Multi-mode electromagnetic field distribution surveying and mapping unmanned aerial vehicle system and application method thereof

Abstract

The invention relates to the technical field of unmanned aerial vehicle application, and discloses a multi-mode electromagnetic field distribution surveying and mapping unmanned aerial vehicle system and an application method thereof. The system comprises a dynamic adjustment module, a multi-band acquisition module, a hybrid positioning module, a three-dimensional modeling module and a multi-modal fusion module. The dynamic adjusting module adjusts the distance between the electromagnetic sensor and the rotor wing according to the rotor wing electromagnetic interference suppression rule; the multi-band acquisition module acquires and processes electromagnetic signals; the hybrid positioning module integrates TDOA and AOA algorithms to position the radiation source; the three-dimensional modeling module generates a three-dimensional electromagnetic field intensity distribution diagram; the multi-modal fusion module fuses the visible light image and the electromagnetic data. The system can also evaluate the risk of the radiation source and plan a flight path. The application method comprises the steps of distance adjustment, data acquisition and processing, radiation source positioning, model construction and the like. According to the method, rotor interference is effectively suppressed, the positioning precision is improved, multi-modal data fusion is realized, and the method has a wide application prospect in the fields of electric power inspection, urban electromagnetic environment evaluation and the like.

Description

Technical Field

[0001] The present invention relates to the technical field of UAV applications, and particularly to a multi-modal electromagnetic field distribution mapping UAV system and its application method. Background Art

[0002] With the rapid development of modern technology, the application of electromagnetic fields has become increasingly widespread in various fields, such as communication, power, industrial production, etc. However, the widespread existence of electromagnetic fields has also brought many problems, and it has become crucial to accurately map and analyze their distribution.

[0003] In the traditional field of electromagnetic field mapping, early on, it mainly relied on manual on-site measurements. Workers had to carry heavy measurement equipment and measure the electromagnetic field intensity point by point in complex environments. This method was not only inefficient but also posed great safety risks in some dangerous or inaccessible areas, such as near high-voltage power towers, mountainous areas, etc. For example, near high-voltage transmission lines, manual measurements were extremely vulnerable to strong electromagnetic field interference, posing a potential threat to the physical health of the measurement personnel and also making it difficult to ensure the accuracy of the measurement data.

[0004] Subsequently, some measurement methods based on fixed monitoring stations emerged. Although these monitoring stations can continuously monitor the electromagnetic field intensity within a certain range, the monitoring range is limited and cannot comprehensively cover complex and variable areas. Moreover, the construction and maintenance costs of fixed monitoring stations are relatively high, and it is difficult to flexibly adjust the monitoring positions according to actual needs. For example, during the urban construction process, new electromagnetic radiation sources continuously appear, and fixed monitoring stations cannot effectively monitor these new radiation sources in a timely manner.

[0005] In recent years, with the continuous maturity of UAV technology, UAVs have gradually been applied in the field of electromagnetic field mapping. However, existing UAV electromagnetic field mapping systems still have many deficiencies. On the one hand, the UAV rotors generate electromagnetic interference during rotation, affecting the measurement accuracy of electromagnetic sensors. Although some current systems are aware of this problem, the anti-interference measures adopted have poor effects and cannot fundamentally eliminate the influence of rotor electromagnetic interference on the measurement data. On the other hand, in terms of radiation source positioning, the accuracy of existing positioning algorithms is not high. Some systems only use a single positioning algorithm, such as the positioning algorithm based on time difference of arrival (TDOA) or angle of arrival (AOA), which is easily interfered by environmental factors, resulting in large positioning errors. In addition, most existing UAV electromagnetic field mapping systems only focus on the acquisition and analysis of electromagnetic data and lack the fusion processing of multi-modal data. They cannot effectively combine electromagnetic data with other information such as visible light image data, making it difficult to comprehensively analyze the electromagnetic field distribution from multiple angles and restricting the in-depth understanding and application of the electromagnetic environment.

[0006] In practical application scenarios, such as power facility inspection, it is necessary to accurately understand the electromagnetic field distribution to detect abnormal conditions such as equipment leakage; in urban electromagnetic environment assessment, geographical information and electromagnetic data need to be comprehensively considered to plan electromagnetic sensitive areas. However, existing technical means cannot meet these complex requirements. Therefore, it is urgent to develop a multi-modal electromagnetic field distribution mapping unmanned aerial vehicle system and its application method that can effectively suppress rotor electromagnetic interference, improve the positioning accuracy of radiation sources, and achieve multi-modal data fusion analysis. Summary of the Invention

[0007] The purpose of the present invention is to provide a multi-modal electromagnetic field distribution mapping unmanned aerial vehicle system and its application method to solve the problems raised in the above background technology.

[0008] To achieve the above purpose, the present invention provides the following technical solution: A multi-modal electromagnetic field distribution mapping unmanned aerial vehicle system, the system includes: A dynamic adjustment module, which is used to adjust the distance between the electromagnetic sensor and the unmanned aerial vehicle rotor through a telescopic carbon fiber bracket according to the preset rotor electromagnetic interference suppression rule, and generate a dynamic distance adjustment instruction; A multi-band acquisition module, which is used to control a high-precision three-axis electromagnetic field sensor to acquire electromagnetic field signals within a preset frequency band, and perform anti-interference filtering processing on the acquired original signals to generate a standardized electromagnetic data set; A hybrid positioning module, which is used to perform joint solution of time difference and angle of arrival on the standardized electromagnetic data set based on the TDOA and AOA fusion positioning algorithm, and output the three-dimensional space coordinates of abnormal radiation sources; A three-dimensional modeling module, which is used to perform correlation mapping on the three-dimensional space coordinates and electromagnetic field intensity data, generate a three-dimensional electromagnetic field intensity distribution map by using a spatial interpolation algorithm, and mark the spatial position of abnormal radiation sources; A multi-modal fusion module, which is used to synchronously obtain visible light image data, establish a spatio-temporal correlation matrix of geographical coordinates and electromagnetic parameters, extract multi-dimensional fusion features through a tensor decomposition algorithm, and generate a superimposed analysis model.

[0009] Preferably, the generation of the dynamic distance adjustment instruction includes: Real-time monitor the rotation speed and current parameters of the unmanned aerial vehicle rotor, and calculate the electromagnetic interference intensity distribution generated by the rotor; Determine the minimum safe distance of the electromagnetic sensor according to the preset interference intensity threshold; Dynamically adjust the sensor to above the minimum safe distance by driving the carbon fiber bracket to stretch in sections through a stepper motor.

[0010] Preferably, the anti-interference filtering processing includes: Identify the periodic rotor interference waveform in the original signal, and construct an adaptive notch filter; The wavelet transform algorithm is adopted to separate high-frequency noise and low-frequency electromagnetic signals; The filtered signal is subjected to amplitude normalization processing to generate standardized electromagnetic data calibrated with the root mean square value.

[0011] Preferably, the TDOA and AOA fusion positioning algorithm includes: Extract the time difference of arrival of electromagnetic signals and the phase difference data of multi-antenna arrays, and construct an overdetermined system of equations; Use the weighted least squares method to iteratively solve the overdetermined system of equations to obtain the initial coordinates of the radiation source; Perform spatio-temporal smoothing processing on the initial coordinates through the Kalman filtering algorithm to optimize the positioning accuracy.

[0012] Preferably, the spatial interpolation algorithm includes: Map the electromagnetic intensity data of discrete acquisition points to three-dimensional grid nodes; Use the radial basis function interpolation method to calculate the field strength values in the unsampled area; Combine the terrain elevation data to correct the boundaries of the interpolation results and generate a continuous three-dimensional field strength distribution surface.

[0013] Preferably, the construction of the spatio-temporal correlation matrix includes: Perform georegistration on the visible light image and extract the conversion relationship between pixel coordinates and longitude and latitude; Align the electromagnetic field strength data with the corresponding geographical coordinates by timestamp; Construct a three-dimensional tensor matrix including spatial position, field strength value and image features.

[0014] Preferably, the extraction of the multi-dimensional fusion features includes: Perform high-order singular value decomposition on the spatio-temporal correlation matrix and extract the principal component feature vectors; Encode the terrain features in the visible light image through a convolutional neural network; Perform weighted fusion on the electromagnetic principal component features and the terrain encoded features to generate a joint feature vector.

[0015] Preferably, the system further includes: Construct a radiation source risk assessment model and calculate the regional safety level according to the electromagnetic field strength and the radiation source coordinates; Generate a visualization map with safety level markings and transmit it to the ground control terminal in real time through the wireless communication module.

[0016] Preferably, the system further includes: Embed the electromagnetic field gradient tracking algorithm in the UAV flight path planning to dynamically adjust the flight path to focus on the high field strength area.

[0017] Preferably, the present invention further includes an application method based on the above multi-modal electromagnetic field distribution mapping UAV system, and the method includes: Step S1: According to the preset rotor electromagnetic interference suppression rule, dynamically adjust the distance between the electromagnetic sensor and the UAV rotor through the retractable carbon fiber bracket, and generate and execute a dynamic distance adjustment instruction; Step S2: Control the high-precision three-axis electromagnetic field sensor to collect the electromagnetic field signals within the preset frequency band, perform anti-interference filtering and frequency domain conversion processing on the original signals, and generate a standardized electromagnetic data set; Step S3: Based on the TDOA and AOA fusion positioning algorithm, perform time difference measurement and arrival angle joint resolution on the standardized electromagnetic data set, and output the three-dimensional space coordinates of the abnormal radiation source after iterative optimization; Step S4: Perform spatial correlation mapping on the three-dimensional space coordinates and the corresponding electromagnetic field intensity data, reconstruct the three-dimensional electromagnetic field intensity distribution surface by using the Kriging interpolation algorithm, and mark the spatial position information of the abnormal radiation source; Step S5: Synchronously obtain visible light image data, construct a spatio-temporal correlation matrix of geographical coordinates, electromagnetic parameters and timestamps, extract multi-dimensional fusion features through high-order singular value tensor decomposition, and generate an electromagnetic-optical superposition analysis model.

[0018] Compared with the prior art, the beneficial effects of the present invention are: In terms of suppressing rotor electromagnetic interference, the system demonstrates excellent performance through the dynamic adjustment module. The dynamic adjustment module monitors the rotation speed and current parameters of the UAV rotor in real time, thereby accurately calculating the electromagnetic interference intensity distribution generated by the rotor. Determine the minimum safe distance of the electromagnetic sensor according to the preset interference intensity threshold, and then drive the carbon fiber bracket to expand and contract in segments through the stepper motor, and dynamically adjust the sensor to above the minimum safe distance. This method effectively avoids the influence of rotor electromagnetic interference on the measurement accuracy of the sensor, ensuring that the collected electromagnetic data is true and reliable. Compared with the traditional method, it no longer simply tries to weaken the interference, but reduces the influence of the interference on the measurement from the source, greatly improving the accuracy of the measurement data and providing a solid foundation for subsequent analysis and processing. For example, when performing electromagnetic field mapping near power facilities, the traditional method may cause large deviations in measurement data due to rotor electromagnetic interference, and it is impossible to accurately judge the electromagnetic leakage of power facilities, while the present invention can effectively solve this problem, accurately measure the electromagnetic field intensity, and timely discover potential safety hazards.

[0019] The improvement of the radiation source positioning accuracy is another prominent advantage of this system. The hybrid positioning module adopts the TDOA and AOA fusion positioning algorithm. By extracting the time difference of arrival of electromagnetic signals and the phase difference data of multi-antenna arrays, an overdetermined equation set is constructed. The weighted least squares method is used for iterative solution to obtain the initial coordinates of the radiation source. Then, the Kalman filtering algorithm is used to perform spatio-temporal smoothing processing on the initial coordinates to further optimize the positioning accuracy. This fusion positioning algorithm fully combines the advantages of the TDOA and AOA algorithms, effectively reducing the interference of environmental factors on positioning. In practical applications, whether in complex urban environments or mountainous areas with complex terrains, the three-dimensional spatial coordinates of abnormal radiation sources can be accurately determined. This is crucial for promptly discovering and handling electromagnetic radiation anomalies. For example, in the detection of excessive radiation from communication base stations, the location of the base station with excessive radiation can be quickly located, facilitating the relevant departments to take timely measures for adjustment or repair.

[0020] The multi-modal data fusion analysis is a major highlight of the present invention. The multi-modal fusion module synchronously acquires visible light image data. By constructing a spatio-temporal correlation matrix of geographical coordinates and electromagnetic parameters, a tensor decomposition algorithm is used to extract multi-dimensional fusion features and generate a superimposed analysis model. When constructing the spatio-temporal correlation matrix, the visible light image is georegistered, the conversion relationship between pixel coordinates and longitude and latitude is extracted, the electromagnetic field intensity data is time-stamped aligned with the corresponding geographical coordinates, and a three-dimensional tensor matrix containing spatial position, field strength value, and image features is constructed. Then, the principal component eigenvectors are extracted through high-order singular value decomposition, the terrain features in the visible light image are encoded using a convolutional neural network, and the electromagnetic principal component features and the terrain encoded features are weighted and fused to generate a joint eigenvector. This multi-modal data fusion method enables the system to analyze the electromagnetic environment from multiple dimensions. In the assessment of the urban electromagnetic environment, electromagnetic data can be combined with geographical information, terrain, etc. to comprehensively understand the distribution law of electromagnetic radiation, provide a scientific basis for urban planning, reasonably demarcate electromagnetic sensitive areas, and avoid residents being affected by excessive electromagnetic radiation.

[0021] In addition, the system also has a series of practical expansion functions. A radiation source risk assessment model is constructed. According to the electromagnetic field intensity and the radiation source coordinates, the regional safety level is calculated, and a visualization map with safety level markings is generated, which is transmitted to the ground control terminal in real time through the wireless communication module, facilitating the operators to promptly understand the safety status of the on-site electromagnetic environment and make corresponding decisions. The electromagnetic field gradient tracking algorithm is embedded in the UAV flight path planning to dynamically adjust the flight path to focus on high-field-strength areas, improve the mapping efficiency, and ensure comprehensive and detailed monitoring of key areas. These functions cooperate with each other, making the entire system have higher practicality and application value in the field of electromagnetic field mapping and being able to meet the diverse needs of electromagnetic field mapping in different scenarios. Description of the Drawings

[0022] Figure 1 This is the working principle diagram of the multi-modal electromagnetic field distribution mapping UAV system of the present invention; Figure 2 This is the working flow chart of anti-interference filtering processing; Figure 3 This is the working flow chart of the spatial interpolation algorithm; Figure 4 This is the working principle diagram of the UAV flight path planning. Specific implementation manners

[0023] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0024] Please refer to Figures 1 - 4 , the present invention provides a technical solution: a multi-modal electromagnetic field distribution mapping UAV system, the system includes: a dynamic adjustment module, a multi-band acquisition module, a hybrid positioning module, a three-dimensional modeling module, and a multi-modal fusion module.

[0025] During the operation of the system, the dynamic adjustment module adjusts the distance between the electromagnetic sensor and the UAV rotor by means of a telescopic carbon fiber bracket according to the preset rotor electromagnetic interference suppression rule. In this way, the influence of the electromagnetic interference generated by the rotor on the data collected by the sensor can be effectively reduced, and then a dynamic distance adjustment instruction is generated, laying a foundation for subsequent accurate data collection.

[0026] The multi-band acquisition module is responsible for controlling the high-precision three-axis electromagnetic field sensor to collect the electromagnetic field signals within the preset frequency band. After the original signals are collected, this module will perform anti-interference filtering processing on them to generate a standardized electromagnetic data set, providing high-quality data support for subsequent analysis and processing.

[0027] The hybrid positioning module processes the standardized electromagnetic data set generated by the multi-band acquisition module based on the TDOA and AOA fusion positioning algorithm. By jointly solving the time difference and arrival angle of the electromagnetic signals, the three-dimensional space coordinates of the abnormal radiation source are finally output, so as to realize the accurate positioning of the abnormal radiation source.

[0028] The three-dimensional modeling module correlates and maps the three-dimensional space coordinates output by the hybrid positioning module with the electromagnetic field intensity data. Using the spatial interpolation algorithm, a three-dimensional electromagnetic field intensity distribution map is generated, and the spatial position of the abnormal radiation source is marked on the map, intuitively presenting the distribution of the electromagnetic field.

[0029] The multi-modal fusion module synchronously acquires visible light image data. By establishing a spatio-temporal correlation matrix of geographical coordinates and electromagnetic parameters, it uses a tensor decomposition algorithm to extract multi-dimensional fusion features, and then generates a superposition analysis model to achieve in-depth fusion and analysis of multi-modal data.

[0030] The present invention will be further described below in conjunction with Embodiments 1 to 5: Embodiment 1:

[0031] This embodiment focuses on the generation process of the dynamic spacing adjustment instruction in the dynamic adjustment module. When the drone is running, the system will monitor the rotation speed and current parameters of the drone's rotor in real time. Let the rotation speed of the rotor be n (unit: revolutions per minute), and the current parameter be I (unit: ampere). The electromagnetic interference intensity distribution generated by the rotor is calculated through a specific algorithm. Generally speaking, there is a certain functional relationship between the electromagnetic interference intensity E and the rotor rotation speed n and current I, which can be expressed as E = k 1 n + k 2 I, where k 1 and k 2 are coefficients determined according to the characteristics of the drone's rotor.

[0032] According to the preset interference intensity threshold E th , the minimum safe distance d min of the electromagnetic sensor is determined. When the calculated electromagnetic interference intensity E is greater than the interference intensity threshold E th , it indicates that the current sensor is subject to greater interference and the distance needs to be adjusted.

[0033] The system drives the carbon fiber bracket to expand and contract in segments through a stepper motor. The expansion and contraction amount of each step of the stepper motor is Δd. Assuming that after m steps of adjustment, the distance between the sensor and the rotor reaches d, then d = d0 + mΔd, where d 0 is the initial distance. During the adjustment process, ensure that d is greater than the minimum safe distance d min , so as to dynamically adjust the sensor to a suitable position, reduce electromagnetic interference, and ensure the accuracy of the collected data.

[0034] In an actual application scenario, for example, when conducting electromagnetic field mapping in an urban environment, the environment around the drone is complex and there are many electromagnetic interference sources. At this time, the dynamic adjustment module monitors the rotation speed and current parameters of the rotor in real time. When the drone encounters a change in air flow that causes the rotor rotation speed to increase, the calculated electromagnetic interference intensity increases. Based on the above calculation and adjustment process, the system drives the carbon fiber bracket to extend through a stepper motor, increasing the distance between the electromagnetic sensor and the rotor, successfully avoiding the strong electromagnetic interference area generated by the rotor, making the collected electromagnetic field signals more accurate and reliable, and providing a high-quality data basis for subsequent analysis.

[0035] Embodiment 2: Refer to the appendix Figure 2 , this embodiment details the anti-interference filtering process in the multi-band acquisition module. The anti-interference filtering process includes: First, identify the periodic rotor interference waveform in the original signal and construct an adaptive notch filter based on this.

[0036] Use the wavelet transform algorithm to separate high-frequency noise from low-frequency electromagnetic signals.

[0037] Perform amplitude normalization on the filtered signal.

[0038] Generate standardized electromagnetic data with root mean square value calibration, and then check whether the standardized electromagnetic data meets the requirements. If it meets the requirements, the process ends; if it does not meet the requirements, return to the step of performing amplitude normalization on the filtered signal and operate again.

[0039] After the high-precision three-axis electromagnetic field sensor collects the original signal within the preset frequency band, first identify the periodic rotor interference waveform in the original signal. Through the analysis of signal characteristics, construct an adaptive notch filter. This filter can specifically filter out the periodic rotor interference signal. Let the original signal be x(t), and the signal after being processed by the adaptive notch filter be y 1 (t), whose working principle is based on the adaptive filtering algorithm. By continuously adjusting the parameters of the filter, the rotor interference component in the output signal y 1 (t) is minimized.

[0040] Use the wavelet transform algorithm to separate high-frequency noise from low-frequency electromagnetic signals. Wavelet transform is a time-frequency analysis method that can decompose the signal at different frequency scales. Let the high-frequency component after wavelet transform be H(f), and the low-frequency component be L(f). By reasonably selecting the wavelet basis function and the number of decomposition layers, the high-frequency noise is mainly concentrated in the high-frequency component H(f), while the low-frequency electromagnetic signal is mainly contained in the low-frequency component L(f).

[0041] Perform amplitude normalization on the filtered signal. Let the signal after the above two steps of processing be y 2 (t), whose amplitude is A. For the convenience of subsequent analysis and comparison, it is necessary to normalize its amplitude to a certain range. By calculating the root mean square value RMS of the signal and performing calibration, generate standardized electromagnetic data with root mean square value calibration. The calculation formula of the root mean square value RMS is , where T is the sampling time of the signal. After amplitude normalization, the generated standardized electromagnetic data can more intuitively reflect the true characteristics of the electromagnetic field and provide reliable data support for subsequent positioning and analysis.

[0042] For example, when conducting electromagnetic field mapping near power facilities, the original signal not only contains the electromagnetic field signals generated by power facilities, but also is mixed with the interference generated by the drone rotors and the high-frequency noise of the surrounding environment. Through the anti-interference filtering process of the multi-band acquisition module, these interferences and noises are successfully removed, and accurate standardized electromagnetic data is obtained. These data play a key role in subsequent determining the electromagnetic radiation range and intensity distribution of power facilities, etc., and provide a strong basis for evaluating the impact of power facilities on the surrounding environment.

[0043] Embodiment 3:

[0044] This embodiment focuses on elaborating the TDOA and AOA fusion positioning algorithm in the hybrid positioning module. First, extract the time difference of arrival of electromagnetic signals and the phase difference data of the multi-antenna array from the standardized electromagnetic data set. Let the arrival times of the electromagnetic signals received by different antennas be t 1 , t 2 , then the time difference of arrival Δt = t 2 - t 1 . The phase difference data of the multi-antenna array is set as Δφ. According to these data, an overdetermined system of equations is constructed.

[0045] Use the weighted least squares method to iteratively solve the overdetermined system of equations to obtain the initial coordinates of the radiation source. The goal of the weighted least squares method is to minimize the weighted sum of squares of the errors. Let the error vector be , and the weight matrix be W, then the objective function is . By continuously iteratively adjusting the coordinate values, the objective function J is minimized, so as to obtain the initial coordinates (x 0 , y 0 , z 0 ) of the radiation source.

[0046] Since there may be certain errors in the initial coordinates, the Kalman filtering algorithm is used to perform spatio-temporal smoothing processing on the initial coordinates to optimize the positioning accuracy. The Kalman filtering algorithm is an optimal estimation method based on the state space model. It continuously fuses new measurement data and previous estimation results through two steps of prediction and update to improve the positioning accuracy. Let the state vector be X, the measurement vector be Z, the prediction equation be , and the update equation be , where F k is the state transition matrix, Q k is the process noise covariance matrix, K k is the Kalman gain, and H k is the observation matrix. After being processed by the Kalman filtering algorithm, more accurate three-dimensional space coordinates of the abnormal radiation source can be obtained, improving the positioning accuracy of the abnormal radiation source.

[0047] For example, when positioning the radiation source of a communication base station, the hybrid positioning module uses the collected electromagnetic signal data and accurately calculates the three-dimensional space coordinates of the communication base station through the TDOA and AOA fusion positioning algorithm. In the city, communication base stations are densely distributed, and traditional positioning methods are easily interfered. The fusion positioning algorithm of the present invention overcomes the interference problem by considering multiple aspects of information such as time difference of arrival and phase difference, and combines the Kalman filtering algorithm for optimization, realizing high-precision positioning of the radiation source of the communication base station and providing accurate data support for communication network optimization and electromagnetic environment assessment.

[0048] Embodiment 4: Refer to the appendix Figure 3 , and this embodiment details the spatial interpolation algorithm in the three-dimensional modeling module. The process of the spatial interpolation algorithm includes: First, obtain the electromagnetic intensity data of discrete acquisition points.

[0049] Map the electromagnetic intensity data of discrete acquisition points to three-dimensional grid nodes.

[0050] Use the radial basis function interpolation method to calculate the field strength value of the unsampled area.

[0051] Obtain the terrain elevation data and correct the boundary of the interpolation result in combination with the terrain elevation data.

[0052] Judge whether the boundary correction meets the requirements. If it meets the requirements, generate a continuous three-dimensional field strength distribution surface; if it does not meet the requirements, return to the step of correcting the boundary of the interpolation result in combination with the terrain elevation data for reprocessing.

[0053] When performing three-dimensional modeling, first map the electromagnetic intensity data of discrete acquisition points to three-dimensional grid nodes. Let the electromagnetic intensity data of the acquisition points be E i (i = 1, 2,..., n), and the three-dimensional grid node coordinates are (x j , y j , z j )(j = 1, 2,..., m). Through a specific mapping rule, the acquisition point data is assigned to the corresponding grid nodes.

[0054] Use the radial basis function interpolation method to calculate the field strength value of the unsampled area. The radial basis function interpolation method is a commonly used interpolation method. Let the radial basis function be φ(r), where r is the distance. Generally, a Gaussian function or the like can be selected as the radial basis function. For the unsampled point (x, y, z), the calculation formula for its field strength value E(x, y, z) is , where λ i is the weight coefficient, which is determined by solving a system of linear equations.

[0055] The boundary of the interpolation result is corrected by combining terrain elevation data to generate a continuous three-dimensional field strength distribution surface. Let the terrain elevation data be h(x, y). Considering the influence of terrain on the propagation of electromagnetic fields, the field strength values obtained by interpolation are corrected. For example, in mountainous areas and other regions with complex terrain, the propagation of electromagnetic fields is affected by terrain undulations. By combining terrain elevation data, the field strength values in areas close to mountains are adjusted, making the generated three-dimensional field strength distribution surface more in line with the actual situation and accurately reflecting the distribution characteristics of electromagnetic fields under different terrain conditions.

[0056] For example, when mapping the electromagnetic field distribution of power transmission lines in mountainous areas, the three-dimensional modeling module uses a spatial interpolation algorithm to convert discrete collected point data into a continuous three-dimensional field strength distribution surface. Considering the complex terrain in mountainous areas, through boundary correction by combining terrain elevation data, the changes in electromagnetic fields in different terrain areas such as valleys and hillsides are successfully presented. This provides an intuitive and accurate basis for evaluating the impact of power transmission lines on the ecological environment and residents' lives in mountainous areas, and helps to formulate reasonable protection measures and planning schemes.

[0057] Example 5:

[0058] This example mainly introduces the construction of the spatio-temporal correlation matrix in the multi-modal fusion module, the extraction of multi-dimensional fusion features, and the implementation of other functions of the system.

[0059] When constructing the spatio-temporal correlation matrix, first perform georegistration on the visible light image to extract the conversion relationship between pixel coordinates and longitude and latitude. Let the pixel coordinates be (u, v), and the longitude and latitude coordinates be (φ, λ). Through the georegistration algorithm, the conversion formula (φ, λ)=f(u, v) is obtained, where f is the conversion function.

[0060] Align the electromagnetic field strength data with the corresponding geographical coordinates by timestamp. Let the electromagnetic field strength data be E(t), the geographical coordinates be (φ(t), λ(t)), and the timestamp be t, ensuring the accurate association of the electromagnetic field strength data at each moment with the corresponding geographical coordinates.

[0061] Construct a three-dimensional tensor matrix containing spatial position, field strength value, and image features. This tensor matrix fuses multi-modal data and provides a basis for subsequent analysis.

[0062] When extracting multi-dimensional fusion features, perform high-order singular value decomposition on the spatio-temporal correlation matrix to extract the principal component eigenvectors. High-order singular value decomposition can decompose the tensor matrix into the product of multiple low-dimensional matrices, thereby extracting the main feature information. Let the spatio-temporal correlation matrix be T. After high-order singular value decomposition, the principal component eigenvectors U 1 、U 2 、U 3 etc.

[0063] Encode the terrain features in the visible light image through a convolutional neural network. The convolutional neural network has powerful feature extraction capabilities and can automatically learn the terrain features in the image. Let the output of the convolutional neural network be F, which contains rich terrain feature information.

[0064] Weightedly fuse the electromagnetic principal component features and the terrain encoding features to generate a joint feature vector. Let the electromagnetic principal component feature vector be E, the terrain encoding feature vector be T, and the weighting coefficients be ω 1 and ω 2 , then the joint feature vector J = ω 1 E + ω 2 T.

[0065] In addition, the system also constructs a radiation source risk assessment model to calculate the regional safety level based on the electromagnetic field intensity and the radiation source coordinates. Let the electromagnetic field intensity be E, the radiation source coordinates be (x, y, z), and calculate the regional safety level R through the risk assessment model R = g(E, x, y, z), where g is the risk assessment function. Generate a visualization map with safety level markings and transmit it to the ground control terminal in real time through the wireless communication module, facilitating the operators to timely understand the electromagnetic field distribution and safety conditions in the surveyed area.

[0066] Refer to the appendix Figure 4 , which is the flowchart for the UAV flight path planning. Initialize the system, plan the initial flight path of the UAV, and embed the electromagnetic field gradient tracking algorithm in the UAV flight path planning. During the flight, monitor the electromagnetic field intensity data and determine whether to adjust the trajectory according to the electromagnetic field gradient tracking algorithm: if so, dynamically adjust the flight path to focus on the high-field-strength area; otherwise, fly according to the original flight path. Let the electromagnetic field gradient be , and adjust the flight direction and speed of the UAV according to the direction and magnitude of the electromagnetic field gradient, so that the UAV can fly towards the high-field-strength area, improving the surveying and mapping efficiency and accuracy.

[0067] For example, when conducting multi-modal electromagnetic field distribution surveying and mapping of a chemical industrial park, the multi-modal fusion module realizes the comprehensive analysis of the electromagnetic environment and terrain and landform of the chemical industrial park by constructing a spatio-temporal correlation matrix and extracting multi-dimensional fusion features. At the same time, using the radiation source risk assessment model and the visualization map, timely discover potential electromagnetic safety hazard areas, and through the electromagnetic field gradient tracking algorithm in the UAV flight path planning, guide the UAV to conduct detailed surveying and mapping of key areas, providing comprehensive and accurate data support for the safety management and environmental assessment of the chemical industrial park.

[0068] It should be noted that in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or elements inherent to such process, method, article or device.

[0069] Although embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A multi-modal electromagnetic field distribution mapping UAV system, characterized in that: The system comprises: A dynamic adjustment module is used to adjust the distance between the electromagnetic sensor and the UAV rotor through a retractable carbon fiber bracket according to the preset rotor electromagnetic interference suppression rules, and generate a dynamic distance adjustment instruction; The multi-band acquisition module is used to control the high-precision three-axis electromagnetic field sensor to collect electromagnetic field signals within a preset frequency band, and to perform anti-interference filtering on the collected original signals to generate a standardized electromagnetic data set; A hybrid positioning module is used to perform a joint solution of the time difference and the angle of arrival on the standardized electromagnetic data set based on the TDOA and AOA fusion positioning algorithm, and output the three-dimensional spatial coordinates of the abnormal radiation source; A three-dimensional modeling module is used to associate and map the three-dimensional spatial coordinates with the electromagnetic field intensity data, generate a three-dimensional electromagnetic field intensity distribution map using a spatial interpolation algorithm, and mark the spatial position of the abnormal radiation source; The multimodal fusion module is used to synchronously acquire visible light image data, establish the spatiotemporal correlation matrix of geographic coordinates and electromagnetic parameters, extract multidimensional fusion features through tensor decomposition algorithm, and generate a superposition analysis model.

2. The multi-modal electromagnetic field distribution mapping UAV system according to claim 1, characterized in that: The generation of the dynamic spacing adjustment instruction includes: Monitor the speed and current parameters of the drone rotor in real time, and calculate the electromagnetic interference intensity distribution generated by the rotor; Determine the minimum safe distance between electromagnetic sensors according to a preset interference intensity threshold; The stepper motor drives the carbon fiber bracket to extend and retract in sections, and dynamically adjusts the sensor to above the minimum safety distance.

3. The multi-modal electromagnetic field distribution mapping UAV system according to claim 1, characterized in that: The anti-interference filtering process includes: Identify the periodic rotor interference waveform in the original signal and construct an adaptive notch filter; Wavelet transform algorithm is used to separate high-frequency noise and low-frequency electromagnetic signals; The filtered signal is amplitude normalized to generate RMS-calibrated standardized electromagnetic data.

4. The multi-modal electromagnetic field distribution mapping UAV system according to claim 1, characterized in that: The TDOA and AOA fusion positioning algorithm includes: Extract the arrival time difference of electromagnetic signals and the phase difference data of multi-antenna arrays to construct an overdetermined set of equations; Iteratively solving the overdetermined equations using a weighted least squares method to obtain initial coordinates of the radiation source; The initial coordinates are smoothed in time and space by the Kalman filter algorithm to optimize the positioning accuracy.

5. The multi-modal electromagnetic field distribution mapping UAV system according to claim 1, characterized in that: The spatial interpolation algorithm includes: Mapping electromagnetic intensity data of discrete acquisition points to three-dimensional grid nodes; The radial basis function interpolation method is used to calculate the field strength value of the unsampled area; The interpolation results are corrected by combining terrain elevation data to generate a continuous three-dimensional field intensity distribution surface.

6. The multi-modal electromagnetic field distribution mapping UAV system according to claim 1, characterized in that: The construction of the spatiotemporal correlation matrix includes: Perform georeferencing on visible light images and extract the conversion relationship between pixel coordinates and longitude and latitude; Align the electromagnetic field intensity data with the corresponding geographic coordinates by time stamp; Construct a three-dimensional tensor matrix containing spatial position, field strength value and image characteristics.

7. The multi-modal electromagnetic field distribution mapping UAV system according to claim 1, characterized in that: The extraction of the multi-dimensional fusion features includes: Performing high-order singular value decomposition on the spatiotemporal correlation matrix to extract principal component eigenvectors; Encoding terrain features in visible light images through convolutional neural networks; The electromagnetic principal component features and terrain coding features are weightedly fused to generate a joint feature vector.

8. The multi-modal electromagnetic field distribution mapping UAV system according to claim 1, characterized in that: The system further comprises: Construct a radiation source risk assessment model to calculate the regional safety level based on the electromagnetic field intensity and radiation source coordinates; Generate a visual map with security level annotations and transmit it to the ground control terminal in real time through the wireless communication module.

9. The multi-modal electromagnetic field distribution mapping UAV system according to any one of claims 1 to 8, characterized in that: The system further comprises: An electromagnetic field gradient tracking algorithm is embedded in the UAV flight path planning to dynamically adjust the track to focus on the high field strength area.

10. An application method of the multi-modal electromagnetic field distribution mapping UAV system according to any one of claims 1 to 9, characterized in that: include: Step S1: According to the preset rotor electromagnetic interference suppression rules, the distance between the electromagnetic sensor and the UAV rotor is dynamically adjusted through the retractable carbon fiber bracket, and a dynamic distance adjustment instruction is generated and executed; Step S2: Control the high-precision three-axis electromagnetic field sensor to collect electromagnetic field signals within a preset frequency band, perform anti-interference filtering and frequency domain conversion processing on the original signal, and generate a standardized electromagnetic data set; Step S3: Based on the TDOA and AOA fusion positioning algorithm, the standardized electromagnetic data set is subjected to time difference measurement and arrival angle joint solution, and the three-dimensional spatial coordinates of the abnormal radiation source are output after iterative optimization; Step S4: spatially correlate and map the three-dimensional spatial coordinates with the corresponding electromagnetic field intensity data, reconstruct the three-dimensional electromagnetic field intensity distribution surface using the Kriging interpolation algorithm, and mark the spatial position information of the abnormal radiation source; Step S5: synchronously acquire visible light image data, construct a spatiotemporal correlation matrix of geographic coordinates, electromagnetic parameters and timestamps, extract multidimensional fusion features through high-order singular value tensor decomposition, and generate an electromagnetic-optical superposition analysis model.

Images