according to claim 1, characterized in that, After determining the target location result corresponding to the target UAV, the method further includes: Receive the positioning data reported by the target drone; Based on the target positioning result and the positioning data, a consistency check is performed to obtain positioning deviation data; Based on the signal-to-noise ratio (SNR) dynamic adjustment threshold and the positioning deviation data, it is determined whether to trigger an early warning prompt; wherein, the SNR dynamic adjustment threshold is determined based on a second SNR associated with the target UAV; If so, an early warning message will be generated based on the positioning deviation data.

Images