Unmanned aerial vehicle cooperative reconnaissance positioning method and device and storage medium

Abstract

The invention provides an unmanned aerial vehicle cooperative reconnaissance positioning method and device and a storage medium, and belongs to the technical field of reconnaissance positioning, and the method comprises the steps: obtaining an unmanned aerial vehicle host coordinate position and original unmanned aerial vehicle host business data from unmanned aerial vehicle hosts in an unmanned aerial vehicle cluster, obtaining unmanned aerial vehicle slave coordinate positions and original unmanned aerial vehicle slave business data from a plurality of unmanned aerial vehicle slaves in the unmanned aerial vehicle cluster; and the original unmanned aerial vehicle host business data and the original unmanned aerial vehicle slave business data are preprocessed to obtain preprocessed unmanned aerial vehicle host business data and preprocessed unmanned aerial vehicle slave business data. According to the invention, accurate control of task load business data volume is realized, real-time transmission of key data is ensured, the problem of load overrun of ad hoc network communication radio data transmission in multi-unmanned aerial vehicle cooperative reconnaissance is solved, and stability of cooperative reconnaissance tasks and intelligence accuracy are improved.

Description

Technical Field

[0001] This invention mainly relates to the field of reconnaissance and positioning technology, specifically to a method, device, and storage medium for collaborative reconnaissance and positioning of unmanned aerial vehicles (UAVs). Background Technology

[0002] In multi-UAV collaborative reconnaissance missions, ad hoc network communication radios serve as the core carrier for information exchange between nodes. With escalating mission requirements, the size and number of nodes in UAV swarms are growing exponentially. The payload data carried by a single UAV includes PDW, EDW, UAV position and attitude data, and payload status data, leading to a significant increase in the rate of generation of multi-source heterogeneous data. Concurrent transmission by multiple nodes will cause a sharp rise in the total network load, far exceeding the maximum effective transmission bandwidth of the ad hoc network radio. This will directly cause network congestion, transmission delays, a surge in packet loss, and even serious consequences such as errors in reconnaissance intelligence and interruptions in collaborative missions.

[0003] In traditional UAV collaborative reconnaissance and localization, pulse pairing and time difference pairing rely on histogram methods or correlation methods. Histogram methods determine the dominant time difference by statistically analyzing the cumulative frequency of pulses within different time difference intervals and identifying the peak position. Correlation methods, based on the similarity characteristics of signal waveforms, find the maximum matching point by calculating the delay correlation function. However, both methods face complex electromagnetic scenarios such as dense pulse streams, multipath reflections, or overlapping targets. Interference signals can cause pulse overlap distortion, time difference blurring, or spurious correlation peaks, making it difficult for traditional histogram or correlation methods to accurately distinguish between the real signal and interference components. Especially in scenarios with multiple signals present simultaneously, interference signals may exhibit similar time difference statistical or correlation characteristics to the target signal, further exacerbating the risk of mispairing and missed pairing, severely restricting the reliability and stability of multi-UAV collaborative time difference localization systems. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method, device and storage medium for collaborative reconnaissance and positioning of unmanned aerial vehicles (UAVs) to address the shortcomings of the prior art.

[0005] The technical solution of this invention to solve the above-mentioned technical problems is as follows: A method for cooperative reconnaissance and positioning of unmanned aerial vehicles (UAVs), comprising the following steps: The coordinates and original business data of the drone host are obtained from the drone host in the drone cluster, and the coordinates and original business data of the drone slave corresponding to each drone slave are obtained from the multiple drone slaves in the drone cluster. The original UAV host service data and the original UAV slave service data corresponding to each of the UAV slaves are preprocessed to obtain the preprocessed UAV host service data corresponding to the UAV host and the preprocessed UAV slave service data corresponding to each of the UAV slaves. Time difference parameter analysis was performed on the preprocessed UAV host business data and the preprocessed UAV slave business data corresponding to each UAV slave to obtain the time difference parameter corresponding to each UAV slave. A time-difference positioning model is constructed, and the coordinate positions of the UAV host, all UAV slaves, and all time-difference parameters are analyzed using the time-difference positioning model to obtain the UAV collaborative reconnaissance positioning results.

[0006] Another technical solution of the present invention to solve the above-mentioned technical problems is as follows: A UAV cooperative reconnaissance and positioning device, comprising: The data acquisition module is used to obtain the coordinate position of the drone host and the original drone host business data from the drone host in the drone cluster, and to obtain the coordinate position of the drone slave corresponding to each drone slave and the original drone slave business data corresponding to each drone slave from multiple drone slaves in the drone cluster. The preprocessing module is used to preprocess the original UAV host business data and the original UAV slave business data corresponding to each of the UAV slaves, respectively, to obtain the preprocessed UAV host business data corresponding to the UAV host and the preprocessed UAV slave business data corresponding to each of the UAV slaves. The time difference parameter analysis module is used to perform time difference parameter analysis on the preprocessed UAV host business data and the preprocessed UAV slave business data corresponding to each of the UAV slaves, and obtain the time difference parameters corresponding to each of the UAV slaves. The positioning result acquisition module is used to construct a time difference positioning model. Through the time difference positioning model, the positioning analysis is performed on the coordinate position of the UAV host, the coordinate positions of all UAV slaves, and all the time difference parameters to obtain the UAV collaborative reconnaissance positioning result.

[0007] Based on the above-mentioned UAV collaborative reconnaissance and positioning method, the present invention also provides a UAV collaborative reconnaissance and positioning system.

[0008] Another technical solution of the present invention to solve the above-mentioned technical problems is as follows: a UAV cooperative reconnaissance and positioning system, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the UAV cooperative reconnaissance and positioning method described above is implemented.

[0009] Based on the above-mentioned UAV collaborative reconnaissance and positioning method, the present invention also provides a computer-readable storage medium.

[0010] Another technical solution of the present invention to solve the above-mentioned technical problems is as follows: a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the UAV collaborative reconnaissance and positioning method as described above.

[0011] The beneficial effects of this invention are as follows: By preprocessing the original UAV host business data and the original UAV slave business data, preprocessed UAV host business data and preprocessed UAV slave business data are obtained. Time difference parameters are obtained through time difference parameter analysis of the preprocessed UAV host business data and the preprocessed UAV slave business data. The collaborative reconnaissance positioning results of the UAVs are obtained through the positioning analysis of the coordinate positions of the UAV host, the UAV slave, and the time difference parameters using a time difference positioning model. This achieves precise control of the workload of mission payload business data, ensures real-time transmission of key data, solves the problem of excessive data transmission load of ad hoc network communication radios in multi-UAV collaborative reconnaissance, improves the stability and intelligence accuracy of collaborative reconnaissance missions, effectively solves the problems of weak anti-interference capability and difficulty in multi-target discrimination of traditional methods in complex electromagnetic environments, improves the accuracy and robustness of pulse pairing and time difference pairing in complex scenarios, ensures the reliability of the time difference positioning system, and can meet the reconnaissance and positioning needs of large UAV swarms, thus having great application value. Attached Figure Description

[0012] Figure 1 This is a flowchart illustrating the UAV collaborative reconnaissance and positioning method provided in an embodiment of the present invention. Figure 2 This is a schematic diagram showing the layout of the drone swarm location in the drone cooperative reconnaissance and positioning method provided in an embodiment of the present invention. Figure 3 This is a block diagram of a UAV collaborative reconnaissance and positioning device provided in an embodiment of the present invention. Detailed Implementation

[0013] The principles and features of the present invention are described below with reference to the accompanying drawings. The examples given are only for explaining the present invention and are not intended to limit the scope of the present invention.

[0014] Figure 1 This is a flowchart illustrating a UAV collaborative reconnaissance and positioning method provided in an embodiment of the present invention.

[0015] like Figure 1 As shown, a method for cooperative reconnaissance and positioning using unmanned aerial vehicles (UAVs) includes the following steps: S1: Obtain the coordinates and original business data of the drone host from the drone host in the drone cluster, and obtain the coordinates and original business data of the drone slave corresponding to each drone slave from the multiple drone slaves in the drone cluster. S2: Preprocess the original UAV host service data and the original UAV slave service data corresponding to each of the UAV slaves respectively to obtain the preprocessed UAV host service data corresponding to the UAV host and the preprocessed UAV slave service data corresponding to each of the UAV slaves. S3: Perform time difference parameter analysis on the preprocessed UAV host business data and the preprocessed UAV slave business data corresponding to each UAV slave to obtain the time difference parameters corresponding to each UAV slave. S4: Construct a time difference positioning model, and use the time difference positioning model to perform positioning analysis on the coordinate position of the UAV host, the coordinate positions of all UAV slaves, and all the time difference parameters to obtain the UAV collaborative reconnaissance positioning results.

[0016] It should be understood that, according to the requirements of the UAV swarm collaborative reconnaissance and positioning system, the flight grid area and flight path of each UAV are determined, the UAV swarm is decomposed into multiple UAV sub-swarms, each sub-swarm has ≤5 UAVs, one of which is designated as the master (i.e., UAV master) and the other UAVs are slaves (i.e., UAV slaves). Target signals within the reconnaissance area are reconnoitered through frequency domain collaboration or air domain collaboration.

[0017] In the above embodiments, preprocessing of the original UAV host business data and the original UAV slave business data yields preprocessed UAV host business data and preprocessed UAV slave business data. Time difference parameters are obtained through time difference parameter analysis of the preprocessed UAV host business data and preprocessed UAV slave business data. The collaborative reconnaissance positioning results are obtained through positioning analysis of the UAV host coordinate position, UAV slave coordinate position, and time difference parameters using a time difference positioning model. This achieves precise control of the mission payload business data volume, ensures real-time transmission of key data, solves the problem of excessive data transmission load of ad hoc network communication radios in multi-UAV collaborative reconnaissance, improves the stability and intelligence accuracy of collaborative reconnaissance missions, effectively solves the problems of weak anti-interference capability and difficulty in multi-target discrimination of traditional methods in complex electromagnetic environments, improves the accuracy and robustness of pulse pairing and time difference pairing in complex scenarios, ensures the reliability of the time difference positioning system, and can meet the reconnaissance and positioning needs of large UAV swarms, demonstrating significant application value.

[0018] Optionally, as an embodiment of the present invention, the original UAV host service data includes original PDW pulse descriptor host data, original EDW event descriptor host data, original UAV host self-parameters, and original payload status host data; the original UAV slave service data includes original PDW pulse descriptor slave data, original EDW event descriptor slave data, original UAV slave self-parameters, and original payload status slave data. The process of preprocessing the original UAV host service data and the original UAV slave service data corresponding to each of the UAV slaves to obtain the preprocessed UAV host service data corresponding to the UAV host and the preprocessed UAV slave service data corresponding to each of the UAV slaves includes: PDW preprocessing is performed on the original PDW pulse descriptor host data and the original PDW pulse descriptor slave data corresponding to each of the UAV slaves to obtain the preprocessed PDW pulse descriptor host data corresponding to the UAV host and the preprocessed PDW pulse descriptor slave data corresponding to each of the UAV slaves. Import pulse signals, and sort the preprocessed PDW pulse descriptor host data and the preprocessed PDW pulse descriptor slave data corresponding to each UAV slave according to the pulse signals to obtain the PDW host cluster list corresponding to the UAV host and the PDW slave cluster list corresponding to each UAV slave. Sort all PDW pulse descriptor host data in the PDW host clustering list according to the signal arrival time, and take the first N PDW pulse descriptor host data as the target PDW pulse descriptor host data, thereby obtaining the target PDW pulse descriptor host dataset; Sort all PWD pulse descriptor slave data in each of the PDW slave cluster lists according to the signal arrival time, and take the first N PWD pulse descriptor slave data as the target PWD pulse descriptor slave data, thereby obtaining the target PWD pulse descriptor slave dataset corresponding to each of the UAV slaves; The target PDW pulse descriptor host dataset, the original EDW event descriptor host data, the original UAV host self-parameters, and the original payload status host data are processed by a pre-built data classification model to obtain classified PDW pulse descriptor host dataset, classified EDW event descriptor host data, classified UAV host self-parameters, and classified payload status host data. By using a pre-built data grading model, the target PDW pulse descriptor slave dataset, the original EDW event descriptor slave data corresponding to each UAV slave, the original UAV slave self parameters corresponding to each UAV slave, and the original load status slave data corresponding to each UAV slave are processed to obtain the graded PDW pulse descriptor slave dataset, the graded EDW event descriptor slave data, the graded UAV slave self parameters, and the graded load status slave data corresponding to each UAV slave; The host data acquisition quality and host network load are obtained from the host drone, and the slave data acquisition quality and slave network load corresponding to each of the slave drones are obtained from each of the slave drones respectively. Based on the host data acquisition quality, the transmission priority of the graded PDW pulse descriptor host dataset, the graded EDW event descriptor host data, the graded UAV host self-parameters, and the graded payload status host data are optimized to obtain the optimized PDW pulse descriptor host dataset, the optimized EDW event descriptor host data, the optimized UAV host self-parameters, and the optimized payload status host data. Based on the data acquisition quality of each slave device, the transmission priority of each tiered PDW pulse descriptor slave dataset, the tiered EDW event descriptor slave data corresponding to each UAV slave device, the tiered UAV slave device self-parameters corresponding to each UAV slave device, and the tiered load status slave data corresponding to each UAV slave device are optimized to obtain the optimized PDW pulse descriptor slave dataset, the optimized EDW event descriptor slave data, the optimized UAV slave device self-parameters, and the optimized load status slave data corresponding to each UAV slave device. Based on the host network load, bandwidth ratio allocation is performed on the optimized PDW pulse descriptor host dataset, the optimized EDW event descriptor host data, the optimized UAV host self-parameters, and the optimized load status host data to obtain the allocated PDW pulse descriptor host dataset, the allocated EDW event descriptor host data, the allocated UAV host self-parameters, and the allocated load status host data. Based on the network load of each slave device, bandwidth ratio allocation processing is performed on each optimized PDW pulse descriptor slave dataset, the optimized EDW event descriptor slave data corresponding to each UAV slave, the optimized UAV slave self parameters corresponding to each UAV slave, and the optimized load status slave data corresponding to each UAV slave. This results in the allocated PDW pulse descriptor slave dataset, the allocated EDW event descriptor slave data, the allocated UAV slave self parameters, and the allocated load status slave data corresponding to each UAV slave. The preprocessed PDW pulse descriptor host data includes the allocated PDW pulse descriptor host dataset, the allocated EDW event descriptor host data, the allocated UAV host self-parameters, and the allocated payload status host data. The preprocessed UAV slave service data includes the allocated PDW pulse descriptor slave dataset, the allocated EDW event descriptor slave data, the allocated UAV slave self-parameters, and the allocated payload status slave data.

[0019] As should be understood, as shown in Table 1, when a single UAV node's payload is powered on, it generates raw business data including PDW, EDW, UAV position and attitude data, and payload status data (i.e., raw PDW pulse descriptor host data, raw EDW event descriptor host data, raw UAV host parameters, raw payload status host data, raw PDW pulse descriptor slave data, raw EDW event descriptor slave data, raw UAV slave parameters, and raw payload status slave data), which are stored in the storage module of the payload device. Table 1 shows the design structure of the PDW data generated by the payload.

[0020] Table 1 It should be understood that by utilizing a multi-dimensional perception adaptive data transmission optimization method, the amount of redundant raw data generated by the task payload is reduced through the fusion preprocessing of PDW data (i.e., raw PDW pulse descriptor host data and raw PDW pulse descriptor slave data). By integrating task data hierarchical and dynamic transmission scheduling, the amount of business data in the task payload can be accurately controlled, ensuring that the key data of each slave node can be transmitted to the host (i.e., the UAV host) in real time.

[0021] Specifically, based on the raw business data generated by the mission payload equipment, the pulse signals (i.e., raw PDW pulse descriptor word host data and raw PDW pulse descriptor word slave data) are preprocessed and sorted using multi-dimensional information of the pulse signals in the time domain, frequency domain, and spatial domain, and PDW cluster lists (i.e. PDW host cluster list and PDW slave cluster list) of signals from the same radiation source are selected.

[0022] Specifically, the selected PDW cluster list of signals from the same radiation source (i.e., the PDW host cluster list and the PDW slave cluster list) is sorted according to the arrival time of the radiation source signal. Only a certain number (50-100) (i.e. N) of pulse parameters after the start pulse of each radiation source signal are extracted for transmission, which greatly reduces the amount of redundant raw data generated by the task load.

[0023] It should be understood that a data classification model of "task priority-data type-time threshold" (i.e., a pre-built data classification model) is constructed to classify the PDW, EDW, UAV position and attitude data and payload status data (i.e., target PDW pulse descriptor slave dataset, original EDW event descriptor slave data, original UAV slave self parameters and original payload status slave data) generated by the task payload, and to prioritize the real-time transmission of task payload PDW data.

[0024] Specifically, based on the real-time data acquisition quality of each UAV node (i.e., the data acquisition quality of the host and the data acquisition quality of the slave) and the network load (i.e., the network load of the host and the network load of the slave), the data transmission priority and bandwidth allocation ratio are dynamically adjusted to ensure that the critical data of each slave node can be transmitted to the host in real time.

[0025] In the above embodiments, the original UAV host business data and the original UAV slave business data are preprocessed to obtain preprocessed UAV host business data and preprocessed UAV slave business data, which reduces the amount of redundant original data generated by the task payload, realizes precise control of the amount of task payload business data, and ensures that the key data of each slave node can be transmitted to the host in real time.

[0026] Optionally, as an embodiment of the present invention, the process of performing time difference parameter analysis on the preprocessed UAV host service data and the preprocessed UAV slave service data corresponding to each of the UAV slaves to obtain the time difference parameters corresponding to each of the UAV slaves includes: Feature extraction is performed on the preprocessed UAV host service data to obtain the UAV host time domain feature, the UAV host frequency domain feature, and the UAV host air domain feature corresponding to the UAV host. Feature extraction is performed on each of the preprocessed UAV slave business data to obtain the UAV slave time domain features, UAV slave frequency domain features, and UAV slave air domain features corresponding to each UAV slave. The time-domain features of the UAV host, the frequency-domain features of the UAV host corresponding to the UAV host, the spatial features of the UAV host corresponding to the UAV host, the time-domain features of the UAV slave corresponding to each of the UAV slaves, the frequency-domain features of the UAV slave corresponding to each of the UAV slaves, and the spatial features of the UAV slave corresponding to each of the UAV slaves are fused respectively, and all feature fusion results are combined to obtain the original fused feature matrix corresponding to each of the UAV slaves; Outlier processing is performed on each of the original fused feature matrices to obtain the fused feature matrix to be processed corresponding to each of the UAV slaves; Each of the fusion feature matrices to be processed is normalized to obtain a normalized fusion feature matrix corresponding to each of the UAV slaves; By assigning weights to each of the normalized fused feature matrices through a pre-constructed attention mechanism layer, the assigned fused feature matrices corresponding to each of the UAV slaves are obtained. Temporal features are extracted from each of the allocated and fused feature matrices using a pre-constructed bidirectional long short-term memory network to obtain the target temporal feature matrix corresponding to each of the UAV slaves; The time-domain features, frequency-domain features, and spatial features of the UAV host are fused to obtain a fused feature matrix of the UAV host. The pre-trained machine learning model is used to pair the fused feature matrix of the UAV host and the target time-series feature matrix corresponding to each of the UAV slaves to obtain pulse pairs corresponding to each of the UAV slaves. The arrival time of the host and the arrival time of the corresponding slave of each UAV slave are extracted from each pulse pair respectively. The arrival time of the host is the time when the pulse signal arrives at the host of the UAV, and the arrival time of the slave is the time when the pulse signal arrives at the slave of the UAV. The time difference parameters corresponding to each UAV slave are obtained by calculating the difference between the arrival time of the host and the arrival time of each corresponding UAV slave.

[0027] Specifically, the host (i.e., the UAV host) receives the business data (i.e., the pre-processed UAV slave business data) from each slave node (i.e., the UAV slave). It uses a time difference positioning pulse pairing method based on machine learning and multi-dimensional information fusion to achieve joint extraction of time domain, frequency domain, and spatial domain features through a multi-dimensional information feature fusion network. Through a machine learning dynamic pairing algorithm, it fuses observation data from multiple nodes and achieves intelligent dynamic pairing to form the three-dimensional time difference parameters of the target signal.

[0028] It should be understood that a three-dimensional feature set of "time domain-frequency domain-spatial domain" (i.e., UAV host time domain features, UAV host frequency domain features, UAV host spatial domain features, UAV slave time domain features, UAV slave frequency domain features, and UAV slave spatial domain features) is constructed, and the joint extraction of signal parameters in the time domain, frequency domain, and spatial domain features is realized through a multi-dimensional information feature fusion network.

[0029] Specifically, by using multi-dimensional information fusion based on attention mechanism (i.e., pre-built attention mechanism layer), the multi-domain feature vector (i.e. normalized fused feature matrix) and traditional time difference statistical features are dynamically weighted according to signal-to-noise ratio and application scenario, and the temporal correlation between features is learned through bidirectional long short-term memory network (i.e. pre-built bidirectional long short-term memory network).

[0030] It should be understood that by using a machine learning dynamic pairing algorithm (i.e., a pre-trained machine learning model), observation data from multiple UAV nodes are fused and intelligent dynamic pairing is achieved to form multi-UAV three-dimensional time difference parameters for the target signal.

[0031] In the above embodiments, time difference parameters are obtained by analyzing the preprocessed UAV host business data and the preprocessed UAV slave business data, respectively. This achieves the joint extraction of time domain, frequency domain, and spatial domain features, and learns the temporal correlation between features.

[0032] Optionally, as an embodiment of the present invention, all the said UAV slave coordinate positions include a first UAV slave coordinate position, a second UAV slave coordinate position, a third UAV slave coordinate position, and a fourth UAV slave coordinate position, and all the said time difference parameters include a first time difference parameter, a second time difference parameter, a third time difference parameter, and a fourth time difference parameter; The process of performing positioning analysis on the coordinate positions of the UAV host, all the coordinate positions of all UAV slaves, and all the time difference parameters using the time difference positioning model to obtain the UAV collaborative reconnaissance positioning results includes: The coordinates of the UAV host, the first UAV slave, the second UAV slave, the third UAV slave, the fourth UAV slave, the first time difference parameter, the second time difference parameter, the third time difference parameter, and the fourth time difference parameter are calculated using the first formula to obtain the coordinates of the reconnaissance target. This reconnaissance target coordinates are then used as the result of the UAV collaborative reconnaissance and positioning. The first formula is: , in, , in, To detect the target's coordinates, The coordinates of the drone host are as follows: The coordinates of the first UAV slave are: The coordinates of the second UAV slave are given. The coordinates of the third UAV slave are given. The coordinates of the fourth UAV slave are given. This is the first distance difference. This is the second distance difference. This is the third distance difference. This is the fourth distance difference. The distance between the coordinates of the UAV host and the coordinates of the reconnaissance target. Let be the distance between the coordinates of the first UAV slave and the coordinates of the reconnaissance target. The distance between the coordinates of the second UAV slave and the coordinates of the reconnaissance target is given. The distance between the coordinates of the third UAV slave and the coordinates of the reconnaissance target is given. The distance between the coordinates of the fourth UAV slave and the coordinates of the reconnaissance target is given. At the speed of light, It can be the first time difference parameter, the second time difference parameter, the third time difference parameter, or the fourth time difference parameter. For the first A drone from the machine, , For the first The difference between the distance from the drone's host coordinates to the target coordinates and the distance from the drone's host coordinates to the target coordinates.

[0033] It should be understood that, based on the obtained three-dimensional time difference parameters (i.e., the first time difference parameter, the second time difference parameter, the third time difference parameter, and the fourth time difference parameter), the position of the target signal (i.e., the coordinate position of the UAV host, the coordinate position of the first UAV slave, the coordinate position of the second UAV slave, the coordinate position of the third UAV slave, and the coordinate position of the fourth UAV slave) is calculated according to the five-machine three-dimensional time difference positioning principle, so as to achieve the high-precision positioning requirement of the target signal.

[0034] Specifically, taking a drone sub-swarm as a typical mission execution scenario, and based on the optimal configuration of five-drone collaborative time-difference positioning, the relative positional relationships and reconnaissance area of ​​the sub-swarm master and slave drones are set; assuming the coordinate position of target T (i.e., the coordinate position of the reconnaissance target) is... The coordinates of the five drones S0, S1, S2, S3, and S4 (i.e., the coordinates of the main drone, the first drone slave, the second drone slave, the third drone slave, and the fourth drone slave) are as follows: , , , , Where S0 is the master drone (i.e., the coordinates of the master drone), and the other four are slave drones (i.e., the coordinates of the first, second, third, and fourth slave drones). The arrival times of the target T's transmitted signal at each station are as follows: , , , , The formula for calculating the time difference positioning of five machines in a coordinated manner (i.e., the time difference positioning model) is as follows: , In the formula, This represents the distance between the host and the target. , , , The distances between each of the four slave devices and the target. At the speed of light, The time difference between the arrival of the target signal at the slave and master devices. The difference between the theoretical distance and the actual test time difference for the target signal to reach the slave and master devices is calculated. i = 1, 2, ..., 4, representing the i-th slave device.

[0035] The above formula can be converted into matrix form as follows: , in , , , Solving the matrix above, we can obtain the target signal coordinates (i.e., the reconnaissance target coordinates) as follows: .

[0036] In the above embodiments, the coordinate positions of the UAV host, all UAV slaves, and all time difference parameters are analyzed using a time difference positioning model to obtain the UAV collaborative reconnaissance positioning results. This achieves the high-precision positioning requirement for target signals, solves the problem of excessive data transmission load of ad hoc network communication radios in multi-UAV collaborative reconnaissance, improves the stability and intelligence accuracy of collaborative reconnaissance missions, effectively solves the problems of weak anti-interference capability and difficulty in multi-target discrimination of traditional methods in complex electromagnetic environments, improves the accuracy and robustness of pulse pairing and time difference pairing in complex scenarios, ensures the reliability of the time difference positioning system, can meet the reconnaissance and positioning needs of large UAV swarms, and has great application value.

[0037] Optionally, as another embodiment of the present invention, the present invention utilizes a multi-dimensional perception adaptive data transmission optimization method. Through PDW data fusion preprocessing, it reduces the amount of redundant raw data generated by the task payload. By integrating task data hierarchical classification and dynamic transmission scheduling, it achieves precise control over the amount of business data in the task payload, ensuring real-time transmission of key data. This solves the problem of excessive data transmission load for ad hoc network communication radios in multi-UAV collaborative reconnaissance, improving the stability and intelligence accuracy of collaborative reconnaissance missions. Utilizing a time-difference positioning pulse pairing method based on machine learning and multi-dimensional information fusion, it achieves joint extraction of time-domain, frequency-domain, and spatial-domain features through a multi-dimensional information feature fusion network. Through a machine learning dynamic pairing algorithm, it fuses multi-node observation data and achieves intelligent dynamic pairing, effectively solving the problems of weak anti-interference capability and difficulty in multi-target discrimination in complex electromagnetic environments using traditional histogram methods or correlation methods. This improves the accuracy and robustness of pulse pairing and time-difference pairing in complex scenarios, ensuring the reliability of the time-difference positioning system and possessing significant application value.

[0038] Alternatively, as another embodiment of the present invention, such as Figure 2 As shown, this invention uses the optimal configuration of five-machine collaborative time-difference positioning to set the relative positional relationship and reconnaissance area of ​​the sub-cluster master and slave machines; assuming the coordinates of target T are... The coordinates of the five drones S0, S1, S2, S3, and S4 are as follows: , , , , The system consists of four drones, with UAV S0 serving as the master and the other four as slaves. The master drone communicates with the ground control station via a wireless data link, transmitting the target location of the radiation source signal to the ground control station in real time.

[0039] Figure 3 This is a block diagram of a UAV collaborative reconnaissance and positioning device provided in an embodiment of the present invention.

[0040] Alternatively, as another embodiment of the present invention, such as Figure 3 As shown, a UAV collaborative reconnaissance and positioning device includes: The data acquisition module is used to obtain the coordinate position of the drone host and the original drone host business data from the drone host in the drone cluster, and to obtain the coordinate position of the drone slave corresponding to each drone slave and the original drone slave business data corresponding to each drone slave from multiple drone slaves in the drone cluster. The preprocessing module is used to preprocess the original UAV host business data and the original UAV slave business data corresponding to each of the UAV slaves, respectively, to obtain the preprocessed UAV host business data corresponding to the UAV host and the preprocessed UAV slave business data corresponding to each of the UAV slaves. The time difference parameter analysis module is used to perform time difference parameter analysis on the preprocessed UAV host business data and the preprocessed UAV slave business data corresponding to each of the UAV slaves, and obtain the time difference parameters corresponding to each of the UAV slaves. The positioning result acquisition module is used to construct a time difference positioning model. Through the time difference positioning model, the positioning analysis is performed on the coordinate position of the UAV host, the coordinate positions of all UAV slaves, and all the time difference parameters to obtain the UAV collaborative reconnaissance positioning result.

[0041] Optionally, as an embodiment of the present invention, the original UAV host service data includes original PDW pulse descriptor host data, original EDW event descriptor host data, original UAV host self-parameters, and original payload status host data; the original UAV slave service data includes original PDW pulse descriptor slave data, original EDW event descriptor slave data, original UAV slave self-parameters, and original payload status slave data. The preprocessing module is specifically used for: PDW preprocessing is performed on the original PDW pulse descriptor host data and the original PDW pulse descriptor slave data corresponding to each of the UAV slaves to obtain the preprocessed PDW pulse descriptor host data corresponding to the UAV host and the preprocessed PDW pulse descriptor slave data corresponding to each of the UAV slaves. Import pulse signals, and sort the preprocessed PDW pulse descriptor host data and the preprocessed PDW pulse descriptor slave data corresponding to each UAV slave according to the pulse signals to obtain the PDW host cluster list corresponding to the UAV host and the PDW slave cluster list corresponding to each UAV slave. Sort all PDW pulse descriptor host data in the PDW host clustering list according to the signal arrival time, and take the first N PDW pulse descriptor host data as the target PDW pulse descriptor host data, thereby obtaining the target PDW pulse descriptor host dataset; Sort all PWD pulse descriptor slave data in each of the PDW slave cluster lists according to the signal arrival time, and take the first N PWD pulse descriptor slave data as the target PWD pulse descriptor slave data, thereby obtaining the target PWD pulse descriptor slave dataset corresponding to each of the UAV slaves; The target PDW pulse descriptor host dataset, the original EDW event descriptor host data, the original UAV host self-parameters, and the original payload status host data are processed by a pre-built data classification model to obtain classified PDW pulse descriptor host dataset, classified EDW event descriptor host data, classified UAV host self-parameters, and classified payload status host data. By using a pre-built data grading model, the target PDW pulse descriptor slave dataset, the original EDW event descriptor slave data corresponding to each UAV slave, the original UAV slave self parameters corresponding to each UAV slave, and the original load status slave data corresponding to each UAV slave are processed to obtain the graded PDW pulse descriptor slave dataset, the graded EDW event descriptor slave data, the graded UAV slave self parameters, and the graded load status slave data corresponding to each UAV slave; The host data acquisition quality and host network load are obtained from the host drone, and the slave data acquisition quality and slave network load corresponding to each of the slave drones are obtained from each of the slave drones respectively. Based on the host data acquisition quality, the transmission priority of the graded PDW pulse descriptor host dataset, the graded EDW event descriptor host data, the graded UAV host self-parameters, and the graded payload status host data are optimized to obtain the optimized PDW pulse descriptor host dataset, the optimized EDW event descriptor host data, the optimized UAV host self-parameters, and the optimized payload status host data. Based on the data acquisition quality of each slave device, the transmission priority of each tiered PDW pulse descriptor slave dataset, the tiered EDW event descriptor slave data corresponding to each UAV slave device, the tiered UAV slave device self-parameters corresponding to each UAV slave device, and the tiered load status slave data corresponding to each UAV slave device are optimized to obtain the optimized PDW pulse descriptor slave dataset, the optimized EDW event descriptor slave data, the optimized UAV slave device self-parameters, and the optimized load status slave data corresponding to each UAV slave device. Based on the host network load, bandwidth ratio allocation is performed on the optimized PDW pulse descriptor host dataset, the optimized EDW event descriptor host data, the optimized UAV host self-parameters, and the optimized load status host data to obtain the allocated PDW pulse descriptor host dataset, the allocated EDW event descriptor host data, the allocated UAV host self-parameters, and the allocated load status host data. Based on the network load of each slave device, bandwidth ratio allocation processing is performed on each optimized PDW pulse descriptor slave dataset, the optimized EDW event descriptor slave data corresponding to each UAV slave, the optimized UAV slave self parameters corresponding to each UAV slave, and the optimized load status slave data corresponding to each UAV slave. This results in the allocated PDW pulse descriptor slave dataset, the allocated EDW event descriptor slave data, the allocated UAV slave self parameters, and the allocated load status slave data corresponding to each UAV slave. The preprocessed PDW pulse descriptor host data includes the allocated PDW pulse descriptor host dataset, the allocated EDW event descriptor host data, the allocated UAV host self-parameters, and the allocated payload status host data. The preprocessed UAV slave service data includes the allocated PDW pulse descriptor slave dataset, the allocated EDW event descriptor slave data, the allocated UAV slave self-parameters, and the allocated payload status slave data.

[0042] Optionally, as an embodiment of the present invention, the time difference parameter analysis module is specifically used for: Feature extraction is performed on the preprocessed UAV host service data to obtain the UAV host time domain feature, the UAV host frequency domain feature, and the UAV host air domain feature corresponding to the UAV host. Feature extraction is performed on each of the preprocessed UAV slave business data to obtain the UAV slave time domain features, UAV slave frequency domain features, and UAV slave air domain features corresponding to each UAV slave. The time-domain features of the UAV host, the frequency-domain features of the UAV host corresponding to the UAV host, the spatial features of the UAV host corresponding to the UAV host, the time-domain features of the UAV slave corresponding to each of the UAV slaves, the frequency-domain features of the UAV slave corresponding to each of the UAV slaves, and the spatial features of the UAV slave corresponding to each of the UAV slaves are fused respectively, and all feature fusion results are combined to obtain the original fused feature matrix corresponding to each of the UAV slaves; Outlier processing is performed on each of the original fused feature matrices to obtain the fused feature matrix to be processed corresponding to each of the UAV slaves; Each of the fusion feature matrices to be processed is normalized to obtain a normalized fusion feature matrix corresponding to each of the UAV slaves; By assigning weights to each of the normalized fused feature matrices through a pre-constructed attention mechanism layer, the assigned fused feature matrices corresponding to each of the UAV slaves are obtained. Temporal features are extracted from each of the allocated and fused feature matrices using a pre-constructed bidirectional long short-term memory network to obtain the target temporal feature matrix corresponding to each of the UAV slaves; The time-domain features, frequency-domain features, and spatial features of the UAV host are fused to obtain a fused feature matrix of the UAV host. The pre-trained machine learning model is used to pair the fused feature matrix of the UAV host and the target time-series feature matrix corresponding to each of the UAV slaves to obtain pulse pairs corresponding to each of the UAV slaves. The arrival time of the host and the arrival time of the corresponding slave of each UAV slave are extracted from each pulse pair respectively. The arrival time of the host is the time when the pulse signal arrives at the host of the UAV, and the arrival time of the slave is the time when the pulse signal arrives at the slave of the UAV. The time difference parameters corresponding to each UAV slave are obtained by calculating the difference between the arrival time of the host and the arrival time of each corresponding UAV slave.

[0043] Optionally, as an embodiment of the present invention, all the said UAV slave coordinate positions include a first UAV slave coordinate position, a second UAV slave coordinate position, a third UAV slave coordinate position, and a fourth UAV slave coordinate position, and all the said time difference parameters include a first time difference parameter, a second time difference parameter, a third time difference parameter, and a fourth time difference parameter; The process of performing positioning analysis on the coordinate positions of the UAV host, all the coordinate positions of all UAV slaves, and all the time difference parameters using the time difference positioning model to obtain the UAV collaborative reconnaissance positioning results includes: The coordinates of the UAV host, the first UAV slave, the second UAV slave, the third UAV slave, the fourth UAV slave, the first time difference parameter, the second time difference parameter, the third time difference parameter, and the fourth time difference parameter are calculated using the first formula to obtain the coordinates of the reconnaissance target. This reconnaissance target coordinates are then used as the result of the UAV collaborative reconnaissance and positioning. The first formula is: , in, , in, To detect the target's coordinates, The coordinates of the drone host are as follows: The coordinates of the first UAV slave are: The coordinates of the second UAV slave are given. The coordinates of the third UAV slave are given. The coordinates of the fourth UAV slave are given. This is the first distance difference. This is the second distance difference. This is the third distance difference. This is the fourth distance difference. The distance between the coordinates of the UAV host and the coordinates of the reconnaissance target. Let be the distance between the coordinates of the first UAV slave and the coordinates of the reconnaissance target. The distance between the coordinates of the second UAV slave and the coordinates of the reconnaissance target is given. The distance between the coordinates of the third UAV slave and the coordinates of the reconnaissance target is given. The distance between the coordinates of the fourth UAV slave and the coordinates of the reconnaissance target is given. At the speed of light, It can be the first time difference parameter, the second time difference parameter, the third time difference parameter, or the fourth time difference parameter. For the first A drone from the machine, , For the first The difference between the distance from the drone's host coordinates to the target coordinates and the distance from the drone's host coordinates to the target coordinates.

[0044] Optionally, another embodiment of the present invention provides a UAV cooperative reconnaissance and positioning system, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the UAV cooperative reconnaissance and positioning method as described above. This system can be a computer or similar system.

[0045] Optionally, another embodiment of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the UAV cooperative reconnaissance and positioning method as described above.

[0046] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0047] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described apparatus and unit can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0048] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed.

[0049] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of the present invention, depending on actual needs.

[0050] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0051] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0052] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for cooperative reconnaissance and positioning using unmanned aerial vehicles (UAVs), characterized in that, Includes the following steps: The coordinates and original business data of the drone host are obtained from the drone host in the drone cluster, and the coordinates and original business data of the drone slave corresponding to each drone slave are obtained from the multiple drone slaves in the drone cluster. The original UAV host service data and the original UAV slave service data corresponding to each of the UAV slaves are preprocessed to obtain the preprocessed UAV host service data corresponding to the UAV host and the preprocessed UAV slave service data corresponding to each of the UAV slaves. Time difference parameter analysis was performed on the preprocessed UAV host business data and the preprocessed UAV slave business data corresponding to each UAV slave to obtain the time difference parameter corresponding to each UAV slave. A time-difference positioning model is constructed, and the coordinate positions of the UAV host, all UAV slaves, and all time-difference parameters are analyzed using the time-difference positioning model to obtain the UAV collaborative reconnaissance positioning results.

2. The UAV cooperative reconnaissance and positioning method according to claim 1, characterized in that, The original UAV host service data includes original PDW pulse descriptor host data, original EDW event descriptor host data, original UAV host self-parameters, and original payload status host data. The original UAV slave service data includes original PDW pulse descriptor slave data, original EDW event descriptor slave data, original UAV slave self-parameters, and original payload status slave data. The process of preprocessing the original UAV host service data and the original UAV slave service data corresponding to each of the UAV slaves to obtain the preprocessed UAV host service data corresponding to the UAV host and the preprocessed UAV slave service data corresponding to each of the UAV slaves includes: PDW preprocessing is performed on the original PDW pulse descriptor host data and the original PDW pulse descriptor slave data corresponding to each of the UAV slaves to obtain the preprocessed PDW pulse descriptor host data corresponding to the UAV host and the preprocessed PDW pulse descriptor slave data corresponding to each of the UAV slaves. Import pulse signals, and sort the preprocessed PDW pulse descriptor host data and the preprocessed PDW pulse descriptor slave data corresponding to each UAV slave according to the pulse signals to obtain the PDW host cluster list corresponding to the UAV host and the PDW slave cluster list corresponding to each UAV slave. Sort all PDW pulse descriptor host data in the PDW host clustering list according to the signal arrival time, and take the first N PDW pulse descriptor host data as the target PDW pulse descriptor host data, thereby obtaining the target PDW pulse descriptor host dataset; Sort all PWD pulse descriptor slave data in each of the PDW slave cluster lists according to the signal arrival time, and take the first N PWD pulse descriptor slave data as the target PWD pulse descriptor slave data, thereby obtaining the target PWD pulse descriptor slave dataset corresponding to each of the UAV slaves; The target PDW pulse descriptor host dataset, the original EDW event descriptor host data, the original UAV host self-parameters, and the original payload status host data are processed by a pre-built data classification model to obtain classified PDW pulse descriptor host dataset, classified EDW event descriptor host data, classified UAV host self-parameters, and classified payload status host data. By using a pre-built data grading model, the target PDW pulse descriptor slave dataset, the original EDW event descriptor slave data corresponding to each UAV slave, the original UAV slave self parameters corresponding to each UAV slave, and the original load status slave data corresponding to each UAV slave are processed to obtain the graded PDW pulse descriptor slave dataset, the graded EDW event descriptor slave data, the graded UAV slave self parameters, and the graded load status slave data corresponding to each UAV slave; The host data acquisition quality and host network load are obtained from the host drone, and the slave data acquisition quality and slave network load corresponding to each of the slave drones are obtained from each of the slave drones respectively. Based on the host data acquisition quality, the transmission priority of the graded PDW pulse descriptor host dataset, the graded EDW event descriptor host data, the graded UAV host self-parameters, and the graded payload status host data are optimized to obtain the optimized PDW pulse descriptor host dataset, the optimized EDW event descriptor host data, the optimized UAV host self-parameters, and the optimized payload status host data. Based on the data acquisition quality of each slave device, the transmission priority of each tiered PDW pulse descriptor slave dataset, the tiered EDW event descriptor slave data corresponding to each UAV slave device, the tiered UAV slave device self-parameters corresponding to each UAV slave device, and the tiered load status slave data corresponding to each UAV slave device are optimized to obtain the optimized PDW pulse descriptor slave dataset, the optimized EDW event descriptor slave data, the optimized UAV slave device self-parameters, and the optimized load status slave data corresponding to each UAV slave device. Based on the host network load, bandwidth ratio allocation is performed on the optimized PDW pulse descriptor host dataset, the optimized EDW event descriptor host data, the optimized UAV host self-parameters, and the optimized load status host data to obtain the allocated PDW pulse descriptor host dataset, the allocated EDW event descriptor host data, the allocated UAV host self-parameters, and the allocated load status host data. Based on the network load of each slave device, bandwidth ratio allocation processing is performed on each optimized PDW pulse descriptor slave dataset, the optimized EDW event descriptor slave data corresponding to each UAV slave, the optimized UAV slave self parameters corresponding to each UAV slave, and the optimized load status slave data corresponding to each UAV slave. This results in the allocated PDW pulse descriptor slave dataset, the allocated EDW event descriptor slave data, the allocated UAV slave self parameters, and the allocated load status slave data corresponding to each UAV slave. The preprocessed PDW pulse descriptor host data includes the allocated PDW pulse descriptor host dataset, the allocated EDW event descriptor host data, the allocated UAV host self-parameters, and the allocated payload status host data. The preprocessed UAV slave service data includes the allocated PDW pulse descriptor slave dataset, the allocated EDW event descriptor slave data, the allocated UAV slave self-parameters, and the allocated payload status slave data.

3. The UAV cooperative reconnaissance and positioning method according to claim 1, characterized in that, The process of performing time difference parameter analysis on the preprocessed UAV host service data and the preprocessed UAV slave service data corresponding to each UAV slave to obtain the time difference parameters corresponding to each UAV slave includes: Feature extraction is performed on the preprocessed UAV host service data to obtain the UAV host time domain feature, the UAV host frequency domain feature, and the UAV host air domain feature corresponding to the UAV host. Feature extraction is performed on each of the preprocessed UAV slave business data to obtain the UAV slave time domain features, UAV slave frequency domain features, and UAV slave air domain features corresponding to each UAV slave. The time-domain features of the UAV host, the frequency-domain features of the UAV host corresponding to the UAV host, the spatial features of the UAV host corresponding to the UAV host, the time-domain features of the UAV slave corresponding to each of the UAV slaves, the frequency-domain features of the UAV slave corresponding to each of the UAV slaves, and the spatial features of the UAV slave corresponding to each of the UAV slaves are fused respectively, and all feature fusion results are combined to obtain the original fused feature matrix corresponding to each of the UAV slaves; Outlier processing is performed on each of the original fused feature matrices to obtain the fused feature matrix to be processed corresponding to each of the UAV slaves; Each of the fusion feature matrices to be processed is normalized to obtain a normalized fusion feature matrix corresponding to each of the UAV slaves; By assigning weights to each of the normalized fused feature matrices through a pre-constructed attention mechanism layer, the assigned fused feature matrices corresponding to each of the UAV slaves are obtained. Temporal features are extracted from each of the allocated and fused feature matrices using a pre-constructed bidirectional long short-term memory network to obtain the target temporal feature matrix corresponding to each of the UAV slaves; The time-domain features, frequency-domain features, and spatial features of the UAV host are fused to obtain a fused feature matrix of the UAV host. The pre-trained machine learning model is used to pair the fused feature matrix of the UAV host and the target time-series feature matrix corresponding to each of the UAV slaves to obtain pulse pairs corresponding to each of the UAV slaves. The arrival time of the host and the arrival time of the corresponding slave of each UAV slave are extracted from each pulse pair respectively. The arrival time of the host is the time when the pulse signal arrives at the host of the UAV, and the arrival time of the slave is the time when the pulse signal arrives at the slave of the UAV. The time difference parameters corresponding to each UAV slave are obtained by calculating the difference between the arrival time of the host and the arrival time of each corresponding UAV slave.

4. The UAV cooperative reconnaissance and positioning method according to claim 1, characterized in that, All the aforementioned UAV slave coordinate positions include a first UAV slave coordinate position, a second UAV slave coordinate position, a third UAV slave coordinate position, and a fourth UAV slave coordinate position; all the aforementioned time difference parameters include a first time difference parameter, a second time difference parameter, a third time difference parameter, and a fourth time difference parameter. The process of performing positioning analysis on the coordinate positions of the UAV host, all the coordinate positions of all UAV slaves, and all the time difference parameters using the time difference positioning model to obtain the UAV collaborative reconnaissance positioning results includes: The coordinates of the UAV host, the first UAV slave, the second UAV slave, the third UAV slave, the fourth UAV slave, the first time difference parameter, the second time difference parameter, the third time difference parameter, and the fourth time difference parameter are calculated using the first formula to obtain the coordinates of the reconnaissance target. This reconnaissance target coordinates are then used as the result of the UAV collaborative reconnaissance and positioning. The first formula is: ， in, , in, To detect the target's coordinates, The coordinates of the drone host are as follows: The coordinates of the first UAV slave are: The coordinates of the second UAV slave are given. The coordinates of the third UAV slave are given. The coordinates of the fourth UAV slave are given. This is the first distance difference. This is the second distance difference. This is the third distance difference. This is the fourth distance difference. The distance between the coordinates of the UAV host and the coordinates of the reconnaissance target. Let be the distance between the coordinates of the first UAV slave and the coordinates of the reconnaissance target. The distance between the coordinates of the second UAV slave and the coordinates of the reconnaissance target is given. The distance between the coordinates of the third UAV slave and the coordinates of the reconnaissance target is given. The distance between the coordinates of the fourth UAV slave and the coordinates of the reconnaissance target is given. At the speed of light, It can be the first time difference parameter, the second time difference parameter, the third time difference parameter, or the fourth time difference parameter. For the first A drone from the machine, , For the first The difference between the distance from the drone's host coordinates to the target coordinates and the distance from the drone's host coordinates to the target coordinates.

5. A UAV cooperative reconnaissance and positioning device, characterized in that, include: The data acquisition module is used to obtain the coordinate position of the drone host and the original drone host business data from the drone host in the drone cluster, and to obtain the coordinate position of the drone slave corresponding to each drone slave and the original drone slave business data corresponding to each drone slave from multiple drone slaves in the drone cluster. The preprocessing module is used to preprocess the original UAV host business data and the original UAV slave business data corresponding to each of the UAV slaves, respectively, to obtain the preprocessed UAV host business data corresponding to the UAV host and the preprocessed UAV slave business data corresponding to each of the UAV slaves. The time difference parameter analysis module is used to perform time difference parameter analysis on the preprocessed UAV host business data and the preprocessed UAV slave business data corresponding to each of the UAV slaves, and obtain the time difference parameters corresponding to each of the UAV slaves. The positioning result acquisition module is used to construct a time difference positioning model. Through the time difference positioning model, the coordinate positions of the UAV host, the coordinate positions of all UAV slaves, and all the time difference parameters are analyzed to obtain the UAV collaborative reconnaissance positioning results.

6. The UAV cooperative reconnaissance and positioning device according to claim 5, characterized in that, The original UAV host service data includes original PDW pulse descriptor host data, original EDW event descriptor host data, original UAV host self-parameters, and original payload status host data. The original UAV slave service data includes original PDW pulse descriptor slave data, original EDW event descriptor slave data, original UAV slave self-parameters, and original payload status slave data. The preprocessing module is specifically used for: PDW preprocessing is performed on the original PDW pulse descriptor host data and the original PDW pulse descriptor slave data corresponding to each of the UAV slaves to obtain the preprocessed PDW pulse descriptor host data corresponding to the UAV host and the preprocessed PDW pulse descriptor slave data corresponding to each of the UAV slaves. Import pulse signals, and sort the preprocessed PDW pulse descriptor host data and the preprocessed PDW pulse descriptor slave data corresponding to each UAV slave according to the pulse signals to obtain the PDW host cluster list corresponding to the UAV host and the PDW slave cluster list corresponding to each UAV slave. Sort all PDW pulse descriptor host data in the PDW host clustering list according to the signal arrival time, and take the first N PDW pulse descriptor host data as the target PDW pulse descriptor host data, thereby obtaining the target PDW pulse descriptor host dataset; Sort all PWD pulse descriptor slave data in each of the PDW slave cluster lists according to the signal arrival time, and take the first N PWD pulse descriptor slave data as the target PWD pulse descriptor slave data, thereby obtaining the target PWD pulse descriptor slave dataset corresponding to each of the UAV slaves; The target PDW pulse descriptor host dataset, the original EDW event descriptor host data, the original UAV host self-parameters, and the original payload status host data are processed by a pre-built data classification model to obtain classified PDW pulse descriptor host dataset, classified EDW event descriptor host data, classified UAV host self-parameters, and classified payload status host data. By using a pre-built data grading model, the target PDW pulse descriptor slave dataset, the original EDW event descriptor slave data corresponding to each UAV slave, the original UAV slave self parameters corresponding to each UAV slave, and the original load status slave data corresponding to each UAV slave are processed to obtain the graded PDW pulse descriptor slave dataset, the graded EDW event descriptor slave data, the graded UAV slave self parameters, and the graded load status slave data corresponding to each UAV slave; The host data acquisition quality and host network load are obtained from the host drone, and the slave data acquisition quality and slave network load corresponding to each of the slave drones are obtained from each of the slave drones respectively. Based on the host data acquisition quality, the transmission priority of the graded PDW pulse descriptor host dataset, the graded EDW event descriptor host data, the graded UAV host self-parameters, and the graded payload status host data are optimized to obtain the optimized PDW pulse descriptor host dataset, the optimized EDW event descriptor host data, the optimized UAV host self-parameters, and the optimized payload status host data. Based on the data acquisition quality of each slave device, the transmission priority of each tiered PDW pulse descriptor slave dataset, the tiered EDW event descriptor slave data corresponding to each UAV slave device, the tiered UAV slave device self-parameters corresponding to each UAV slave device, and the tiered load status slave data corresponding to each UAV slave device are optimized to obtain the optimized PDW pulse descriptor slave dataset, the optimized EDW event descriptor slave data, the optimized UAV slave device self-parameters, and the optimized load status slave data corresponding to each UAV slave device. Based on the host network load, bandwidth ratio allocation is performed on the optimized PDW pulse descriptor host dataset, the optimized EDW event descriptor host data, the optimized UAV host self-parameters, and the optimized load status host data to obtain the allocated PDW pulse descriptor host dataset, the allocated EDW event descriptor host data, the allocated UAV host self-parameters, and the allocated load status host data. Based on the network load of each slave device, bandwidth ratio allocation processing is performed on each optimized PDW pulse descriptor slave dataset, the optimized EDW event descriptor slave data corresponding to each UAV slave, the optimized UAV slave self parameters corresponding to each UAV slave, and the optimized load status slave data corresponding to each UAV slave. This results in the allocated PDW pulse descriptor slave dataset, the allocated EDW event descriptor slave data, the allocated UAV slave self parameters, and the allocated load status slave data corresponding to each UAV slave. The preprocessed PDW pulse descriptor host data includes the allocated PDW pulse descriptor host dataset, the allocated EDW event descriptor host data, the allocated UAV host self-parameters, and the allocated payload status host data. The preprocessed UAV slave service data includes the allocated PDW pulse descriptor slave dataset, the allocated EDW event descriptor slave data, the allocated UAV slave self-parameters, and the allocated payload status slave data.

7. The UAV cooperative reconnaissance and positioning device according to claim 5, characterized in that, The time difference parameter analysis module is specifically used for: Feature extraction is performed on the preprocessed UAV host service data to obtain the UAV host time domain feature, the UAV host frequency domain feature, and the UAV host air domain feature corresponding to the UAV host. Feature extraction is performed on each of the preprocessed UAV slave business data to obtain the UAV slave time domain features, UAV slave frequency domain features, and UAV slave air domain features corresponding to each UAV slave. The time-domain features of the UAV host, the frequency-domain features of the UAV host corresponding to the UAV host, the spatial features of the UAV host corresponding to the UAV host, the time-domain features of the UAV slave corresponding to each of the UAV slaves, the frequency-domain features of the UAV slave corresponding to each of the UAV slaves, and the spatial features of the UAV slave corresponding to each of the UAV slaves are fused respectively, and all feature fusion results are combined to obtain the original fused feature matrix corresponding to each of the UAV slaves; Outlier processing is performed on each of the original fused feature matrices to obtain the fused feature matrix to be processed corresponding to each of the UAV slaves; Each of the fusion feature matrices to be processed is normalized to obtain a normalized fusion feature matrix corresponding to each of the UAV slaves; By assigning weights to each of the normalized fused feature matrices through a pre-constructed attention mechanism layer, the assigned fused feature matrices corresponding to each of the UAV slaves are obtained. Temporal features are extracted from each of the allocated and fused feature matrices using a pre-constructed bidirectional long short-term memory network to obtain the target temporal feature matrix corresponding to each of the UAV slaves; The time-domain features, frequency-domain features, and spatial features of the UAV host are fused to obtain a fused feature matrix of the UAV host. The pre-trained machine learning model is used to pair the fused feature matrix of the UAV host and the target time-series feature matrix corresponding to each of the UAV slaves to obtain pulse pairs corresponding to each of the UAV slaves. The arrival time of the host and the arrival time of the corresponding slave of each UAV slave are extracted from each pulse pair respectively. The arrival time of the host is the time when the pulse signal arrives at the host of the UAV, and the arrival time of the slave is the time when the pulse signal arrives at the slave of the UAV. The time difference parameters corresponding to each UAV slave are obtained by calculating the difference between the arrival time of the host and the arrival time of each corresponding UAV slave.

8. The UAV cooperative reconnaissance and positioning device according to claim 5, characterized in that, All the aforementioned UAV slave coordinate positions include a first UAV slave coordinate position, a second UAV slave coordinate position, a third UAV slave coordinate position, and a fourth UAV slave coordinate position; all the aforementioned time difference parameters include a first time difference parameter, a second time difference parameter, a third time difference parameter, and a fourth time difference parameter. The process of performing positioning analysis on the coordinate positions of the UAV host, all the coordinate positions of all UAV slaves, and all the time difference parameters using the time difference positioning model to obtain the UAV collaborative reconnaissance positioning results includes: The coordinates of the UAV host, the first UAV slave, the second UAV slave, the third UAV slave, the fourth UAV slave, the first time difference parameter, the second time difference parameter, the third time difference parameter, and the fourth time difference parameter are calculated using the first formula to obtain the coordinates of the reconnaissance target. This reconnaissance target coordinates are then used as the result of the UAV collaborative reconnaissance and positioning. The first formula is: ， in, , in, To detect the target's coordinates, The coordinates of the drone host are as follows: The coordinates of the first UAV slave are: The coordinates of the second UAV slave are given. The coordinates of the third UAV slave are given. The coordinates of the fourth UAV slave are given. This is the first distance difference. This is the second distance difference. This is the third distance difference. This is the fourth distance difference. The distance between the coordinates of the UAV host and the coordinates of the reconnaissance target. Let be the distance between the coordinates of the first UAV slave and the coordinates of the reconnaissance target. The distance between the coordinates of the second UAV slave and the coordinates of the reconnaissance target is given. The distance between the coordinates of the third UAV slave and the coordinates of the reconnaissance target is given. The distance between the coordinates of the fourth UAV slave and the coordinates of the reconnaissance target is given. At the speed of light, It can be the first time difference parameter, the second time difference parameter, the third time difference parameter, or the fourth time difference parameter. For the first A drone from the machine, , For the first The difference between the distance from the drone's host coordinates to the target coordinates and the distance from the drone's host coordinates to the target coordinates.

9. A UAV cooperative reconnaissance and positioning device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the UAV cooperative reconnaissance and positioning method as described in any one of claims 1 to 4.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the UAV cooperative reconnaissance and positioning method as described in any one of claims 1 to 4.

Images