A target radiation source positioning method and executor based on unmanned aerial vehicle cooperation

Abstract

The application discloses a target radiation source positioning method and executor based on cooperation of unmanned aerial vehicles, and belongs to the technical field of radiation source positioning; the method comprises the following steps: firstly, a signal source is monitored through networking of unmanned aerial vehicles; secondly, a target area of the signal source is captured; when a preset feature of the target signal source is monitored by any unmanned aerial vehicle, the unmanned aerial vehicle formation determines the target area of the target signal source by outlining the contour of the signal coverage range; then, a spatial geometric relationship between the unmanned aerial vehicle and the signal source is established; finally, three-dimensional coordinates of the signal source are obtained; the target radiation source positioning method and executor based on cooperation of unmanned aerial vehicles have the advantages that through orthogonal direction node optimization and an event triggering diffusion mechanism, combined with high-precision Beidou timing and self-organizing networking communication, the positioning precision of the radiation source is improved, the wide-area search efficiency is improved, the system energy consumption is reduced, and the anti-interference and anti-damage capabilities in complex environments are enhanced.

Description

Technical Field

[0001] This invention specifically relates to a target radiation source localization method and actuator based on UAV collaboration, belonging to the field of radiation source localization technology. Background Technology

[0002] Radio positioning technology is widely used in military reconnaissance, spectrum monitoring, and emergency search and rescue. Currently, radiation source positioning methods based on UAV platforms mainly include direction finding, time difference positioning, and time-frequency difference positioning. For example, Chinese Patent Publication No. CN121995313A discloses a spin-scanning direction finding and cross-positioning method based on a single UAV platform. This method utilizes the UAV's spin to achieve mechanical scanning of the antenna, realizing single-station multi-point direction finding and positioning. However, this positioning method is energy-intensive and has low accuracy. Existing direction finding methods measure the arrival angle of the target signal using an airborne antenna array, determining the target position through multiple measurements at a single station or cross-positioning at multiple stations. The positioning accuracy decreases sharply with increasing distance and is significantly affected by multipath effects. Time difference positioning uses multiple receiving stations to measure the arrival time difference of the same target signal and calculates the position through the intersection of hyperbolas. This method has high positioning accuracy, but requires time synchronization between receiving stations. The requirements are stringent, and at least three receiving stations are needed to achieve two-dimensional positioning. Traditional TDOA positioning typically uses 2-3 UAVs for signal acquisition and positioning. When some UAVs do not meet the positioning geometry conditions due to flight attitude or environmental obstruction, they need to be temporarily dispatched to fly to a suitable location, resulting in long response time and limited coverage. Time-frequency difference positioning adds frequency difference information to TDOA, but this system has extremely high requirements for the measurement accuracy of frequency synchronization, local oscillator stability, and the motion state of the receiving station itself, making the system complex to implement. In addition, most existing technologies rely on a small number of fixed or semi-fixed receiving nodes and lack the ability to coordinate dynamic, large-scale nodes. When facing wide-area, time-varying, and complex electromagnetic environments, the small number of nodes leads to poor geometry and limited positioning accuracy. The nodes rely on ground communication or pre-planned links, resulting in response lag. It is impossible to achieve integrated operation of rapid search and accurate positioning of radiation source areas. Summary of the Invention

[0003] To address the aforementioned issues, this invention proposes a target radiation source localization method and actuator based on UAV collaboration. By optimizing orthogonal direction nodes and using an event-triggered diffusion mechanism, combined with BeiDou high-precision timing and self-organizing network communication, the method improves the positioning accuracy of the radiation source, while also enhancing wide-area search efficiency, reducing system energy consumption, and strengthening anti-interference and anti-damage capabilities in complex environments.

[0004] The target radiation source localization method based on UAV cooperation of the present invention includes the following steps: S1. UAV Network Monitoring Signal Source: By constructing a self-organizing communication network for the target area through UAV formation and continuously monitoring it, the preset characteristics of the target signal source are collected. S2. Capture the target area of ​​the signal source: When any drone detects the preset characteristics of the target signal source, the drone formation determines the target area of ​​the target signal source by outlining the contour of the signal coverage area. S3. Establish the spatial geometric relationship between the UAV and the signal source: Based on the positions of all UAVs participating in the boundary exploration, calculate the virtual center point of the target area. Using the virtual center point as the origin, determine the positioning UAVs located in the three orthogonal directions, and determine the spatial geometric relationship between the positioning UAVs and the target signal source. S4. Obtain the three-dimensional coordinates of the signal source: Based on spatial geometric relationships, determine the three-dimensional coordinates of the target signal source.

[0005] The target radiation source localization method based on UAV collaboration of the present invention optimizes the TDOA observation geometry by determining the center point of the area where the target signal source is located and selecting UAVs in three orthogonal directions to participate in the localization. This makes the hyperbola intersection angle close to the optimal state, significantly reducing the geometric accuracy factor and achieving sub-meter or even higher accuracy localization under the same measurement error. At the same time, the event-triggered diffusion mechanism realizes the transformation from blind traversal of the whole area to local adaptive focusing, avoiding invalid flight and significantly shortening the target detection and tracking response time. The system also achieves precise boundary exploration and node selection, allowing only key nodes to perform high-frequency acquisition and high-computing calculation, while the other nodes keep low-power standby, effectively reducing the overall energy consumption of the group and extending the operation endurance. Combining BeiDou's high-precision timing and networking, the system possesses decentralized resilience, enabling rapid reconstruction of the positioning configuration even after some nodes fail. Furthermore, it can maintain nanosecond-level time synchronization and reliable positioning even under GPS denial or strong electromagnetic interference environments. In addition, the network simultaneously observes targets from multiple angles and iteratively corrects the target signal source position through algorithms, effectively suppressing multipath effects and non-line-of-sight propagation errors. This significantly improves the stability and reliability of positioning results in complex environments such as densely populated urban areas.

[0006] Furthermore, the construction of a self-organizing communication network for the target area through drone swarms specifically includes: S11. Each drone is equipped with a communication module, which loads preset network parameters and initializes the network protocol stack after power-on. S12. Each UAV periodically broadcasts a beacon frame containing its own identifier and status on a preset channel, while simultaneously listening to the beacon frames of other UAVs. S13. Upon receiving the beacon frame, the drone records the identifier and signal strength of the sending drone and generates a neighbor table; S14. Each UAV exchanges routing information based on its neighbor table and through a self-organizing routing protocol, dynamically establishing a routing table and forming a self-organizing communication network. S15. During network operation, each UAV continuously monitors the link status of the self-organizing communication network. When a link interruption or node change is detected, route repair and topology reconstruction are automatically triggered.

[0007] Furthermore, determining the target region of the target signal source specifically includes: S16. When any one or more drones detect that the strength of the target signal exceeds a preset threshold, it is determined that a suspected target signal has been detected, and a trigger signal is broadcast to nearby drones through the self-organizing communication network. S17. Nearby drones that receive the trigger signal move toward the triggering drone, forming a local high-density cluster around the suspected target signal. S18. Each UAV in the cluster measures the signal-to-noise ratio (SNR) of the target signal it receives and aggregates the SNR information to the ground server. The ground server determines the locations where the SNR intensity is lower than a preset threshold as the signal coverage edge based on the spatial distribution of the SNR. S19. The UAV located at the edge of the signal coverage hovers or scans, dynamically measures the signal boundary, gradually delineates the coverage outline of the signal source, and determines the area inside the outline as the target area where the target signal source may exist.

[0008] Furthermore, each UAV is equipped with a BeiDou positioning and timing unit, and the preset characteristics of the target signal source for acquisition specifically include: S21. Each UAV receives the time signal from the Beidou satellite and calibrates and corrects the frequency and phase of the local crystal oscillator in real time to obtain the real-time position coordinates of each UAV. S22. The drone formation conducts a distributed wide-area search in the target area according to the preset coverage path. Each drone continuously monitors the spectrum environment of the target area and detects the preset characteristics of the target signal source in real time.

[0009] Furthermore, the real-time calibration and correction of the local crystal oscillator's frequency and phase specifically includes: S211. Receive BeiDou pulses and time information sent by BeiDou satellites, and use BeiDou pulses as an absolute time reference; S212. Divide the high-frequency clock signal output by the local crystal oscillator to generate a local pulse; S213. Compare the time difference between the rising edges of the BeiDou pulse and the local pulse using a phase detector, and output the phase difference signal. S214. After processing the phase difference signal through a loop filter, a control voltage is generated; S215. Apply control voltage to the control terminal of the voltage-controlled crystal oscillator to adjust the frequency and phase of the local crystal oscillator in real time.

[0010] Furthermore, the virtual center point of the calculated target region specifically includes: S31. Obtain the position coordinates of all drones participating in the boundary exploration and the received signal-to-noise ratio of the same target signal measured by each drone; S32. Use the received signal-to-noise ratio value of each drone as the weight of each drone in the calculation, and add up the signal-to-noise ratio weights of all drones to obtain the total weight. S33. Multiply the longitude, latitude, and altitude values ​​of each UAV by the corresponding signal-to-noise ratio weights to obtain the weighted longitude, weighted latitude, and weighted altitude of each UAV. Sum the weighted longitude, weighted latitude, and weighted altitude of all UAVs to obtain the total weighted longitude, total weighted latitude, and total weighted altitude. S34. Divide the total weighted longitude by the total weight to obtain the longitude value of the virtual center point; divide the total weighted latitude by the total weight to obtain the latitude value of the virtual center point; divide the total weighted altitude by the total weight to obtain the altitude value of the virtual center point. S35. By integrating the longitude, latitude, and altitude values, the three-dimensional coordinates of the virtual center point are obtained.

[0011] Furthermore, determining the spatial geometric relationship between the positioning drone and the target signal source specifically includes: S36. Establish a local three-dimensional coordinate system with the virtual center point of the target area as the origin; S37. Obtain the position coordinates and attitude information of each UAV participating in the exploration in the global coordinate system; S38. Using the direction cosine matrix, the position coordinates and attitude information are transformed into a local three-dimensional coordinate system to obtain the three-dimensional direction vector of each UAV relative to the virtual center point; calculate the axial proximity between the three-dimensional direction vector of each UAV and the three orthogonal coordinate axes in the local three-dimensional coordinate system. S39. Based on axial proximity, the UAV closest to the three orthogonal axes is selected as the positioning UAV, and the spatial geometric relationship between the positioning UAV and the target signal source is determined.

[0012] Furthermore, determining the three-dimensional coordinates of the target signal source specifically includes: S41. Simultaneously intercept signals emitted by the target signal source by positioning the drone, and record the arrival timestamp of each intercepted signal; S42. Calculate the initial position of the target signal source based on the arrival timestamp; S43. Iteratively correct the initial position to obtain the three-dimensional coordinates of the target signal source.

[0013] Furthermore, the calculation of the initial position of the target signal source specifically includes: S421. Obtain the three-dimensional coordinates of at least three reference UAVs and determine the reference UAV; S422. Obtain the arrival time difference of the target signal source to each reference UAV relative to the reference UAV, and convert the time difference into a distance difference; S423. Introduce an auxiliary variable, which is the distance from the target signal source to the reference UAV; S424. Using the distance difference and auxiliary variables, the nonlinear TDOA hyperbolic equation system is transformed into a pseudo-linear equation system about the three-dimensional coordinates of the target signal source and auxiliary variables. S425. The pseudo-linear equation system is solved by the first weighted least squares method to obtain the initial solution of the target coordinates and auxiliary variables; S426. Using the geometric constraint relationship between the three-dimensional coordinates of the target signal source and the auxiliary variables, construct the second weighted least squares equation, correct the initial solution, and obtain the initial position of the target signal source.

[0014] Furthermore, the iterative correction of the initial position specifically includes: S431. Based on the initial position, construct a system of linear equations about the position correction using Taylor series, and solve them to obtain the correction amount for the current position. S432. Add the correction amount to the initial position to obtain the candidate new position, and calculate the error index of the candidate new position; S433. If the error index of the candidate new position is less than the error index of the initial position, then accept the candidate new position as the estimated position for the next iteration; otherwise, recalculate the candidate position. S434. When the correction amount reaches the preset iteration condition, stop the iteration and output the three-dimensional coordinates of the target signal source.

[0015] Furthermore, it also includes dynamic tracking of the target radiation source: when the target signal source is determined to be in a moving state, the motion state parameters of the target are estimated based on the positioning results of multiple consecutive frames; the position area of ​​the target at the next moment is predicted according to the motion state parameters, and the UAV formation is dynamically scheduled to adjust its position, maintaining the orthogonal spatial geometric relationship between the positioning UAV and the target throughout the process; the calculation is repeated according to the preset positioning cycle, and the continuous motion trajectory of the target signal source is output.

[0016] Furthermore, the dynamic scheduling of UAV formations to adjust their positions specifically includes: using extended Kalman filtering to perform joint state estimation of the target's position, velocity, and acceleration, generating a predicted target position and confidence interval for the next moment; using the predicted position as the new virtual center, recalculating the axial proximity of each UAV relative to the virtual center; when the axial proximity of the currently positioned UAV is lower than a preset configuration threshold, selecting the replacement node with the highest matching degree from the surrounding standby UAVs, and using a smooth transition trajectory scheduling to switch the replacement node into the positioning configuration while the original positioning node retreats to the standby state, ensuring that at least three UAVs maintain an orthogonal observation configuration throughout the switching process.

[0017] Furthermore, the dynamic tracking step also includes orthogonal configuration-assisted multipath signal discrimination and correction: using the signal arrival direction vectors of the three orthogonal positioning UAVs, spatial matching verification is performed on the multipath components of the same signal frame; components whose direction vectors deviate from the target main signal direction by more than a preset angle threshold are identified as reflected multipath signals and are removed in the arrival time calculation; the arrival times of the remaining main path signals are fused by signal-to-noise ratio weighting, and the corrected arrival time difference is output for positioning calculation.

[0018] Furthermore, it also includes hierarchical power consumption control for tracking, dividing the tracking mode into three levels based on the target's movement speed: when the target's movement speed is below the first speed threshold, it enters the low-power tracking mode, maintaining 3 positioning drones in operation, reducing the signal sampling frequency to 1 / 2 of the baseline value, and the remaining drones enter a low-power cruise standby state; when the target's movement speed is above the first speed threshold but below the second speed threshold, it enters the standard tracking mode, maintaining 4 positioning drones in operation, using the baseline sampling frequency; when the target's movement speed is above the second speed threshold, it enters the high-precision tracking mode, enabling 6 positioning drones to form a biorthogonal redundant configuration, increasing the sampling frequency to twice the baseline value, and simultaneously increasing the number of iterations for positioning calculation.

[0019] Furthermore, the dynamic tracking step also includes a signal loss predictive recapture mechanism: when the target signal is briefly lost, the predicted activity area of ​​the target is extrapolated based on the motion state parameters of the last frame before the loss, and the surrounding standby UAVs are scheduled to perform an involute spiral scan search along the predicted area; at the same time, according to the spatial distribution of the signal-to-noise ratio before the signal loss, the search density is tilted towards the high probability area; after the target signal is recaptured, the orthogonal positioning configuration and continuous tracking are immediately restored within one positioning cycle.

[0020] Furthermore, determining the three-dimensional coordinates of the target signal source also includes virtual baseline super-resolution enhancement: controlling the positioning UAVs in three orthogonal directions to perform small-amplitude uniform maneuvers along their respective axes, and collecting the arrival time difference sequence of the target signal at N consecutive sampling times; equating the sampling points at different times to virtual array nodes, and constructing a space-time joint long baseline observation equation; using a sparse Bayesian learning algorithm to perform super-resolution reconstruction of the target position, while using the spatial constraints of orthogonal trajectories to suppress multipath pseudo-peaks, and outputting target three-dimensional coordinates with a resolution better than those sampled at a single time; the construction of the space-time joint long baseline observation equation specifically includes: using the UAV position at the kth sampling time as the virtual node coordinates, and using the TDOA measurement value at that time as the observation value of the corresponding virtual node; combining the virtual nodes and observation values ​​at all times to form an extended TDOA equation system; the equivalent length of the virtual baseline is equal to the sum of the UAV maneuver distance and the physical node spacing, and the equivalent array aperture is 5-10 times larger than that of a single time.

[0021] A target radiation source localization actuator based on UAV collaboration is provided, employing a target radiation source localization method based on UAV collaboration. The localization actuator has a built-in target radiation source localization system, and the localization actuator is a localization device or storage medium; the target radiation source localization system comprises components connected in sequence. The signal frequency acquisition module is used to build a self-organizing communication network for the target area through UAV formation and continuously monitor and acquire preset characteristics of the target signal source; The signal edge determination module is used to control the drone formation to delineate the outline of the signal coverage area and determine the target area of ​​the target signal source when any drone detects the preset characteristics of the target signal source. The target node determination module is used to calculate the virtual center point of the target area based on the positions of all UAVs participating in the boundary exploration, and to determine the positioning UAVs located in three orthogonal directions with the virtual center point as the origin, and to determine the spatial geometric relationship between the positioning UAVs and the target signal source. The signal coordinate determination module is used to determine the three-dimensional coordinates of the target signal source based on spatial geometric relationships; The modules work together to complete the entire process from wide-area search to precise positioning.

[0022] Compared with existing technologies, the target radiation source localization method and actuator based on UAV cooperation of the present invention have the following advantages: 1. Significantly improved positioning accuracy: By determining the center point of the area where the target signal source is located and selecting UAVs in three orthogonal directions to participate in positioning, the TDOA observation geometry configuration is significantly optimized, making the hyperbola intersection angle close to the optimal state, greatly reducing the geometric accuracy factor, and achieving sub-meter or even higher positioning accuracy under the same measurement error. 2. Improved search efficiency and response speed: The event-triggered diffusion mechanism transforms the search from blind traversal of the entire area to local adaptive focusing, avoiding ineffective flight and significantly shortening the target discovery and tracking response time, making it suitable for wide-area fast search scenarios; 3. Reduced system energy consumption: Through precise boundary exploration and node optimization, only key nodes are allowed to perform high-frequency data acquisition and high-computing power calculation, while the remaining nodes remain in low-power standby mode, effectively reducing the overall energy consumption of the cluster and extending the operating endurance. 4. Strong anti-interference and damage resistance: Combining BeiDou high-precision time synchronization and self-organizing networking, the system has decentralized damage resistance capabilities, and can quickly reconstruct the positioning configuration after some nodes fail; it can still maintain nanosecond-level time synchronization and reliable positioning in GPS denial or strong electromagnetic interference environments. 5. Good adaptability to complex environments: The network observes the target from multiple angles simultaneously. The target signal source position is corrected by coarse localization using the Chan algorithm combined with Taylor series iteration. This effectively suppresses multipath effects and non-line-of-sight propagation errors, significantly improving the stability and reliability of localization results in complex environments such as densely populated urban areas. 6. Dynamic tracking: Continuous tracking of moving targets without interruption, stable positioning accuracy, significantly reducing formation energy consumption and shortening signal reacquisition time, and adapting to continuous monitoring of high-speed mobile radiation sources; 7. Virtual baseline super-resolution positioning can significantly improve positioning resolution and accuracy in weak signal and long-distance scenarios. Attached Figure Description

[0023] Figure 1 This is a schematic diagram illustrating the process of constructing a self-organizing communication network for a target environment using drone swarms, as per the present invention.

[0024] Figure 2 This is a schematic diagram of the signal frequency flow of the target signal source in this invention.

[0025] Figure 3 This is a schematic diagram illustrating the real-time calibration and correction process for the local crystal oscillator's frequency and phase according to the present invention.

[0026] Figure 4 This is a schematic diagram of the process for determining the target region of the target signal source according to the present invention.

[0027] Figure 5 This is a schematic diagram of the process for calculating the virtual center point of the target region according to the present invention.

[0028] Figure 6 This is a schematic diagram illustrating the process of determining the spatial geometric relationships according to the present invention.

[0029] Figure 7 This is a schematic diagram illustrating the process of determining the three-dimensional coordinates of the target signal source according to the present invention.

[0030] Figure 8 This is a schematic diagram of the process for calculating the initial position of the target signal source according to the present invention.

[0031] Figure 9 This is a schematic diagram of the iterative correction process for the initial position according to the present invention.

[0032] Figure 10 This is a schematic diagram of the target radiation source dynamic tracking process of the present invention.

[0033] Figure 11 This is a schematic diagram of the virtual baseline super-resolution enhancement process of the present invention.

[0034] Figure 12 This is a schematic diagram of the target radiation source positioning actuator based on UAV collaboration of the present invention. Detailed Implementation

[0035] Example 1: The working process of the target radiation source localization method based on UAV cooperation of the present invention is as follows: S1. Construction of self-organizing communication networks: such as... Figure 1 As shown, building a self-organizing communication network for a target area using drone swarms includes the following steps: S11. Each drone is equipped with a communication module, which loads preset network parameters and initializes the network protocol stack after power-on. S12. Each UAV periodically broadcasts a beacon frame containing its own identifier and status on a preset channel, while simultaneously listening to the beacon frames of other UAVs. S13. Upon receiving the beacon frame, the drone records the identifier and signal strength of the sending drone and generates a neighbor table; S14. Each UAV exchanges routing information based on its neighbor table and through a self-organizing routing protocol, dynamically establishing a routing table and forming a self-organizing communication network. S15. During network operation, each UAV continuously monitors the link status of the self-organizing communication network. When a link interruption or node change is detected, route repair and topology reconstruction are automatically triggered. Once the drone formation takes off, a mesh network based on OFDM wireless IP is first established using the networking communication module: the physical layer uses OFDM modulation to support non-line-of-sight (NLOS) communication in complex urban or mountainous environments, and combines MIMO multi-antenna technology to improve channel capacity and link reliability; the network layer adopts a decentralized, multi-hop self-organizing routing protocol, where each node is both a data transceiver and a relay forwarder, and can dynamically discover and maintain the optimal transmission path; it supports the TCP / IP protocol stack for seamless integration with existing wired or wireless networks, and has built-in BeiDou positioning function to support location-based routing strategies; The establishment of a mesh network (a self-organizing communication network) is an automated and distributed process: after powering on, nodes load preset network parameters and initialize the protocol stack; then, they periodically broadcast "Hello" beacon frames on preset channels while simultaneously listening to the beacons of other nodes, building a neighbor table by recording the other party's ID and signal strength (RSSI); as neighbor relationships are established, nodes broadcast route requests (RREQ) and receive route responses (RREP) through on-demand routing protocols (such as AODV), gradually forming a complete routing table and constituting a dynamic mesh topology; after the network is established, topology maintenance is continuously performed, and if a node moves, fails, or its signal is interrupted, neighboring nodes will quickly detect this and trigger route repair. The system automatically reconfigures to ensure continuous communication links; through the aforementioned self-organizing network, the system achieves high reliability without a central hub, and the failure of any node will not lead to overall paralysis, demonstrating strong resilience; the network can be automatically established within seconds and achieve millisecond-level self-healing when nodes move or links are interrupted, adapting to high-speed dynamic topologies; with the help of OFDM and multi-hop relays, signals can bypass obstacles such as buildings to achieve beyond-line-of-sight communication, with a single node's air-to-ground communication distance reaching 30-50 kilometers, supporting coverage expansion of more than 9 hops; the system can provide a continuous data transmission rate of over 60Mbps, supporting real-time transmission of multiple high-definition video and sensor data, with low end-to-end latency, meeting the needs of real-time command and control.

[0036] S2. Target signal feature acquisition and time synchronization: such as Figure 2 As shown, each UAV completes time synchronization and collects target signal characteristics through the BeiDou positioning and timing unit, specifically including the following sub-steps: S21. Each UAV receives the time signal from the Beidou satellite and calibrates and corrects the frequency and phase of the local crystal oscillator in real time to obtain the real-time position coordinates of each UAV. S22. The drone formation conducts a distributed wide-area search in the target area according to the preset coverage path. Each drone continuously monitors the spectrum environment of the target area and detects the preset characteristics of the target signal source in real time. During operation, the airborne BeiDou positioning and timing module is used to achieve space-based time synchronization: by receiving high-precision time signals (1PPS pulse per second and TOD time information) from BeiDou satellites as an absolute time reference source, the frequency and phase of the airborne local high-stability crystal oscillator are calibrated and corrected in real time with the help of an internal phase-locked loop and discipline algorithm, eliminating clock deviations and drifts between UAV nodes; at the same time, each UAV obtains its own initial position coordinates and the absolute time of each signal acquisition moment, providing a unified time reference and position basis for subsequent TDOA time difference positioning; on this basis, the UAV formation conducts wide-area distributed cruise search in the target area according to a preset coverage path (such as a bow shape or random dispersion), and each UAV continuously monitors the spectrum environment, focusing on detecting preset characteristics of target signal sources (such as RSSI abrupt changes in received signal strength at a specific frequency), efficiently discovering suspected radiation sources; like Figure 3 As shown, real-time calibration and correction of the local crystal oscillator's frequency and phase specifically includes the following steps: S31. Receive BeiDou pulses and time information sent by BeiDou satellites, and use BeiDou pulses as absolute time references; S32. Divide the high-frequency clock signal output by the local crystal oscillator to generate a local pulse; S33. Compare the time difference between the rising edges of the BeiDou pulse and the local pulse using a phase detector, and output the phase difference signal. S34. After processing the phase difference signal through a loop filter, a control voltage is generated; S35. Apply the control voltage to the control terminal of the voltage-controlled crystal oscillator to adjust the frequency and phase of the local crystal oscillator in real time.

[0037] During operation, the onboard BeiDou time synchronization module receives 1PPS second pulses and TOD time information transmitted by BeiDou satellites. The rising edge of 1PPS corresponds to a whole second in UTC time, with absolute time accuracy reaching the nanosecond level. TOD is serial time data, providing the absolute time (year, month, day, hour, minute, second) corresponding to the current 1PPS. The onboard oven-controlled crystal oscillator (OCXO) outputs a high-frequency clock signal (e.g., 10MHz), which is divided by a digital frequency divider to obtain the local second pulse. Due to factors such as initial frequency error, temperature drift, and aging of the crystal oscillator, the local second pulse will have a phase deviation from the BeiDou 1PPS. The BeiDou 1PPS and the local second pulse are sent to a phase detector (Time-to-Digital Converter, TDC). The phase detector detects the time difference between the rising edges of the two pulses and outputs a phase difference signal proportional to the time difference. The phase difference signal is then filtered by a loop filter (a low-pass filter combined with PID control). The system uses a control algorithm to smooth noise, extract long-term average deviation, and generate a stable control voltage. This control voltage is converted to an analog voltage via a digital-to-analog converter and applied to the control terminal of the voltage-controlled crystal oscillator (VCO) to linearly adjust the crystal oscillation frequency. If the local second pulse lags, the control voltage is increased to raise the crystal oscillator frequency and reduce the lag; if it leads, the frequency is lowered to slow down the local clock. To address long-term aging and drift of the crystal oscillator, a discipline algorithm is added: historical phase difference data is recorded, and the frequency offset trend and aging rate of the crystal oscillator are statistically fitted to predict and output optimized compensation amounts. When satellite signals are briefly lost, the discipline parameters can be used to maintain short-term high-precision timekeeping. This feedback adjustment process runs periodically, ultimately locking the phase difference between the local second pulse and the BeiDou 1PPS within ±5 nanoseconds, achieving nanosecond-level clock synchronization between various UAV nodes and providing a unified high-precision timestamp basis for TDOA positioning. S4: Delineating and defining the boundary of the target area, such as Figure 4 As shown, determining the target area of ​​the target signal source specifically includes the following steps: S41. When any one or more drones detect that the strength of a target signal exceeds a preset threshold, it is determined that a suspected target signal has been detected, and a trigger signal is broadcast to nearby drones through a self-organizing communication network. S42. Nearby drones that receive the trigger signal move toward the triggering drone, forming a local high-density cluster around the suspected target signal. S43. Each UAV in the cluster measures the signal-to-noise ratio of the target signal it receives and aggregates the signal-to-noise ratio information to the ground server. S44. Based on the spatial distribution of the signal-to-noise ratio, the ground server determines the location where the signal-to-noise ratio strength is lower than a preset threshold as the signal coverage edge; S45. The UAV located at the edge of the signal coverage hovers or scans, dynamically measures the signal boundary, gradually delineates the coverage outline of the signal source, and determines the area inside the outline as the target area where the target signal source may exist. During operation, an event-triggered mechanism facilitates a shift from blind wide-area searching to targeted collaborative detection: when any UAV detects a target signal strength exceeding a preset threshold, it determines a suspected target and immediately broadcasts a trigger signal containing information such as the target signal's center frequency and bandwidth to neighboring nodes via a self-organizing network, activating the collaborative detection mode. Upon receiving the trigger signal, neighboring UAVs converge towards the triggering node, forming a local high-density cluster around the suspected target. Each UAV within the cluster aggregates its measured signal-to-noise ratio (SNR) to a ground server. The ground server determines the signal coverage edge based on the spatial distribution of the SNR: areas with SNR above the threshold are the core area, while areas below the threshold are the signal coverage edge. The system schedules UAVs located at the signal edge to hover or scan, dynamically measuring the signal boundary and gradually outlining the complete coverage contour of the signal source. The collaborative boundary continues to expand outward until the target signal can no longer be detected, ultimately determining the final boundary of the signal coverage; the area within this contour is the potential target region where the target signal source may exist.

[0038] S5, Virtual center point calculation, such as Figure 5 As shown, calculating the virtual center point of the target area specifically includes the following steps: S51. Obtain the position coordinates of all drones participating in the boundary exploration and the received signal-to-noise ratio of the same target signal measured by each drone; S52. Use the received signal-to-noise ratio value of each drone as the weight of each drone in the calculation, and add up the signal-to-noise ratio weights of all drones to obtain the total weight. S53. Multiply the longitude, latitude, and altitude values ​​of each UAV by the corresponding signal-to-noise ratio weights to obtain the weighted longitude, weighted latitude, and weighted altitude of each UAV. Sum the weighted longitude, weighted latitude, and weighted altitude of all UAVs to obtain the total weighted longitude, total weighted latitude, and total weighted altitude. S54. Divide the total weighted longitude by the total weight to obtain the longitude value of the virtual center point; divide the total weighted latitude by the total weight to obtain the latitude value of the virtual center point; divide the total weighted altitude by the total weight to obtain the altitude value of the virtual center point. S55. By integrating the longitude, latitude, and altitude values, the three-dimensional coordinates of the virtual center point are obtained.

[0039] During operation, a weighted centroid algorithm is used to calculate the virtual center point: the signal-to-noise ratio (SNR) of each UAV is used as the weight, with nodes having a higher SNR having a greater influence on the center point's location. Weighted summation is performed on the longitude, latitude, and altitude dimensions, and then divided by the total weight to obtain the three-dimensional center point coordinates. The calculated virtual center point naturally leans towards areas with high SNR, more closely approximating the actual location of the target signal source, and serves as the reference coordinates for subsequent location node selection.

[0040] S6. Selection and geometric relationship determination of orthogonal orientation positioning UAVs, such as Figure 6 As shown, determining the spatial geometric relationship between the positioning drone and the target signal source specifically includes the following sub-steps: S61. Establish a local three-dimensional coordinate system with the virtual center point of the target area as the origin; S62. Obtain the position coordinates and attitude information of each UAV participating in the exploration in the global coordinate system; S63. Using the direction cosine matrix, the position coordinates and attitude information are transformed into a local three-dimensional coordinate system to obtain the three-dimensional direction vector of each UAV relative to the virtual center point. S64. Calculate the axial proximity between the three-dimensional direction vector of each UAV and the three orthogonal coordinate axes in the local three-dimensional coordinate system; S65. Based on axial proximity, the UAV closest to the three orthogonal axes is selected as the positioning UAV, and the spatial geometric relationship between the positioning UAV and the target signal source is determined.

[0041] During operation, the system establishes a local 3D coordinate system with the virtual center point as the origin. It then transforms the position of each UAV in the global ECEF coordinate system to the local coordinate system using a direction cosine matrix, obtaining the 3D direction vector of each node relative to the center point. The system calculates the angle between each direction vector and the three orthogonal coordinate axes, selects the UAV nodes closest to each axis, and finally chooses 3-4 UAVs arranged in an approximately orthogonal distribution as the positioning reference UAVs.

[0042] The selection process follows the principle of minimizing the geometrical precision factor (GDOP). The selected nodes can form the optimal TDOA observation geometry, making the intersection angle of the hyperbolic equation system close to the optimal state, greatly reducing the geometric amplification effect of positioning error, and laying the geometric foundation for subsequent high-precision solution.

[0043] S7. Calculation of three-dimensional coordinates of the target signal source: (e.g.) Figure 7 As shown, determining the three-dimensional coordinates of the target signal source specifically includes the following steps: S71. Simultaneously intercept signals emitted by the target signal source by positioning the drone, and record the arrival timestamp of each intercepted signal; S72. Calculate the initial position of the target signal source based on the arrival timestamp; S73. Iteratively correct the initial position to obtain the three-dimensional coordinates of the target signal source.

[0044] During operation, a two-step solution strategy combining the Chan algorithm and Taylor series expansion method is adopted: first, the Chan algorithm is used to quickly obtain the initial position, and then the Taylor series is used for iterative refinement, taking into account both real-time performance and positioning accuracy.

[0045] like Figure 8 As shown, calculating the initial position of the target signal source (Chan algorithm flow) specifically includes the following sub-steps: S81. Obtain the three-dimensional coordinates of at least three reference UAVs and determine the reference UAV; S82. Obtain the arrival time difference of the target signal source to each reference UAV relative to the reference UAV, and convert the time difference into a distance difference; S83. Introduce an auxiliary variable, which is the distance from the target signal source to the reference UAV; S84. Using distance difference and auxiliary variables, the nonlinear TDOA hyperbolic equation system is transformed into a pseudo-linear equation system with respect to the three-dimensional coordinates of the target signal source and auxiliary variables. S85. The pseudo-linear equation system is solved using the first weighted least squares method to obtain the initial solution of the target coordinates and auxiliary variables; S86. Using the geometric constraint relationship between the three-dimensional coordinates of the target signal source and the auxiliary variables, construct the second weighted least squares equation, correct the initial solution, and obtain the initial position of the target signal source.

[0046] During operation, the Chan algorithm transforms the nonlinear hyperbolic equation system into a pseudo-linear form by introducing auxiliary variables. It obtains the coarse positioning result through two weighted least squares solutions, which eliminates the need for iteration, has a fast calculation speed, and can approach the Cramer-Rao lower bound in positioning accuracy under moderate noise. At the same time, it avoids the local convergence problem of iterative algorithms and provides stable and accurate initial values ​​for subsequent fine iterations.

[0047] like Figure 9 As shown, the initial position is iteratively corrected (Taylor series iteration process), which specifically includes the following steps: S91. Based on the initial position, construct a system of linear equations about the position correction using Taylor series, and solve them to obtain the correction amount for the current position. S92. Add the correction amount to the initial position to obtain the candidate new position, and calculate the error index of the candidate new position; S93. If the error index of the candidate new position is less than the error index of the initial position, then accept the candidate new position as the estimated position for the next iteration; otherwise, recalculate the candidate position. S94. When the correction amount reaches the preset iteration condition, stop the iteration and output the three-dimensional coordinates of the target signal source.

[0048] During operation, the Taylor series iterative method uses the initial position output by the Chan algorithm as the initial value. It performs a first-order Taylor expansion of the nonlinear distance difference function at the current estimated position, transforming the residual into a linear function of the position correction. The correction is then solved using weighted least squares. After each iteration, the residual between the theoretical and measured time differences is recalculated. If the residual decreases, the new position is accepted; otherwise, the correction step size is adjusted and recalculated. Iteration stops when the correction magnitude is less than a preset threshold (e.g., 1 meter), the error rate of change is less than the threshold, or the maximum number of iterations is reached. The final three-dimensional coordinates of the target signal source are then output. Through the combined algorithm of Chan coarse positioning and Taylor fine iteration, the system ensures real-time positioning response while effectively suppressing measurement noise and multipath errors, outputting stable, high-precision positioning results even in complex electromagnetic environments.

[0049] Example 2: like Figures 1 to 10 As shown, this embodiment is used for emergency search and rescue missions in densely populated urban areas. The target is the VHF distress signal source carried by the missing person, with an operating frequency of 156.8MHz and a transmission power of 0.5W. Six industrial-grade multi-rotor UAVs are used in a coordinated formation to work with a ground server to complete wide-area search and high-precision positioning of the radiation source. First, system hardware configuration: The core equipment and parameters of each UAV are as follows: Radio frequency signal acquisition unit: Broadband radio frequency receiving module, supporting 100MHz-1GHz frequency band reception, sampling rate 500MS / s, dynamic range 110dB, with nanosecond-level signal arrival time stamping capability; Beidou positioning and timing unit: Beidou-3 dual-mode timing chip, supporting B1I / B2a frequency band, planar positioning accuracy 10cm, timing accuracy 20ns, equipped with temperature-controlled crystal oscillator, daily aging rate ≤1×10-10; Self-organizing network communication unit: Software-defined radio Mesh module, operating frequency band 1.4GHz, single-hop communication radius 1.2km, peak transmission rate 60Mbps, supporting AODV self-organizing routing protocol; Flight control and power system: Maximum flight time 2.5h, supporting dynamic route planning and cluster scheduling; Ground server configured with 8-core CPU + 16GB memory, responsible for aggregating data from all nodes and positioning calculation.

[0050] Second, the entire process of positioning and implementation: 1. Self-organizing communication network construction: Six drones take off at the edge of the mission area. After powering on, they automatically load preset network parameters: working channel 1.42GHz, beacon broadcast period 100ms, and routing protocol is on-demand distance vector routing protocol. After takeoff, each drone periodically broadcasts beacon frames containing its own ID, battery level, and real-time location, while listening for signals from neighboring nodes. About 3 seconds after takeoff, all drones complete the construction of neighbor tables and the generation of routing tables, forming a decentralized mesh self-organizing network with data transmission latency of less than 50ms within 3 hops. During flight, each node checks the link status every 200ms. When a drone is blocked by a building, causing a link interruption, neighboring nodes trigger route repair within 100ms, automatically switching relay paths to maintain network connectivity. 2. BeiDou Time Synchronization and Wide-Area Search: After all UAVs are powered on, the BeiDou time synchronization training process is initiated: receiving BeiDou 1PPS second pulses and TOD time information, using the BeiDou pulse as the absolute time reference, comparing the local second pulse generated by the local crystal oscillator frequency division through a digital phase detector, and adjusting the voltage-controlled crystal oscillator through a loop filter output control voltage; after the training is stable, the phase difference between the local clock of each UAV and the BeiDou reference is ≤5ns, and the time synchronization error between nodes within the formation is ≤8ns; after time synchronization is completed, the UAV formation performs a wide-area search in a bow-shaped coverage route, with a flight altitude of 150m, a cruising speed of 8m / s, a lateral spacing of 800m, and a search coverage width of approximately 1km per sortie. Each UAV continuously scans the 156.8MHz frequency point and detects the signal strength in real time; when the received signal strength is ≥-85dBm, it is identified as a suspected target; 3. Target Area Boundary Delineation: 12 minutes after mission initiation, U3 detected the target signal for the first time, with a received signal strength of -78dBm. It immediately broadcast a trigger signal to neighboring nodes via its ad hoc network, including the target frequency, signal strength, and its own position. Upon receiving the trigger signal, the remaining five UAVs converged on U3's location within 90 seconds, forming a ring-shaped detection cluster with a radius of approximately 600m around the target. Each UAV within the cluster simultaneously measured the signal-to-noise ratio (SNR) of the target signal and transmitted its position and SNR data back to the ground server once per second. The ground server set a SNR of 6dB as the edge threshold, marking locations with SNR below 6dB as signal coverage boundaries. Three UAVs located near the boundary were scheduled to perform circular circumferential scanning along the boundary at a speed of 3m / s, with the circumference gradually increasing until the SNR at three consecutive sampling points was below the threshold. This process took approximately 2 minutes to delineate the signal coverage outline, determining the circular target area with a radius of approximately 320m. 4. Calculation of Virtual Center Point of Target Area: The position coordinates and corresponding signal-to-noise ratio (SNR) data of the 5 UAVs participating in the boundary exploration are obtained as follows (using a planar coordinate system as an example, unit: meters, height uniformly 150m): U1: coordinates (200, 100), SNR 12dB; U2: coordinates (180, 350), SNR 8dB; U3: coordinates (400, 280), SNR 15dB; U4: coordinates (520, 120), SNR 7dB; U5: coordinates (350, 420), SNR 9dB; Using the SNR as the weight, the virtual center point is calculated using a weighted centroid algorithm: Total weight = 12 + 8 + 15 + 7 + 9 = 51; Total weighted average X = 200×12 + 180×8 + 400×15 + 520×7 + 350×9 = 16630; Total weighted Y = 100×12 + 350×8 + 280×15 + 120×7 + 420×9 = 12820; The virtual center point X coordinate = 16630 ÷ 51 ≈ 326.1m; The virtual center point Y-coordinate = 12820 ÷ 51 ≈ 251.4m; The final three-dimensional coordinates of the virtual center point are (326.1, 251.4, 150), which will serve as the reference origin for subsequent node selection. 5. Selection of UAVs for Orthogonal Direction Positioning: A local 3D coordinate system is established with the virtual center point as the origin, the X-axis pointing due east, the Y-axis pointing due north, and the Z-axis pointing to the zenith. The global coordinates of the five UAVs are converted into direction vectors in the local coordinate system using the direction cosine matrix. The cosine values ​​of the angles (axial proximity) between each vector and the three orthogonal coordinate axes are calculated: U1 has a proximity of 0.92 with the X-axis and 0.31 with the Y-axis; U2 has a proximity of 0.89 with the Y-axis and 0.27 with the X-axis; U3 has a proximity of less than 0.7 with each axis; U4 has a proximity of 0.94 with the X-axis and 0.18 with the Y-axis; U5 has a proximity of 0.87 with the Z-axis (flight altitude 220m). Based on the axial proximity, U4 (closest in the positive X-axis direction), U2 (closest in the positive Y-axis direction), and U5 (closest in the positive Z-axis direction) are selected as the positioning reference UAVs. The three points form an approximately orthogonal spatial geometric configuration relative to the center point, with a calculated geometric precision factor of 1.2, which is significantly better than the 2.8 of randomly selected points. 6. Target signal source three-dimensional coordinate calculation: Three positioning UAVs synchronously collect target signals and record the signal arrival timestamps; combined with the BeiDou unified time reference, the following calculations are made: with U2 as the reference node, the arrival time difference between U4 and U2 is 1.2μs, and the arrival time difference between U5 and U2 is 0.8μs, which translates to distance differences of 360m and 240m respectively; The first step uses the Chan algorithm to calculate the initial position: the distance from the target to the reference node U2 is introduced as an auxiliary variable, and the nonlinear TDOA hyperbolic equation system is transformed into a pseudo-linear equation system. After two weighted least squares solutions, the initial position coordinates of the target are obtained as (312.5, 248.7, 12.3); The second step uses this initial position as the initial value for Taylor series iterative correction: the correction amount for the first iteration is (2 The second iteration used the correction factor (0.1, -1.5, 0.8), and the magnitude of the correction factor was (0.3, 0.2, 0.1). Since the magnitude of the correction factor was less than the convergence threshold of 0.5m, the iteration stopped. The final output three-dimensional coordinates of the target signal source were (314.8, 247.2, 13.1). After on-site verification, the positioning error was about 2.7m, achieving sub-meter level positioning accuracy. In this embodiment, the total time from target discovery to outputting accurate positioning results was about 4.5 minutes. Compared with the traditional fixed-route point-by-point scanning method, the target positioning response time was greatly shortened. Since only 3 UAVs participated in high-precision acquisition and calculation, and the remaining nodes maintained low-power cruise, the overall energy consumption of the formation could be reduced. In urban environments where multipath effects caused by building obstruction were observed, the stability of the positioning results was improved compared with the traditional single-station direction finding method.

[0051] Example 3: like Figure 10As shown, this embodiment is applied to the continuous tracking of illegal mobile radio interference sources in border areas. The target is a vehicle-mounted 400MHz band interference source with a maximum speed of 60km / h, which can move and turn with the road. A tracking formation of 8 multi-rotor UAVs is used to achieve continuous high-precision tracking of the moving target based on the current positioning function.

[0052] First, system hardware configuration: Based on the hardware of Example 2, each UAV is equipped with an additional airborne edge computing unit with a computing power of 1 TOPS, which can complete motion state estimation and trajectory planning locally; the flight control system supports dynamic and smooth trajectory switching, and the attitude adjustment response time is ≤200ms; the positioning reference sampling frequency is 10Hz, the first rate threshold is set to 10km / h, and the second rate threshold is set to 40km / h; the configuration threshold is set to axial proximity of 0.85, and the multipath determination angle threshold is 15°; II. The dynamic tracking process is as follows: 1. Target movement determination and state initialization: In the initial stage, the target is located for the first time according to the fixed signal source positioning process. After outputting three consecutive frames of positioning results, the system calculates the target displacement rate as 25km / h, determines that the target is in a moving state, and starts the dynamic tracking mode. Based on the positioning data of the first three frames, the system initializes the extended Kalman filter, inputs the target position and velocity observation values, estimates the target's motion state vector, and outputs the target position prediction value and 95% confidence interval for the next moment (0.1s later). 2. Dynamic Maintenance of Orthogonal Configuration: The system uses the predicted target location as the new virtual center and calculates the axial proximity of 8 UAVs relative to the predicted center in real time. When the target moves 120m eastward along the road, the axial proximity of the original X-axis positioning UAV drops from 0.94 to 0.82, which is below the configuration threshold of 0.85. The system immediately selects the node with the highest axial proximity from the two UAVs waiting on the east side as a replacement, plans a smooth entry trajectory, and the replacement node moves towards the predetermined station at a speed of 12m / s. The original positioning node simultaneously withdraws to the periphery. The entire switching process lasts for 8 seconds, during which 3 effective orthogonal positioning nodes are always maintained. The positioning is uninterrupted throughout the configuration switching process, and the geometric accuracy factor fluctuation does not exceed 0.3, avoiding a cliff-like drop in positioning accuracy when nodes are switched. 3. Multipath Signal Identification and Correction: When tracking through mountainous reflection areas, the target signal generates multipath components due to reflection from the mountains, significantly increasing the time difference of arrival measurement error. The system utilizes the spatial geometric relationship of three orthogonal positioning UAVs to calculate the arrival direction vector of each signal. One of the reflected signals has an angle of 22° with the direction of the main signal, exceeding the 15° judgment threshold. The system identifies it as a multipath component and removes it. The remaining main path signals are then fused using signal-to-noise ratio weighting. The corrected time difference of arrival error is reduced from 120ns to 25ns, and the positioning error is narrowed from 12.6m to 3.8m, effectively suppressing multipath disturbances in moving scenarios. 4. Tiered power consumption control operation: The tracking mode automatically switches according to the target vehicle speed throughout the tracking process: When the target is traveling at low speed (8km / h), it enters low power consumption mode, with only 3 drones participating in positioning, the sampling frequency is reduced to 5Hz, and the other 5 drones remain hovering at low altitude, greatly reducing the overall power consumption of the formation; when the target is traveling at normal speed (25km / h), it switches to standard mode, with 4 drones positioning and 4 on standby, the sampling frequency is 10Hz, balancing accuracy and endurance; when the target accelerates to escape (55km / h), it triggers high-precision mode, with 6 drones forming a double orthogonal redundant configuration, the sampling frequency is increased to 20Hz, the number of iterations is increased from 2 to 5, and the positioning output delay is controlled within 50ms, ensuring the continuity of tracking under high-speed maneuvering; 5. Signal Loss Reacquisition: When the target enters the tunnel and loses signal, the positioning actuator immediately initiates predictive reacquisition. Based on the movement state before the loss (speed 50km / h, direction due east), the tunnel exit area is extrapolated as a high-probability reacquisition zone. Three standby UAVs are dispatched to perform an involute spiral scan centered on the tunnel exit, with a scanning step of 50m, prioritizing coverage of the predicted path in the due east direction. The target is reacquired 1.2s after exiting the tunnel. The positioning actuator immediately reassembles the orthogonal configuration within one positioning cycle (0.1s) to resume continuous tracking. The tracking interruption time caused by the loss of signal is only 1.3s, which is much lower than the 8-15s of the traditional blind search method. In this embodiment, the average tracking error of mobile radio interference sources is 4.1m, and the tracking error at the highest speed does not exceed 6.5m; the continuous tracking endurance of the formation is increased from 1.8 hours to 2.7 hours; the reacquisition speed after signal loss is greatly improved, realizing high-precision, long-endurance, and high-reliability collaborative tracking of interference sources.

[0053] Example 4: like Figure 11 As shown, this embodiment is designed for long-distance, low-power weak radiation source positioning scenarios. The target is a low-power signal source 10km away, with a received signal-to-noise ratio of about 0dB. The single-moment TDOA positioning error is large and the resolution is insufficient. Three orthogonal configuration UAVs are used to perform positioning. First, parameter configuration: Three UAVs perform uniform small-amplitude maneuvers along the three orthogonal axes X, Y, and Z, respectively, with a maneuver speed of 5m / s, a sampling frequency of 10Hz, a continuous sampling duration of 2s, and generate a total of 20 virtual nodes; the equivalent virtual baseline length is 8 times larger than the physical baseline; the number of sparse Bayesian learning iterations is set to 30.

[0054] Second, the super-resolution positioning implementation process: Three positioning UAVs maintain orthogonal positions and simultaneously translate at a uniform speed of 5m / s along their respective axes, moving 10m each within 2s to form three orthogonal sampling trajectories. One frame of TDOA data is collected every 100ms, corresponding to the UAV position at a given moment, resulting in a total of 20 sets of observation samples. The UAV position corresponding to each set of samples is used as a virtual array node, and the 20 sets of data are combined to form an extended TDOA equation system, which is equivalent to forming a virtual baseline of approximately 80m in length. The extended equation system is solved using a sparse Bayesian learning algorithm to model the target position as a spatially sparse vector. The posterior probability distribution of the position is iteratively updated through Bayesian estimation. At the same time, the spatial symmetry of the orthogonal trajectories is used as a constraint to eliminate spurious peaks caused by multipath reflections. After iterative convergence, the three-dimensional coordinates of the target are output, completing the super-resolution positioning.

[0055] Third, implementation results: Under the condition of 0dB low signal-to-noise ratio, the traditional single-time Chan algorithm has a positioning error of 28.6m and is subject to multipath pseudo-peak interference; after adopting virtual baseline super-resolution, the positioning error is reduced to 6.2m, the positioning resolution is improved by more than 4 times, and the pseudo-peak suppression rate is greatly improved, making it suitable for long-distance and weak signal reconnaissance scenarios.

[0056] like Figure 12 The target radiation source localization actuator shown employs a UAV-based collaborative target radiation source localization method. The actuator has a built-in target radiation source localization system and functions as a localization device or storage medium. The target radiation source localization system includes a signal frequency acquisition module, a signal edge determination module, a target node determination module, and a signal coordinate determination module connected in sequence. The target radiation source localization actuator collaboratively completes the entire process from wide-area search to precise localization. Each UAV corresponding to the target radiation source localization actuator is equipped with the following hardware components: Signal acquisition equipment: adopts radio frequency sensors that support multi-band signal reception, with a sampling rate of not less than 500MS / s, a dynamic range of not less than 110dB, and signal strength indication (RSSI) and arrival time stamp functions; Beidou positioning and timing module: It adopts the Beidou-3 dual-mode timing chip, supports B1I / B2a / B3I frequency bands, has a positioning accuracy better than 10cm, a timing accuracy better than 20ns, and is equipped with an oven-controlled crystal oscillator (OCXO) to maintain local clock stability. Network communication equipment: Software-defined radio (SDR) modules that support multi-hop Mesh networks, with a transmission rate of not less than 50Mbps, a communication radius of not less than 1km, and support for MAVLink / UAVCAN communication protocols; Power management system: It adopts a high-energy-density lithium battery pack, which can support no less than 2 hours of continuous flight and has an adaptive power consumption adjustment function; The ground server is equipped with GPU computing resources and can receive signal data collected by each UAV, as well as the corresponding collection location and timestamp, to complete time difference calculation and radiation source location calculation. The workflow is as follows: First, the signal frequency acquisition module constructs a self-organizing communication network for the target area through UAV swarms and continuously monitors the spectrum environment, acquiring preset characteristics of the target signal source. This allows for rapid discovery of suspected targets in a wide-area search, avoiding blind traversal. In the signal edge determination module, when any UAV detects preset characteristics of the target signal source, the network immediately uses an event-triggered mechanism to collaboratively delineate the signal coverage area, accurately determining the target region of the signal source. This achieves a shift from full-area search to local adaptive focusing, significantly shortening the positioning response time. Subsequently, the target node determination module determines the target node based on the participating boundary detection... The system locates all UAV positions, calculates the virtual center point of the target area, and uses this center point as the origin to determine the positioning UAVs located in three orthogonal directions, constructing the optimal TDOA observation geometry configuration. This significantly reduces the geometric accuracy factor and lays the geometric foundation for high-precision positioning. Finally, the signal coordinate determination module, based on this spatial geometric relationship, uses the time difference of arrival data collected by the selected positioning UAVs to solve the three-dimensional coordinates of the target signal source through the Chan-Taylor combined algorithm. This achieves sub-meter positioning accuracy under the same measurement error, while only selecting a small number of key nodes to participate in high-computation calculations, effectively reducing the overall system energy consumption.

[0057] The above embodiments are merely preferred embodiments of the present invention. Therefore, all equivalent changes or modifications made to the structure, features and principles described in the claims of the present invention are included within the scope of the present invention.

Claims

1. A method for locating a target emitter based on cooperation of unmanned aerial vehicles, characterized in that, Includes the following steps: S1. UAV Network Monitoring Signal Source: By constructing a self-organizing communication network for the target area through UAV formation and continuously monitoring it, the preset characteristics of the target signal source are collected. S2. Target Region of Signal Source: When any drone detects the preset characteristics of the target signal source, the drone formation determines the target region of the signal source by outlining the signal coverage area, as follows: S21. When any one or more drones detect that the strength of the target signal exceeds a preset threshold, it is determined that a suspected target signal has been detected, and a trigger signal is broadcast to nearby drones through a self-organizing communication network. S22. Nearby drones that receive the trigger signal move toward the triggering drone, forming a local high-density cluster around the suspected target signal. S23. Each UAV in the cluster measures the signal-to-noise ratio of the target signal it receives and aggregates the signal-to-noise ratio information to the ground server. S24. Based on the spatial distribution of the signal-to-noise ratio, the ground server determines the location where the signal-to-noise ratio strength is lower than a preset threshold as the signal coverage edge; S25. The UAV located at the edge of the signal coverage hovers or scans, dynamically measures the signal boundary, gradually delineates the coverage outline of the signal source, and determines the area inside the outline as the target area where the target signal source may exist. S3. Establish the spatial geometric relationship between the UAV and the signal source: Based on the positions of all UAVs participating in the boundary exploration, calculate the virtual center point of the target area. Using the virtual center point as the origin, determine the positioning UAVs located in the three orthogonal directions, and determine the spatial geometric relationship between the positioning UAVs and the target signal source. S4. Obtain the three-dimensional coordinates of the signal source: Based on spatial geometric relationships, determine the three-dimensional coordinates of the target signal source.

2. The method of claim 1, wherein: Each UAV is equipped with a communication module and a BeiDou positioning and timing unit. The process of constructing a self-organizing communication network and continuously monitoring the target area is as follows: First, after the UAV is powered on, the communication module automatically loads preset network parameters and initializes the network protocol stack. Then, each UAV periodically broadcasts beacon frames containing its own identifier and status on a preset channel, while listening to the beacon frames of other UAVs. When a UAV receives a beacon frame, it records the identifier and signal strength of the sending UAV and generates a neighbor table. Based on the neighbor table, each UAV exchanges routing information through a self-organizing routing protocol, dynamically establishes a routing table, and forms a self-organizing communication network. During network operation, each drone continuously monitors the link status of the self-organizing communication network. When a link interruption or node change is detected, route repair and topology reconstruction are automatically triggered. The process of acquiring the preset characteristics of the target signal source is as follows: Each UAV receives the time signal from the BeiDou satellite and calibrates and corrects the frequency and phase of the local crystal oscillator in real time to obtain the real-time position coordinates of each UAV; then, the UAV formation performs a distributed wide-area search in the target area according to the preset coverage path, and each UAV continuously monitors the spectrum environment of the target area and detects the preset characteristics of the target signal source in real time; the specific process of real-time calibration and correction of the frequency and phase of the local crystal oscillator is as follows: First, receive the BeiDou pulse and time information sent by the BeiDou satellite and use the BeiDou pulse as the absolute time reference; second, divide the high-frequency clock signal output by the local crystal oscillator to generate a local pulse; The third step is to compare the time difference between the rising edges of the BeiDou pulse and the local pulse using a phase detector, and output the phase difference signal. The fourth step is to process the phase difference signal through a loop filter to generate a control voltage; the fifth step is to apply the control voltage to the control terminal of the voltage-controlled crystal oscillator to adjust the frequency and phase of the local crystal oscillator in real time.

3. The method of claim 1, wherein: The process of calculating the virtual center point of the target area is as follows: First, obtain the position coordinates of all UAVs participating in the boundary exploration and the received signal-to-noise ratio (SNR) of the same target signal measured by each UAV; Second, use the received SNR value of each UAV as the weight of each UAV in the calculation, and sum the SNR weights of all UAVs to obtain the total weight; Third, multiply the longitude, latitude, and altitude values ​​of each UAV by the corresponding SNR weight to obtain the weighted longitude, weighted latitude, and weighted altitude of each UAV, and sum the weighted longitude, weighted latitude, and weighted altitude of all UAVs to obtain the total weighted longitude, total weighted latitude, and total weighted altitude. The fourth step is to divide the total weighted longitude by the total weight to obtain the longitude value of the virtual center point, divide the total weighted latitude by the total weight to obtain the latitude value of the virtual center point, and divide the total weighted altitude by the total weight to obtain the altitude value of the virtual center point. The fifth step is to integrate the longitude, latitude, and altitude values ​​to obtain the three-dimensional coordinates of the virtual center point. 4.The target emitter positioning method based on UAV cooperation according to claim 1, characterized in that: The process of determining the spatial geometric relationship between the positioning UAV and the target signal source is as follows: First, establish a local three-dimensional coordinate system with the virtual center point of the target area as the origin; second, obtain the position coordinates and attitude information of each UAV participating in the exploration in the global coordinate system; third, use the direction cosine matrix to transform the position coordinates and attitude information into the local three-dimensional coordinate system to obtain the three-dimensional direction vector of each UAV relative to the virtual center point; fourth, calculate the axial proximity between the three-dimensional direction vector of each UAV and the three orthogonal coordinate axes in the local three-dimensional coordinate system. The fifth step is to determine the spatial geometric relationship between the positioning drone and the target signal source based on the axial proximity, using the drone closest to the three orthogonal axes as the positioning drone.

5. The method of claim 1, wherein: The process of determining the three-dimensional coordinates of the target signal source is as follows: First, the drone is positioned to simultaneously intercept the signal emitted by the target signal source and record the arrival timestamp of each intercepted signal; The second step is to calculate the initial position of the target signal source based on the arrival timestamp; the third step is to iteratively correct the initial position to obtain the three-dimensional coordinates of the target signal source. The initial position calculation of the target signal source is as follows: First, obtain the three-dimensional coordinates of at least three reference UAVs and determine the reference UAV; The second step is to obtain the time difference of arrival of the target signal source to each reference UAV relative to the reference UAV, and convert the time difference into a distance difference; The third step is to use the distance from the target signal source to the reference UAV as an auxiliary variable, and use the distance difference and the auxiliary variable to transform the nonlinear TDOA hyperbolic equation system into a pseudo-linear equation system about the three-dimensional coordinates of the target signal source and the auxiliary variable. The fourth step is to use the first weighted least squares method to solve the pseudo-linear equation system and obtain the initial solution for the target coordinates and auxiliary variables. The fifth step involves using the geometric constraints between the three-dimensional coordinates of the target signal source and the auxiliary variables to construct a second weighted least squares equation, which corrects the initial solution and yields the initial position of the target signal source.

6. The method of claim 1, wherein: The iterative correction of the initial position is carried out as follows: First, based on the initial position, a system of linear equations about the position correction amount is constructed using Taylor series and solved to obtain the correction amount of the current position; Second, the initial position is added to the correction amount to obtain a candidate new position, and the error index of the candidate new position is calculated. The third step is to accept the candidate new position as the estimated position for the next iteration if the error index of the candidate new position is less than that of the initial position; otherwise, the candidate position is recalculated. The fourth step is to stop the iteration when the correction amount reaches the preset iteration condition and output the three-dimensional coordinates of the target signal source.

7. The method of claim 1, wherein: It also includes dynamic tracking of target radiation sources: When the target signal source is determined to be in a moving state, the motion state parameters of the target are estimated based on the positioning results of multiple consecutive frames; the position area of ​​the target at the next moment is predicted according to the motion state parameters, and the UAV formation is dynamically scheduled to adjust its position, maintaining the orthogonal spatial geometric relationship between the positioning UAV and the target throughout the process. The calculation is repeated according to the preset positioning cycle, and the continuous motion trajectory of the target signal source is output. The dynamic scheduling of drone formations to adjust their positions specifically includes: Extended Kalman filtering is used to perform joint state estimation of the target's position, velocity, and acceleration, generating a predicted target position and confidence interval for the next moment. Using the predicted position as the new virtual center, the axial proximity of each UAV to the virtual center is recalculated. When the axial proximity of the currently positioned UAV is lower than a preset configuration threshold, the replacement node with the highest matching degree is selected from the surrounding standby UAVs. The replacement node is switched into the positioning configuration through a smooth transition trajectory scheduling, and the original positioning node is returned to the standby state. Throughout the switching process, at least three UAVs maintain an orthogonal observation configuration.

8. The method of claim 7, wherein: The target radiation source dynamic tracking also includes orthogonal configuration-assisted multipath signal discrimination and correction, tracking-level power consumption control, and predictive reacquisition of lost signals.

9. The target radiation source localization method based on UAV cooperation according to claim 1, characterized in that, The determination of the three-dimensional coordinates of the target signal source also includes virtual baseline super-resolution enhancement, specifically as follows: control the positioning UAVs in three orthogonal directions to perform small-amplitude uniform maneuvers along their respective axes, and collect the arrival time difference sequence of the target signal at N consecutive sampling times; convert the sampling points at different times into virtual array nodes, and construct a space-time joint long baseline observation equation; A sparse Bayesian learning algorithm is used to perform super-resolution reconstruction of the target position. At the same time, the spatial constraints of orthogonal trajectories are used to suppress multipath pseudo-peaks, and the output resolution is better than the target three-dimensional coordinates sampled at a single time step. The specific steps of constructing the spatiotemporal joint long baseline observation equation include: using the UAV position at the kth sampling time as the virtual node coordinates, and using the TDOA measurement value at that time as the observation value of the corresponding virtual node; By combining the virtual nodes and observations at all times, an extended TDOA equation system is formed; the equivalent length of the virtual baseline is equal to the sum of the UAV maneuver distance and the physical node spacing, and the equivalent array aperture is 5-10 times larger than that at a single time.

10. An unmanned aerial vehicle (UAV) coordination based target emitter positioner, adopting the UAV coordination based target emitter positioning method according to any one of claims 1 to 9. The positioning actuator has a built-in target radiation source positioning system, and the positioning actuator is a positioning device or storage medium; the target radiation source positioning system includes the following components connected in sequence: The signal frequency acquisition module is used to build a self-organizing communication network for the target area through UAV formation and continuously monitor and acquire preset characteristics of the target signal source; The signal edge determination module is used to control the drone formation to delineate the outline of the signal coverage area and determine the target area of ​​the target signal source when any drone detects the preset characteristics of the target signal source. The target node determination module is used to calculate the virtual center point of the target area based on the positions of all UAVs participating in the boundary exploration, and to determine the positioning UAVs located in three orthogonal directions with the virtual center point as the origin, and to determine the spatial geometric relationship between the positioning UAVs and the target signal source. The signal coordinate determination module is used to determine the three-dimensional coordinates of the target signal source based on spatial geometric relationships.

Images