Monitoring unit matching method and system for complex low-altitude environment

Abstract

The invention provides a monitoring unit matching method and system for a complex low-altitude environment, and relates to the technical field of image processing. The system comprises an unmanned aerial vehicle, a sound wave monitoring unit, an image monitoring unit and a monitoring matching unit. On the basis of the image monitoring unit, the sound wave monitoring unit is arranged to make up for the deficiency of single image recognition; on the aspect of resource finiteness, an image monitoring unit and a sound wave monitoring unit are combined through the method steps, optimal distribution of monitoring resources is carried out, and therefore the optimal monitoring effect in limited resources is achieved. According to the invention, the technical problems of insufficient tracking, incomplete display and disturbing of a low-altitude environment when a light device is not triggered, the environment is dark and mutual or object shielding is faced by the unmanned aerial vehicle are solved, dynamic distribution and negative feedback control of monitoring resources are realized, an image positioning position and a sound wave positioning position can be fused according to actual conditions, and the positioning accuracy of the unmanned aerial vehicle is improved. And a reasonable comprehensive positioning position is obtained.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and particularly to a monitoring unit matching method and system for a complex low-altitude environment. Background Art

[0002] During the process of UAV array performance, it is necessary for us to always pay attention to the position of each UAV. On the one hand, it is through the positioning data fed back by the positioning module of the UAV itself. On the other hand, it also requires us to provide independent monitoring and positioning on the ground to correct abnormal positioning data or situations where the UAV does not fly according to the preset path in a timely manner.

[0003] However, in the face of untriggered lighting devices, dark environments, mutual or object occlusion, the monitoring unit based on image algorithms cannot effectively monitor and track each UAV in the UAV cluster. It is reflected that it cannot effectively track the UAV that is not lit in the dark environment, and cannot effectively identify and display the UAV behind that is blocked by the UAV in front. Moreover, the radar tracking system will generate electromagnetic interference to the complex low-altitude environment, thus triggering potential safety problems.

[0004] Therefore, it is necessary to provide a monitoring unit matching method and system for a complex low-altitude environment to solve the technical problems of insufficient tracking, incomplete display, and disruption of the low-altitude environment in the prior art when the UAV faces untriggered lighting devices, dark environments, mutual or object occlusion. Summary of the Invention

[0005] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a monitoring unit matching method and system for a complex low-altitude environment, aiming to solve the technical problems of insufficient tracking, incomplete display, and disruption of the low-altitude environment in the prior art when the UAV faces untriggered lighting devices, dark environments, mutual or object occlusion.

[0006] To achieve the above purpose, the present application proposes a monitoring unit matching system for a complex low-altitude environment, which is used in a UAV array and includes: UAVs, which are provided with acoustic wave generators; wherein, the acoustic wave generators are used to emit characteristic acoustic wave signals uniquely corresponding to the UAVs through low-frequency acoustic waves; An acoustic wave monitoring unit, which includes at least three acoustic wave collectors and an acoustic wave processing module arranged at different positions; wherein, the acoustic wave collectors are used to collect acoustic wave signals and screen out characteristic acoustic wave signals, and the acoustic wave processing module calculates the corresponding acoustic wave positioning positions according to the characteristic acoustic wave signals received by the acoustic wave collectors at different positions; The image monitoring unit includes an image collector and an image processing module; wherein, the image collector is used to collect UAV image data, and the image processing module performs UAV target recognition based on the UAV image data to obtain a UAV target image and acquire the corresponding image positioning position; The monitoring matching unit is used to perform matching management on the acoustic wave monitoring unit and the image monitoring unit, and perform positioning fusion to obtain a comprehensive positioning position; wherein, it includes: The image information acquisition module is used to acquire the clarity and integrity of each UAV in the UAV target image; The image resource matching module dynamically allocates image acquisition resources based on the clarity of each UAV; The acoustic wave resource matching module dynamically allocates acoustic wave processing resources based on the integrity of each UAV; The image monitoring control module dynamically controls the image acquisition area of each image collector according to the image acquisition resource situation; The acoustic wave monitoring control module dynamically controls the acoustic wave processing frequency of each characteristic acoustic wave signal in the acoustic wave processing module according to the acoustic wave processing resource situation; The comprehensive positioning fusion module performs weighted positioning fusion on the image positioning position and the acoustic wave positioning position through dynamic weights to obtain a comprehensive positioning position.

[0007] As a further solution, the monitoring matching unit also acquires the real-time motion parameters of the UAV through the IMU sensor of the UAV, and verifies the credibility of the comprehensive positioning position through the real-time motion parameters.

[0008] On the other hand, the present invention also provides a monitoring unit matching method for a complex low-altitude environment, which is applied to a monitoring unit matching system for a complex low-altitude environment as described in any one of the above, and includes the following steps: Step 1: Collect UAV image data through an image collector, and collect and screen characteristic acoustic wave signals through an acoustic wave collector; Step 2: The image processing module performs UAV target recognition based on the UAV image data to obtain a UAV target image and acquire the corresponding image positioning position; Step 3: The acoustic wave processing module calculates the corresponding acoustic wave positioning position according to the characteristic acoustic wave signals received by the acoustic wave collectors at different positions; Step 4: Acquire the clarity and integrity of each UAV in the UAV target image through the image information acquisition module; Step 5: Dynamically allocate image acquisition resources through the image resource matching module based on the clarity of each UAV; Step 6: Based on the integrity of each drone, the acoustic wave resource matching module dynamically allocates the acoustic wave processing resources; Step 7: According to the image acquisition resource situation, the image monitoring control module dynamically controls the image acquisition area of each image collector; Step 8: According to the acoustic wave processing resource situation, the acoustic wave monitoring control module dynamically controls the acoustic wave processing frequency of each characteristic acoustic wave signal in the acoustic wave processing module; Step 9: The integrated positioning fusion module sets dynamic weights, and performs weighted positioning fusion on the image positioning position and the acoustic wave positioning position and conducts credibility verification; among them, the integrated positioning position that passes the credibility verification is output; Step 10: Repeat Steps 1 to 9 to obtain the integrated positioning position after dynamic control until the matching positioning process ends.

[0009] As a further solution, in Step 2, the image processing module performs drone vision positioning based on ORB feature matching, and then based on the camera pitch angle and image parameters, obtains the image positioning position corresponding to the drone vision positioning.

[0010] As a further solution, in Step 3, the acoustic wave processing module performs spatial positioning on the characteristic acoustic wave signal through the Time Difference of Arrival (TDOA) algorithm, obtains the corresponding sound source position and uses it as the acoustic wave positioning position.

[0011] As a further solution, in Step 5, the image resource matching module obtains the image resource allocation weights of each drone according to 1 - [the clarity of the current drone / the total clarity of the drones], and calculates the image acquisition resources of each drone based on the image resource allocation weights of each drone.

[0012] As a further solution, in Step 6, the acoustic wave resource matching module obtains the acoustic wave resource allocation weights of each drone according to 1 - [the integrity of the current drone / the total integrity of the drones], and calculates the acoustic wave processing resources of each drone based on the acoustic wave resource allocation weights of each drone.

[0013] As a further solution, in Step 7, the image monitoring control module calculates the corresponding drone image acquisition area based on the clarity of the current drone and the drone image acquisition resources, and controls the image collector to make adjustments based on the drone image acquisition area; among them, the drone image acquisition resources = the clarity of the current drone * the drone image acquisition area.

[0014] As a further solution, in step 8, the acoustic wave monitoring and control module calculates the corresponding number of acoustic wave processes of the drone based on the acoustic wave processing frequency and the drone acoustic wave processing resources, and controls the acoustic wave processing module to process the characteristic acoustic wave signal the corresponding number of times based on the number of drone acoustic wave processes; wherein, the drone acoustic wave processing resources = the number of drone acoustic wave processes * the acoustic wave processing frequency.

[0015] As a further solution, in step 9, the weight term of the image positioning position is directly proportional to the drone acoustic wave resource allocation weight, and the weight term of the acoustic wave positioning position is directly proportional to the drone acoustic wave resource allocation weight.

[0016] Compared with the related art, the monitoring unit matching method and system for complex low-altitude environments provided by the present invention have the following advantages: The present invention makes up for the deficiency of single image recognition by setting an acoustic wave monitoring unit on the basis of the image monitoring unit; in terms of the finiteness of resources, the "image monitoring unit + acoustic wave monitoring unit" is combined and the best allocation of monitoring resources is carried out, so as to achieve the best monitoring effect within limited resources; the present invention solves the technical problems that when the drone faces the situation that the lighting device is not triggered, the environment is dark, there is mutual or object occlusion, there is insufficient tracking, incomplete display, and disruption of the low-altitude environment, realizes the dynamic allocation and negative feedback control of monitoring resources, and can fuse the image positioning position and the acoustic wave positioning position according to the actual situation to obtain a reasonable comprehensive positioning position. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings herein are incorporated into the specification and form a part of the specification, showing embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application.

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the following will briefly introduce the drawings required for use in the description of the embodiments or the related art. Obviously, for those of ordinary skill in the art, other drawings can also be obtained based on these drawings without creative efforts.

[0019] Figure 1 It is a schematic structural diagram of a monitoring unit matching system for complex low-altitude environments provided by the present invention; Figure 2 It is a schematic diagram of a drone array provided by the present invention Figure 1 ; Figure 3 It is a schematic diagram of a drone array provided by the present invention Figure 2 ; Figure 4 It is a schematic diagram of a drone array provided by the present invention Figure 3 ; Figure 5 Schematic diagram of the steps of a monitoring unit matching method for complex low-altitude environments provided by the present invention.

[0020] The realization of the purpose, functional features and advantages of this application will be further described with reference to the accompanying drawings in combination with the embodiments. Specific embodiments

[0021] To make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Usually, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0022] Embodiment 1 Please refer to Figure 1 , this application embodiment provides a monitoring unit matching method and system for complex low-altitude environments, aiming to solve the technical problem that the content after fixed optimization cannot meet the different viewing needs of users.

[0023] To achieve the above purpose, this application proposes a monitoring unit matching system for complex low-altitude environments, which is used in an unmanned aerial vehicle (UAV) array and includes: UAVs, equipped with acoustic wave generators; wherein, the acoustic wave generators are used to emit characteristic acoustic wave signals uniquely corresponding to the UAVs through low-frequency acoustic waves; Acoustic wave monitoring unit, including at least three acoustic wave collectors and an acoustic wave processing module arranged at different positions; wherein, the acoustic wave collectors are used to collect acoustic wave signals and screen out characteristic acoustic wave signals, and the acoustic wave processing module calculates the corresponding acoustic wave positioning positions according to the characteristic acoustic wave signals received by the acoustic wave collectors at different positions; Image monitoring unit, including an image collector and an image processing module; wherein, the image collector is used to collect UAV image data, and the image processing module performs UAV target recognition based on the UAV image data to obtain a UAV target image and obtain the corresponding image positioning position; Monitoring matching unit, used to perform matching management on the acoustic wave monitoring unit and the image monitoring unit, and perform positioning fusion to obtain a comprehensive positioning position; wherein, it includes: Image information acquisition module, used to acquire the clarity and integrity of each UAV in the UAV target image; Image resource matching module, dynamically allocates image acquisition resources based on the clarity of each UAV; Acoustic wave resource matching module, dynamically allocates acoustic wave processing resources based on the integrity of each UAV; The image monitoring and control module dynamically controls the image acquisition area of each image collector according to the image acquisition resource situation; The acoustic wave monitoring and control module dynamically controls the acoustic wave processing frequency of each characteristic acoustic wave signal in the acoustic wave processing module according to the acoustic wave processing resource situation; The integrated positioning and fusion module performs weighted positioning and fusion on the image positioning position and the acoustic wave positioning position through dynamic weights to obtain the integrated positioning position.

[0024] It should be noted that: The traditional third-party positioning method for unmanned aerial vehicles is to perform target tracking and positioning through lidar / electromagnetic radar; however, there are a large number of tracking targets in the unmanned aerial vehicle array. Therefore, lidar cannot handle the tracking of such a large number of targets, and electromagnetic radar will produce inevitable signal interference in the face of a complex low-altitude environment, and in severe cases, it may even cause the unmanned aerial vehicles to crash collectively, and the deployment cost is high and the algorithm is complex.

[0025] Therefore, we need to find a method that does not affect the complex low-altitude environment and can track unmanned aerial vehicles from a third party. At this time, we thought of using image algorithms to achieve unmanned aerial vehicle tracking and identification; however, when the lighting device is not triggered, the environment is dark, or there is mutual or object occlusion, the monitoring unit based on image algorithms cannot effectively monitor and track each unmanned aerial vehicle in the unmanned aerial vehicle cluster.

[0026] As Figure 2 shown in the unmanned aerial vehicle array, through this array, we can see that there is a front-back occlusion relationship between the unmanned aerial vehicles. At this time, the obtained unmanned aerial vehicle images are not complete. As Figure 3 shown, only some unmanned aerial vehicles are lit, and the unmanned aerial vehicles in the red part are not lit. As Figure 4 shown, the obtained unmanned aerial vehicle images are not clear and complete at this time; therefore, it is very difficult for us to accurately position through a pure vision-based positioning system.

[0027] Therefore, we further thought of setting up an acoustic wave monitoring unit on the basis of the image monitoring unit to make up for the deficiency of single image recognition. We use low-frequency sound waves that avoid the audible range of the human ear, allocate corresponding characteristic acoustic wave signals to each unmanned aerial vehicle within the low-frequency sound wave frequency range, and collect the characteristic acoustic wave signals through at least three acoustic wave collectors arranged at different positions, and then we can calculate the spatial position of the unmanned aerial vehicle target based on the time difference of arrival algorithm.

[0028] However, simply combining the "image monitoring unit + acoustic wave monitoring unit" cannot truly solve the monitoring problem of the drone array. This is mainly reflected in the limitation of resources. That is, when monitoring a large number of drones simultaneously, how to optimally allocate monitoring resources on the basis of combining the "image monitoring unit + acoustic wave monitoring unit" to achieve the best monitoring effect within limited resources is the core technical problem to be solved by the present invention.

[0029] The following further explains the limitation of resources: In the actual image acquisition process, the image acquisition resources are limited. We can simplify it into the following equation: Image acquisition resources = Image acquisition area * Image acquisition detail. Without adding new image acquisition devices, the image acquisition resources are constant. That is, when the image acquisition area increases, the image acquisition detail will decrease, and when the image acquisition detail increases.

[0030] Similarly, when processing acoustic wave signals, the acoustic wave processing resources are limited. We can simplify it into the following equation: Acoustic wave processing resources = Number of acoustic wave processing * Acoustic wave processing frequency. Since the acoustic wave processing resources of the acoustic wave processing module are constant, when the acoustic wave processing frequency is constant, the allocated acoustic wave processing resources determine the number of acoustic wave processing.

[0031] Based on the above analysis, when actually monitoring the drone array, it is necessary to dynamically adjust the resource allocation to achieve the best combined monitoring effect.

[0032] In addition, the monitoring matching unit also obtains the real-time motion parameters of the drone through the IMU sensor of the drone, and verifies the credibility of the comprehensive positioning position through the real-time motion parameters. For example: The real-time motion parameters of the drone can be substituted to determine whether the change of the position coordinates conforms to the representation of the real-time motion parameters.

[0033] Embodiment 2 Please refer to Figure 5 , the present invention also provides a monitoring unit matching method for a complex low-altitude environment, which is applied to a monitoring unit matching system for a complex low-altitude environment as described in Embodiment 1, and includes the following steps: Step 1: Collect drone image data through an image collector, and collect and screen characteristic acoustic wave signals through an acoustic wave collector; Step 2: The image processing module performs drone target recognition based on the drone image data, obtains the drone target image and the corresponding image positioning position; Step 3: The acoustic wave processing module calculates the corresponding acoustic wave positioning position according to the characteristic acoustic wave signals received by the acoustic wave collectors at different positions; Step 4: Obtain the clarity and integrity of each drone in the drone target image through the image information acquisition module; Step 5: Dynamically allocate image acquisition resources based on the clarity of each drone through the image resource matching module; Step 6: Dynamically allocate acoustic wave processing resources based on the integrity of each drone through the acoustic wave resource matching module; Step 7: Dynamically control the image acquisition area of each image collector according to the image acquisition resource situation through the image monitoring control module; Step 8: Dynamically control the acoustic wave processing frequency of each characteristic acoustic wave signal in the acoustic wave processing module according to the acoustic wave processing resource situation through the acoustic wave monitoring control module; Step 9: Set dynamic weights through the integrated positioning fusion module, perform weighted positioning fusion on the image positioning position and the acoustic wave positioning position, and perform credibility verification; among them, output the integrated positioning position that passes the credibility verification; Step 10: Repeat Steps 1 to 9 to obtain the integrated positioning position after dynamic control until the matching positioning process ends.

[0034] It should be noted that: when performing image positioning, inaccurate image positioning is mainly due to unclear and incomplete images. At this time, acoustic wave positioning needs to be mobilized to make up for this part, and the resources of both are limited. Therefore, we need to perform reasonable scheduling of resource matching.

[0035] Image positioning needs to consider the impact of the image acquisition area on image clarity. For example, when we magnify an image of a certain place, the area of the collected image will shrink, and the image will be clearer than before. These two are in an inverse relationship; therefore, for drones with clear images that can be accurately positioned, there is no need to magnify the image. Instead, a larger drone image acquisition area can be maintained, so as to collect as many drone targets as possible in the same image, thereby saving image acquisition resources.

[0036] For unclear images, we need to measure the total image acquisition resources, allocate image acquisition resources proportionally within the limited resources, and within the allocated drone image acquisition resources, magnify the image area by no more than the allocated limit. In this way, we can ensure the balance of resource allocation and prevent the overall resource allocation from being unbalanced due to magnifying the image.

[0037] Specifically, the acoustic wave resource matching module obtains the acoustic wave resource allocation weights of each drone according to 1 - [the integrity of the current drone / the total integrity of the drones], and calculates the acoustic wave processing resources of each drone based on the acoustic wave resource allocation weights of each drone.

[0038] Furthermore, if the image is incomplete, it cannot be solved by magnification. At this time, we need to dispatch acoustic positioning to make up for it. However, acoustic positioning is restricted by acoustic processing resources (such as FPGA computing resources). At this time, it also requires us to allocate resources preferentially. For example, if the drone image is complete, we do not need to use acoustic positioning. If the drone image is incomplete, it will be allocated according to the degree of incompleteness. The more incomplete it is, the more acoustic processing resources will be allocated to it.

[0039] Specifically, the acoustic monitoring and control module calculates the corresponding number of drone acoustic processing based on the acoustic processing frequency and the drone acoustic processing resources, and controls the acoustic processing module to process the characteristic acoustic signal for the corresponding number of times based on the number of drone acoustic processing. Among them, the drone acoustic processing resources = the number of drone acoustic processing * the acoustic processing frequency.

[0040] The advantage of doing this is that it can increase the number of drone acoustic processing for drones with a higher degree of incompleteness, so as to improve the accuracy and granularity of acoustic positioning (the higher the number of processing, the finer the granularity), and then make up for the missing part of image positioning when it is incomplete.

[0041] Furthermore, in step 9, the weight term of the image positioning position is directly proportional to the weight of the drone acoustic resource allocation, and the weight term of the acoustic positioning position is directly proportional to the weight of the drone acoustic resource allocation.

[0042] It can be understood that: the clearer the image, the more accurate the image positioning position, and the higher the weight. The higher the degree of image incompleteness, the less important whether the image is clear (it cannot be overcome). Therefore, the proportion of the acoustic positioning position is adjusted preferentially through the weight of the drone acoustic resource allocation, so as to obtain a reasonable comprehensive positioning position.

[0043] In addition, in step 2, the image processing module performs drone visual positioning based on ORB feature matching, and then obtains the image positioning position corresponding to the drone visual positioning based on the camera pitch angle and image parameters. ORB (Oriented FAST and Rotated BRIEF) is an efficient algorithm for image feature detection and matching, especially suitable for scenarios with high real-time requirements (such as drone visual positioning). The following is the core principle of ORB feature matching: 1. FAST key point detection Fast corner detection: By analyzing the brightness differences of 16 neighboring points around a pixel, quickly locate the corner points (Corner Points) in the image. If the brightness of N consecutive points (usually N = 9) differs from the center point by more than the threshold, it is determined as a corner point. The calculation speed is more than 10 times faster than traditional SIFT, meeting the real-time processing requirements of drones.

[0044] 2. BRIEF Descriptor Binary Feature Description: Generate a binary description vector through 256 groups of predefined pixel point pairs ((x1, y1), (x2, y2)); Compare the brightness of each pair of points: If I(x1, y1) > I(x2, y2), the corresponding bit is 1, otherwise it is 0; Generate a 256-bit binary code (such as 1011...001) for fast matching.

[0045] 3. Improvement of Direction Invariance Direction Calculation: Calculate the main direction of the key point through the centroid of the image block (IntensityCentroid); Rotation Correction: Rotate the BRIEF sampling pattern according to the main direction θ to make the descriptor rotation invariant.

[0046] 4. Feature Matching Hamming Distance: Calculate the difference between two feature descriptors through the XOR operation (XOR); The smaller the distance, the higher the matching similarity (for example, the Hamming distance between 1010 and 1001 is 2).

[0047] Fast Screening: Use the kNN (k = 2) algorithm to screen the best matching pairs and eliminate incorrect matches.

[0048] In step 3, the acoustic wave processing module performs spatial positioning on the characteristic acoustic wave signal through the Time Difference of Arrival (TDOA) algorithm to obtain the corresponding sound source position and use it as the acoustic wave positioning position. The core method of sound source positioning through three acoustic wave collectors (microphones) and time difference is the Time Difference of Arrival (TDOA) positioning method, and its implementation process is as follows: 1. Time Difference Measurement Three acoustic wave collectors (set as , , ) record the arrival time of the acoustic wave , , Calculate the time difference between each pair: , ; 2. Distance Difference Calculation According to the speed of sound c (about 343 m / s at room temperature), convert the time difference into a distance difference: ; where i and j are the numbers of two relatively positioned acoustic wave collectors. For example: ; where S is the sound source position. 3. Solving the hyperbola equation Each distance difference corresponds to a hyperbola trajectory. The following equations can be established with three groups of sensors (taking two-dimensional positioning as an example): Solve the intersection point of the equations by numerical methods (such as Newton iteration method) to determine the sound source position .

[0049] The above are only some embodiments of the present application, and do not limit the patent scope of the present application accordingly. Any equivalent structural transformation made under the technical concept of the present application by using the content of the specification and drawings of the present application, or any direct / indirect application in other related technical fields is included in the patent protection scope of the present application.

Claims

1. A monitoring unit matching system for complex low-altitude environments, which is used in an unmanned aerial vehicle array, and is characterized in that Comprising: A drone, provided with an acoustic wave generator; wherein, the acoustic wave generator is used to emit a characteristic acoustic wave signal unique to the drone through low-frequency acoustic waves; An acoustic wave monitoring unit, including at least three acoustic wave collectors and an acoustic wave processing module arranged at different positions; wherein, the acoustic wave collectors are used to collect acoustic wave signals and screen out the characteristic acoustic wave signals, and the acoustic wave processing module calculates the corresponding acoustic wave positioning positions according to the characteristic acoustic wave signals received by the acoustic wave collectors at different positions; An image monitoring unit, including an image collector and an image processing module; wherein, the image collector is used to collect drone image data, and the image processing module performs drone target recognition based on the drone image data to obtain a drone target image and acquire the corresponding image positioning position; A monitoring matching unit, used to perform matching management on the acoustic wave monitoring unit and the image monitoring unit, and perform positioning fusion to obtain a comprehensive positioning position; wherein, it includes: An image information acquisition module, used to acquire the clarity and integrity of each drone in the drone target image; An image resource matching module, based on the clarity of each drone, dynamically allocates the image acquisition resources; An acoustic wave resource matching module, based on the integrity of each drone, dynamically allocates the acoustic wave processing resources; An image monitoring control module, according to the image acquisition resource situation, dynamically controls the image acquisition area of each image collector; An acoustic wave monitoring control module, according to the acoustic wave processing resource situation, dynamically controls the acoustic wave processing frequency of each characteristic acoustic wave signal in the acoustic wave processing module; A comprehensive positioning fusion module, through dynamic weights, performs weighted positioning fusion on the image positioning position and the acoustic wave positioning position to obtain a comprehensive positioning position.

2. The monitoring unit matching system for complex low-altitude environment according to claim 1, wherein The monitoring matching unit also acquires the real-time motion parameters of the drone through the IMU sensor of the drone, and verifies the credibility of the comprehensive positioning position through the real-time motion parameters.

3. A monitoring unit matching method for complex low-altitude environments, which is applied to a monitoring unit matching system for complex low-altitude environments as described in any one of claims 1 to 2, characterized in that, Including the following steps: Step 1: Collect drone image data through the image collector, and collect and screen out the characteristic acoustic wave signals through the acoustic wave collector; Step 2: The image processing module performs drone target recognition based on the drone image data to obtain a drone target image and acquire the corresponding image positioning position; Step 3: The acoustic wave processing module calculates the corresponding acoustic wave positioning positions according to the characteristic acoustic wave signals received by the acoustic wave collectors at different positions; Step 4: Acquire the clarity and integrity of each drone in the drone target image through the image information acquisition module; Step 5: Through the image resource matching module, based on the clarity of each drone, dynamically allocate the image acquisition resources; Step 6: Through the acoustic wave resource matching module, based on the integrity of each drone, dynamically allocate the acoustic wave processing resources; Step 7: Through the image monitoring control module, according to the image acquisition resource situation, dynamically control the image acquisition area of each image collector; Step 8: Through the acoustic wave monitoring control module, according to the acoustic wave processing resource situation, dynamically control the acoustic wave processing frequency of each characteristic acoustic wave signal in the acoustic wave processing module; Step 9: Set dynamic weights through the comprehensive positioning fusion module, perform weighted positioning fusion on the image positioning position and the acoustic wave positioning position, and conduct credibility verification; among them, output the comprehensive positioning position that passes the credibility verification. Step 10: Repeat Steps 1 to 9 to obtain the comprehensive positioning position after dynamic control until the matching positioning process ends.

4. A method for matching monitoring units for a complex low-altitude environment according to claim 3, characterized in that In Step 2, the image processing module performs UAV vision positioning based on ORB feature matching, and then obtains the image positioning position corresponding to the UAV vision positioning based on the camera pitch angle and image parameters.

5. A method for matching a monitoring unit for a complex low-altitude environment according to claim 3, characterized in that, In Step 3, the acoustic wave processing module performs spatial positioning on the characteristic acoustic wave signal through the Time Difference of Arrival (TDOA) algorithm to obtain the corresponding sound source position and use it as the acoustic wave positioning position.

6. The matching method of a monitoring unit for a complex low-altitude environment according to claim 3, wherein In Step 5, the image resource matching module obtains the allocation weights of each UAV image resource according to 1 - [the clarity of the current UAV / the total clarity of all UAVs], and calculates the image acquisition resources of each UAV based on the allocation weights of each UAV image resource.

7. A method for matching monitoring units for a complex low-altitude environment according to claim 3, characterized in that, In Step 6, the acoustic wave resource matching module obtains the allocation weights of each UAV acoustic wave resource according to 1 - [the integrity of the current UAV / the total integrity of all UAVs], and calculates the acoustic wave processing resources of each UAV based on the allocation weights of each UAV acoustic wave resource.

8. A method for matching a monitoring unit for a complex low-altitude environment according to claim 6, characterized in that, In Step 7, the image monitoring and control module calculates the corresponding UAV image acquisition area based on the clarity of the current UAV and the UAV image acquisition resources, and controls the image collector to make adjustments based on the UAV image acquisition area; where the UAV image acquisition resources = the clarity of the current UAV * the UAV image acquisition area.

9. A method for matching a monitoring unit for a complex low-altitude environment according to claim 7, characterized in that, In Step 8, the acoustic wave monitoring and control module calculates the corresponding number of UAV acoustic wave processes based on the acoustic wave processing frequency and the UAV acoustic wave processing resources, and controls the acoustic wave processing module to process the characteristic acoustic wave signal the corresponding number of times based on the number of UAV acoustic wave processes; where the UAV acoustic wave processing resources = the number of UAV acoustic wave processes * the acoustic wave processing frequency.

10. A method for matching monitoring units for a complex low-altitude environment according to any one of claims 8 or 9, characterized in that, In Step 9, the weight term of the image positioning position is directly proportional to the allocation weight of the UAV acoustic wave resource, and the weight term of the acoustic wave positioning position is directly proportional to the allocation weight of the UAV acoustic wave resource.

Images