Unmanned aerial vehicle locking monitoring method and system based on cloud control platform

Abstract

The invention relates to the technical field of unmanned aerial vehicle monitoring, and discloses an unmanned aerial vehicle locking monitoring method and system based on a cloud control platform, and the method comprises the following steps: collecting feature data of an unmanned aerial vehicle, and transmitting the feature data to the cloud control platform; the cloud control platform preprocesses the feature data; based on the preprocessed feature data, behavior feature parameters of the unmanned aerial vehicle are extracted; performing abnormal behavior detection on the unmanned aerial vehicle based on the behavior characteristic parameters; acquiring environment data, and performing behavior pattern matching on the unmanned aerial vehicle with the abnormal behavior; illegal unmanned aerial vehicle response and management measures are taken for the unmanned aerial vehicles with illegal matching results; according to the method, the illegal unmanned aerial vehicle identification efficiency and effect are remarkably improved, a systematic and intelligent solution is provided for the unmanned aerial vehicle supervision field, and the method has an important value for promoting standardized development of the unmanned aerial vehicle industry.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle monitoring, and in particular to a method and system for locking and monitoring unmanned aerial vehicles based on a cloud control platform. Background Art

[0002] With the rapid development of science and technology, drone technology has become a booming field worldwide and is widely used in many industries such as agricultural monitoring, topographic mapping, cargo transportation, film and television shooting, emergency rescue, and environmental monitoring. With its advantages of flexibility, cost-effectiveness, and ability to perform dangerous tasks, drones have greatly expanded the boundaries of human activities and promoted the diversified development of social economy. However, the popularization of drone technology has also brought many safety hazards and regulatory challenges. Unauthorized drone flights may disrupt aviation order and cause air traffic accidents. For example, drones illegally approaching airport runways may force flights to be delayed or even canceled, posing a serious threat to aviation safety. In addition, drones may also be used for illegal activities. In view of this, various regions have strengthened the formulation of drone management regulations, requiring drones to be registered and comply with specific flight rules, while increasing the crackdown on illegal drone activities. However, most of the existing regulatory means and technologies rely on human visual observation, ground radar monitoring, or simple electronic fence technology. These methods have obvious deficiencies in coverage, accuracy, response speed, and the ability to identify illegal drones, and are difficult to effectively deal with the growing number of drones and complex and changing usage scenarios. Drones are small in size, fly at low altitudes, and move at high speeds. Traditional radar systems often have difficulty accurately capturing their dynamics, especially in complex urban environments, where obstructions from buildings and terrain make identification even more difficult. Relying solely on a single data source (such as GPS location information) makes it difficult to fully assess whether a drone's flight behavior is abnormal, and lacks the ability to conduct comprehensive multi-dimensional data analysis. Existing monitoring systems do not adequately consider environmental factors, such as the impact of weather changes and the diversity of geographical environments on drone flight characteristics, resulting in high false alarm and missed alarm rates. In summary, although the widespread use of drone technology has brought great value to society, the accompanying safety and regulatory issues cannot be ignored.

[0003] For example, a Chinese patent with authorization announcement number CN108291952B discloses a method and a monitoring system for supervising a UAV and its flight status, wherein the method includes: generating an offline flight certificate for the UAV based on offline flight declaration information, and providing the offline flight certificate to the UAV, thereby limiting the flight behavior of the UAV in the offline flight mode through the offline flight certificate; or, reporting the flight information of the UAV, judging whether the flight information of the UAV complies with the flight safety regulations based on preset safe flight parameters, and generating corresponding flight restriction instructions when judging that the flight information of the UAV violates the flight safety regulations, and sending the flight restriction instructions to the UAV for execution, so as to limit the flight behavior of the UAV in the online flight mode.

[0004] For example, the patent application with publication number CN105912288A discloses a comprehensive processing display method and system for monitoring the flight status of a drone, and establishes a comprehensive processing display interface, which is divided into a fixed display bar interface and a switching display bar interface. The fixed display bar interface is divided into: a drone attitude display sub-interface, a basic flight control sub-interface, and a basic flight status display sub-interface. The switching display bar interface switches between the route planning and photo setting sub-interface, the flight monitoring display sub-interface, and the data analysis sub-interface. Compared with the prior art, the invention can intuitively display the key flight status data of the drone on the flight monitoring display sub-interface through graphics, animations, etc., so that the monitoring personnel can find the required information in time. At the same time, it also has a route planning function and a data comprehensive analysis function combined with photo settings and flight settings.

[0005] The above technical solutions all have the problem raised by the background technology: insufficient consideration of environmental factors leads to high false alarm rate and missed alarm rate.

[0006] The information disclosed in this background technology section is only intended to enhance the understanding of the overall background of the invention and should not be regarded as an acknowledgement or any form of suggestion that the information constitutes the prior art already known to ordinary technicians in this field. Summary of the invention

[0007] The technical problem to be solved by the present invention is to overcome the defects of the prior art, provide a drone locking and monitoring method and system based on a cloud control platform, improve the efficiency and effect of illegal drone identification, and provide a systematic and intelligent solution for the field of drone supervision.

[0008] In order to solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, the present invention provides a method for locking and monitoring a drone based on a cloud control platform, comprising the following steps: S1: Collect characteristic data of the drone and transmit the characteristic data to the cloud control platform; S2: The cloud control platform pre-processes the feature data; S3: Extract the behavior characteristic parameters of the UAV based on the preprocessed feature data; S4: Detecting abnormal behavior of the drone based on the behavior characteristic parameters; S5: Obtain environmental data and perform behavioral pattern matching on drones with abnormal behaviors; S6: Take illegal drone response and management measures for drones that are matched as illegal.

[0009] As a preferred solution of the drone locking monitoring method based on the cloud control platform of the present invention, wherein: the feature data includes three-dimensional coordinate data, communication signal data, and drone images; The three-dimensional coordinate data is obtained through a laser radar; the communication signal data is the communication signal between the drone and the remote controller or the ground station, which is collected through a wireless signal detector; the drone image is obtained through a photoelectric sensor.

[0010] As a preferred solution of the drone locking monitoring method based on the cloud control platform of the present invention, the preprocessing is specifically as follows: Performing data cleaning on the feature data; Perform format calibration and coordinate alignment for each feature data; Arrange the three-dimensional coordinate data into a time series as a coordinate sequence, and draw a flight trajectory diagram based on the coordinate sequence; the flight trajectory diagram is a flight trajectory of the drone composed of coordinate points of the drone, and each coordinate point corresponds to a three-dimensional coordinate data and a corresponding timestamp; Performing frequency domain analysis on the communication signal data to obtain a communication signal spectrum diagram; Perform image enhancement on the drone image.

[0011] As a preferred solution of the drone locking monitoring method based on the cloud control platform of the present invention, wherein: the behavior characteristic parameters include flight trajectory characteristic parameters, flight mode characteristic parameters, communication behavior characteristic parameters, and target shape characteristic parameters; The flight trajectory characteristic parameters include average flight speed, maximum acceleration, and average flight altitude; the extraction method is as follows: The distance between any two adjacent coordinate points in the flight trajectory diagram is calculated based on the three-dimensional coordinate data and divided by the time difference between the two adjacent coordinate points to obtain the average speed between any two adjacent coordinate points; the average of the average speeds between all adjacent coordinate points is calculated to obtain the average flight speed; the three-dimensional coordinate data of each coordinate point in the flight trajectory characteristic parameters is differentiated twice with respect to time to obtain the acceleration of each coordinate point and the maximum value of the acceleration of the coordinate point is extracted as the maximum acceleration; the flight height of each coordinate point in the flight trajectory diagram is extracted and the average is calculated to obtain the average flight height.

[0012] As a preferred solution of the drone locking monitoring method based on the cloud control platform of the present invention, the flight mode characteristic parameters include the frequency and duration of each flight mode; the flight modes include straight flight, turning, circling, ascending, descending, and hovering; the method for extracting the flight mode characteristic parameters is as follows: The trajectory segments corresponding to each flight model are identified from the flight trajectory graph through a machine learning model; the number of times each flight model appears per unit time is counted as the frequency of each flight mode; the total duration of each flight mode per unit time is counted as the duration of each flight mode.

[0013] As a preferred solution of the drone locking monitoring method based on the cloud control platform of the present invention, the communication behavior characteristic parameters include signal strength characteristics and signal frequency characteristics, and the extraction method is as follows: The signal strength mean and variance of the communication signal data are calculated as signal strength features; and the frequency range and center frequency of the communication signal are extracted from the communication signal spectrum diagram as signal frequency features.

[0014] The target appearance characteristic parameters include the model and color of the drone, and the extraction method is as follows: Performing target detection on the drone image, marking the drone area in the drone image and cropping a fixed-size feature map containing the drone area; The feature map is input into the trained convolutional neural network model to output the model and color of the drone.

[0015] As a preferred solution of the drone locking monitoring method based on the cloud control platform of the present invention, the method of abnormal behavior detection is as follows: Collect normal drone feature data and extract behavioral feature parameters as normal samples; normalize and encode the behavioral feature parameters of normal samples to form a behavioral feature vector; the form of any behavioral feature vector is as follows: ; in, represents the behavioral feature vector of the i-th normal sample, the value range of i is 1, 2, ..., m, where m is the number of normal samples; represents the jth normalized and encoded behavior characteristic parameter of the i-th normal sample, and the value range of j is 1, 2, ..., n, where n is the total number of behavior characteristic parameters of any UAV; Construct an abnormal probability model based on Gaussian distribution, the equation is as follows: ; in, express The abnormal probability density function of represents the weight coefficient of the kth Gaussian distribution, the value range of k is 1, 2, ..., K, K is the number of Gaussian distributions in the abnormal probability model; represents the covariance matrix of the k-th Gaussian distribution; express The determinant of ; express The inverse matrix of represents the mean vector of the kth Gaussian distribution; Indicates the transpose of the matrix in brackets; The abnormal probability model is trained as follows: EM algorithm is used for parameter estimation to determine the weight coefficient of each Gaussian distribution , covariance matrix , mean vector The value of and the number K of Gaussian distribution are determined by AIC criterion; the trained abnormal probability model is saved; The behavioral characteristic parameters of the drone to be detected for abnormal behavior are normalized and encoded to form a behavioral characteristic vector, denoted as , and the behavioral feature vector Input the trained anomaly probability model to obtain the abnormal behavior detection results.

[0016] As a preferred solution of the drone locking monitoring method based on the cloud control platform of the present invention, the method of obtaining the abnormal behavior detection result is as follows: Equation calculation based on abnormal probability model The abnormal probability density function is denoted as ; calculate The logarithmic probability function , the formula is as follows: ; like , then the drone does not have abnormal behavior; otherwise, the drone has abnormal behavior; in, Represents the first abnormal threshold, and the calculation formula is as follows: ; in, Represents the behavioral feature vector of the i-th normal sample The logarithmic probability function is calculated as follows: ; Represents the mean of the logarithmic probability function of the behavioral feature vectors of m normal samples, Represents the variance of the logarithmic probability function of the behavior feature vector of m normal samples; Represents the second abnormal threshold, and the calculation formula is as follows: .

[0017] As a preferred solution of the drone locking and monitoring method based on the cloud control platform described in the present invention, wherein: the cloud control platform is configured with a behavior pattern database of the drone; the behavior pattern database contains M reference samples, and the reference samples include legal samples and illegal samples; each reference sample corresponds to a behavior pattern vector, and the behavior pattern vector is spliced ​​by a behavior feature vector and an environmental feature vector; the environmental feature vector of each reference sample is composed of the environmental data corresponding to the reference sample after normalization and encoding processing; the environmental data includes wind speed, wind direction, temperature, relative humidity, visibility, cloud thickness, and flight environment; the flight environment is the geographical environment corresponding to the area where the drone flies, including mountains, hills, plains, plateaus, basins, waters, and urban areas.

[0018] As a preferred solution of the drone locking monitoring method based on the cloud control platform of the present invention, the behavior pattern matching method is as follows: Collect the environmental data of the drone to be matched, normalize and encode it to form an environmental feature vector, and then combine it with the behavioral feature vector to form the behavioral pattern vector of the drone to be matched; Calculating the similarity between the behavior pattern vector of the drone to be matched and the behavior pattern vector of each reference sample in the behavior pattern database; Selecting the N reference samples with the highest similarity as similar samples of the drone to be matched; Calculate the legal matching degree of the drone to be matched, the formula is as follows: ; Among them, R represents the legal matching degree; Indicates the similarity between the behavior pattern vector of the drone to be matched and the behavior pattern vector of the pth similar sample, and the value range of p is 1, 2, ..., N; represents the similarity between the behavior pattern vector of the drone to be matched and the behavior pattern vector of the qth similar sample whose sample type is a legal sample, and the value range of q is 1, 2, ..., Q, where Q represents the number of legal samples in the similar samples; If the legal matching degree R is higher than the legal threshold, the matching result of the drone to be matched is legal; otherwise, the matching result of the drone to be matched is illegal.

[0019] In a second aspect, the present invention provides a drone locking monitoring system based on a cloud control platform, comprising a monitoring locking module, a preprocessing module, a feature extraction module, an anomaly detection module, a data acquisition module, a behavior matching module, and a control module; wherein: The monitoring and locking module is used to locate the drone and continuously collect characteristic data of the drone; The preprocessing module is used to preprocess the characteristic data of the drone; The feature extraction module is used to extract the behavior feature parameters of the UAV based on the preprocessed feature data; The abnormality detection module is used to detect abnormal behavior of the drone based on the behavior characteristic parameters; The data acquisition module is used to collect environmental data of the flight environment of the UAV; The behavior matching module is used to match the behavior patterns of UAVs with abnormal behaviors; The control module is used to take illegal drone response and management measures for drones that are matched as illegal.

[0020] In a third aspect, the present invention provides an electronic device, comprising: a memory for storing instructions; a processor for executing the instructions, so that the device performs operations to implement the drone locking and monitoring method based on a cloud control platform described in the present invention.

[0021] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for locking and monitoring a drone based on a cloud control platform described in the present invention is implemented.

[0022] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: By comprehensively collecting the drone's 3D coordinate data, communication signal data, and image information, this solution constructs a multi-dimensional feature data set, providing a solid foundation for subsequent accurate analysis. Through in-depth analysis of feature data, extracting multi-dimensional behavioral feature parameters such as flight trajectory, mode, communication behavior, and target appearance, this solution can more deeply understand the drone's flight mode and intention, and can effectively identify even complex flight strategies.

[0023] The combined abnormal behavior detection with multiple characteristic parameters can handle the complex distribution in the data, effectively distinguish between normal and abnormal flight modes, and improve the sensitivity and accuracy of detection; incorporating environmental data for behavior pattern matching makes the judgment process closer to the actual situation, reduces misjudgments caused by environmental factors, and improves the robustness and adaptability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them: Figure 1 A flowchart of a method for locking and monitoring a drone based on a cloud control platform provided by the present invention; Figure 2 A schematic diagram of the structure of a drone locking monitoring system based on a cloud control platform provided by the present invention; Figure 3 This is a logic diagram for judging the legality of the drone provided by the present invention. DETAILED DESCRIPTION

[0025] The technical solution of the present invention is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations on the technical solution of the present invention. The embodiments of the present invention and the technical features in the embodiments may be combined with each other unless there is a conflict.

[0026] Example 1 This embodiment introduces a method for monitoring drone locking based on a cloud control platform. Figure 1 , the method comprises the following steps: S1: Collect characteristic data of the drone and transmit the characteristic data to the cloud control platform; The feature data includes three-dimensional coordinate data, communication signal data, and drone images; The three-dimensional coordinate data is obtained by laser radar and can be used to accurately track the position and movement trajectory of the drone, providing a data basis for subsequent analysis of the speed, direction, altitude, speed, heading, etc. of the drone; The communication signal data is the communication signal between the drone and the remote controller or the ground station, which is collected by a wireless signal detector and can assist in identifying the identity of the drone and control instructions, etc.; The drone image is acquired through a photoelectric sensor, such as a visible light camera, an infrared camera, a thermal imager, etc., which can identify the shape, color, thermal characteristics, etc. of the drone.

[0027] S2: The cloud control platform pre-processes the characteristic data; the details are as follows: Performing data cleaning on the characteristic data; removing noise points, outliers and invalid data, such as erroneous radar echoes or interference signals; Perform format calibration and coordinate alignment for each feature data; convert feature data from different sources into a unified time series and coordinate system to ensure synchronization and alignment of feature data from different sources in time and space for subsequent analysis; Arrange the three-dimensional coordinate data into a time series as a coordinate sequence, and draw a flight trajectory diagram based on the coordinate sequence; the flight trajectory diagram is a flight trajectory of the drone composed of coordinate points of the drone, and each coordinate point corresponds to a three-dimensional coordinate data and a corresponding timestamp; Performing frequency domain analysis on the communication signal data to obtain a communication signal spectrum diagram; Perform image enhancement on the drone image; for example, adjust brightness and contrast to better identify features such as size, model, and color of the drone.

[0028] S3: Extract the behavior characteristic parameters of the UAV based on the preprocessed feature data; The behavior characteristic parameters include flight trajectory characteristic parameters, flight mode characteristic parameters, communication behavior characteristic parameters, and target shape characteristic parameters; The flight trajectory characteristic parameters include average flight speed, maximum acceleration, and average flight altitude; the extraction method is as follows: The distance between any two adjacent coordinate points in the flight trajectory diagram is calculated based on the three-dimensional coordinate data and divided by the time difference between the two adjacent coordinate points to obtain the average speed between any two adjacent coordinate points; the average of the average speeds between all adjacent coordinate points is calculated to obtain the average flight speed; the three-dimensional coordinate data of each coordinate point in the flight trajectory characteristic parameters is differentiated twice with respect to time to obtain the acceleration of each coordinate point and the maximum value of the acceleration of the coordinate point is extracted as the maximum acceleration; the flight height of each coordinate point in the flight trajectory diagram is extracted and the average is calculated to obtain the average flight height.

[0029] The flight mode characteristic parameters include the frequency and duration of each flight mode; the flight modes include straight flight, turning, circling, ascending, descending, and hovering; the method for extracting the flight mode characteristic parameters is as follows: Through machine learning models, such as convolutional neural networks, YOLO, SSD, etc., the trajectory segments corresponding to each flight model are identified from the flight trajectory graph; the number of times each flight model appears per unit time is counted as the frequency of each flight mode; the total duration of each flight mode per unit time is counted as the duration of each flight mode.

[0030] The communication behavior characteristic parameters include signal strength characteristics and signal frequency characteristics, and the extraction method is as follows: The signal strength mean and variance of the communication signal data are calculated as signal strength features; and the frequency range and center frequency of the communication signal are extracted from the communication signal spectrum diagram as signal frequency features.

[0031] The target appearance characteristic parameters include the model and color of the drone, and the extraction method is as follows: Performing target detection on the drone image, marking the drone area in the drone image and cropping a fixed-size feature map containing the drone area; The feature map is input into the trained convolutional neural network model to output the model and color of the drone.

[0032] The convolutional neural network model is obtained by fine-tuning and training a pre-trained model (such as VGG, ResNet, etc.) on a drone image set; the drone image set is a large number of feature maps containing drone areas, covering various drone models and colors.

[0033] S4: Detect abnormal behavior of the drone based on the behavior characteristic parameters; the method is as follows: Collect normal drone feature data and extract behavioral feature parameters as normal samples; normalize and encode the behavioral feature parameters of normal samples to form a behavioral feature vector; the form of any behavioral feature vector is as follows: ; in, represents the behavioral feature vector of the i-th normal sample, the value range of i is 1, 2, ..., m, where m is the number of normal samples; represents the jth normalized and encoded behavior characteristic parameter of the i-th normal sample, and the value range of j is 1, 2, ..., n, where n is the total number of behavior characteristic parameters of any UAV; Construct an abnormal probability model based on Gaussian distribution, the equation is as follows: ; in, express The abnormal probability density function of represents the weight coefficient of the kth Gaussian distribution, the value range of k is 1, 2, ..., K, K is the number of Gaussian distributions in the abnormal probability model; represents the covariance matrix of the k-th Gaussian distribution; express The determinant of ; express The inverse matrix of represents the mean vector of the kth Gaussian distribution; Indicates the transpose of the matrix in brackets; The abnormal probability model is trained as follows: EM algorithm is used for parameter estimation to determine the weight coefficient of each Gaussian distribution , covariance matrix , mean vector The value of and the number K of Gaussian distribution are determined by AIC criterion; the trained abnormal probability model is saved; The behavioral characteristic parameters of the drone to be detected for abnormal behavior are normalized and encoded to form a behavioral characteristic vector, denoted as , and the behavioral feature vector Input the trained abnormal probability model to obtain the abnormal behavior detection results; the details are as follows: Equation calculation based on abnormal probability model The abnormal probability density function is denoted as ; calculate The logarithmic probability function , the formula is as follows: ; like , then the drone does not have abnormal behavior; otherwise, the drone has abnormal behavior; in, Represents the first abnormal threshold, and the calculation formula is as follows: ; in, Represents the behavioral feature vector of the i-th normal sample The logarithmic probability function is calculated as follows: ; Represents the mean of the logarithmic probability function of the behavioral feature vectors of m normal samples, Represents the variance of the logarithmic probability function of the behavior feature vector of m normal samples; Represents the second abnormal threshold, and the calculation formula is as follows: .

[0034] S5: Obtain environmental data and perform behavioral pattern matching on drones with abnormal behaviors; The cloud control platform is configured with a behavior pattern database of drones; the behavior pattern database contains M reference samples, and the reference samples include legal samples and illegal samples; each reference sample corresponds to a behavior pattern vector, and the behavior pattern vector is composed of a behavior feature vector and an environment feature vector; the environment feature vector of each reference sample is composed of the environment data corresponding to the reference sample after normalization and encoding processing; the environment data includes wind speed, wind direction, temperature, relative humidity, visibility, cloud thickness, and flight environment; the flight environment is the geographical environment corresponding to the area where the drone flies, including mountains, hills, plains, plateaus, basins, waters, and urban areas.

[0035] Since the behavior patterns and characteristics of drones may be affected by environmental factors, for example, flying in the mountains will cause the drone to fly at a higher altitude and rise and fall more frequently to avoid mountain obstacles; therefore, the abnormal behavior of drones may be caused by environmental factors, and it is necessary to combine environmental data to further judge the legality of drones with abnormal behavior.

[0036] The behavior pattern matching method is as follows: Collect the environmental data of the drone to be matched, normalize and encode it to form an environmental feature vector, and then combine it with the behavioral feature vector to form the behavioral pattern vector of the drone to be matched; Calculate the similarity between the behavior pattern vector of the drone to be matched and the behavior pattern vector of each reference sample in the behavior pattern database; the similarity between the behavior pattern vectors may be calculated using methods such as Euclidean distance and cosine similarity; Selecting the N reference samples with the highest similarity as similar samples of the drone to be matched; Calculate the legal matching degree of the drone to be matched, the formula is as follows: ; Among them, R represents the legal matching degree; Indicates the similarity between the behavior pattern vector of the drone to be matched and the behavior pattern vector of the pth similar sample, and the value range of p is 1, 2, ..., N; represents the similarity between the behavior pattern vector of the drone to be matched and the behavior pattern vector of the qth similar sample whose sample type is a legal sample, and the value range of q is 1, 2, ..., Q, where Q represents the number of legal samples in the similar samples; If the legal matching degree R is higher than the legal threshold, the matching result of the drone to be matched is legal; otherwise, the matching result of the drone to be matched is illegal. The legal threshold is set by technicians in this field based on experience and actual needs. The judgment logic of the legality of the drone in this embodiment is as follows: Figure 3 shown.

[0037] S6: Take illegal drone response and management measures for drones that are matched as illegal.

[0038] The specific response and management measures for illegal drones are as follows: Activate the alarm system and send illegal drone alarm signals, flight trajectory diagrams, and behavior characteristic parameters to relevant departments; Send notifications to illegal drones to slow down or land; use electromagnetic interference technology to block communications between drones and ground control consoles, cut off the drone's remote control signals, data transmission and image transmission signals, and force the drone to enter a self-protection state after signal loss, thereby achieving the purpose of forced landing or driving away; interfere with the drone's satellite navigation signals, making it impossible to accurately locate the drone, and restricting the drone's flight.

[0039] The real-time display of drone locations is achieved through a visual interface, making it easier to continuously track, monitor and manage illegal drones.

[0040] Example 2 This embodiment is the second embodiment of the present invention; it is based on the same inventive concept as embodiment 1, Figure 2 This embodiment introduces a drone locking monitoring system based on a cloud control platform, including a monitoring locking module, a preprocessing module, a feature extraction module, an anomaly detection module, a data acquisition module, a behavior matching module, and a control module; wherein: The monitoring and locking module is used to locate the drone and continuously collect the characteristic data of the drone; it includes a laser radar, a wireless signal detector, and a photoelectric sensor, which are respectively used to collect the three-dimensional coordinate data, communication signal data, and drone images of the drone; The preprocessing module is used to preprocess the characteristic data of the drone, including data cleaning, format calibration, and coordinate alignment; and to sort out the coordinate sequence, draw a flight trajectory diagram; and draw a spectrum diagram of the communication signal; The feature extraction module is used to extract the behavior feature parameters of the UAV, including flight trajectory feature parameters, flight mode feature parameters, communication behavior feature parameters, and target shape feature parameters; The anomaly detection module is used to detect abnormal behavior of the drone; an abnormality detection is performed on the behavior feature vector of the drone through an abnormal probability model to determine whether the drone has abnormal behavior; The data acquisition module is used to collect environmental data of the flight environment of the UAV, including wind speed, wind direction, temperature, relative humidity, visibility, cloud thickness, and flight environment; The behavior matching module is used to match the behavior patterns of drones with abnormal behaviors; the legitimacy of the drone is determined by measuring the similarity between the behavior pattern vector of the drone and the behavior pattern vector of the reference sample in the behavior pattern database and calculating the legal matching degree; The control module is used to respond to and manage illegal drones; including but not limited to sending illegal drone alarm information, sending control signals to drones, such as interference signals that interfere with the flight of illegal drones or notifications of slowing down and landing; and continuously tracking and displaying the location information of illegal drones in real time through the monitoring and locking module.

[0041] The specific functional implementation of each of the above modules refers to the relevant content of the drone locking and monitoring method based on the cloud control platform described in Example 1, and will not be repeated here.

[0042] Example 3 Based on the same inventive concept as other embodiments, this embodiment introduces an electronic device, including a memory and a processor, wherein the memory is used to store instructions and the processor is used to execute the instructions, so that the computer device executes the drone locking and monitoring method based on the cloud control platform provided in the above embodiments.

[0043] Since the electronic device introduced in this embodiment is an electronic device used to implement the drone locking monitoring method based on the cloud control platform in the embodiment of this application, based on the drone locking monitoring method based on the cloud control platform introduced in the embodiment of this application, the technical personnel of this field can understand the specific implementation of the electronic device of this embodiment and its various variations, so how the electronic device implements the method in the embodiment of this application is not introduced in detail here. As long as the technical personnel of this field implement the electronic device used by the drone locking monitoring method based on the cloud control platform in the embodiment of this application, it belongs to the scope of protection of this application.

[0044] Example 4

[0045] Based on the same inventive concept as other embodiments, this embodiment introduces a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the drone locking and monitoring method based on the cloud control platform provided in the above embodiments is implemented.

[0046] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0047] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the purpose and scope of protection of the present invention, which are all within the protection of the present invention.

Claims

1. A method for locking and monitoring a drone based on a cloud control platform, characterized in that: The following steps are involved: S1: Collect characteristic data of the drone and transmit the characteristic data to the cloud control platform; S2: The cloud control platform pre-processes the feature data; S3: Extract the behavior characteristic parameters of the UAV based on the preprocessed feature data; S4: Detecting abnormal behavior of the drone based on the behavior characteristic parameters; S5: Obtain environmental data and perform behavioral pattern matching on drones with abnormal behaviors; S6: Take illegal drone response and management measures for drones that are matched as illegal.

2. The method for locking and monitoring a drone based on a cloud control platform according to claim 1, characterized in that: The feature data includes three-dimensional coordinate data, communication signal data, and drone images; The three-dimensional coordinate data is obtained through a laser radar; the communication signal data is the communication signal between the drone and the remote controller or the ground station, which is collected through a wireless signal detector; the drone image is obtained through a photoelectric sensor.

3. The method for locking and monitoring a drone based on a cloud control platform as claimed in claim 2, characterized in that: The pre-processing is specifically as follows: Performing data cleaning on the feature data; Perform format calibration and coordinate alignment for each feature data; Arrange the three-dimensional coordinate data into a time series as a coordinate sequence, and draw a flight trajectory diagram based on the coordinate sequence; the flight trajectory diagram is a flight trajectory of the drone composed of coordinate points of the drone, and each coordinate point corresponds to a three-dimensional coordinate data and a corresponding timestamp; Performing frequency domain analysis on the communication signal data to obtain a communication signal spectrum diagram; Perform image enhancement on the drone image.

4. The method for locking and monitoring a drone based on a cloud control platform as claimed in claim 3, characterized in that: The behavior characteristic parameters include flight trajectory characteristic parameters, flight mode characteristic parameters, communication behavior characteristic parameters, and target shape characteristic parameters; The flight trajectory characteristic parameters include average flight speed, maximum acceleration, and average flight altitude; the extraction method is as follows: The distance between any two adjacent coordinate points in the flight trajectory diagram is calculated based on the three-dimensional coordinate data and divided by the time difference between the two adjacent coordinate points to obtain the average speed between any two adjacent coordinate points; the average of the average speeds between all adjacent coordinate points is calculated to obtain the average flight speed; the three-dimensional coordinate data of each coordinate point in the flight trajectory characteristic parameters is differentiated twice with respect to time to obtain the acceleration of each coordinate point and the maximum value of the acceleration of the coordinate point is extracted as the maximum acceleration; the flight height of each coordinate point in the flight trajectory diagram is extracted and the average is calculated to obtain the average flight height.

5. The method for locking and monitoring a drone based on a cloud control platform as claimed in claim 4, characterized in that: The flight mode characteristic parameters include the frequency and duration of each flight mode; the flight modes include straight flight, turning, circling, ascending, descending, and hovering; the method for extracting the flight mode characteristic parameters is as follows: The trajectory segments corresponding to each flight model are identified from the flight trajectory graph through a machine learning model; the number of times each flight model appears per unit time is counted as the frequency of each flight mode; the total duration of each flight mode per unit time is counted as the duration of each flight mode.

6. The method for locking and monitoring a drone based on a cloud control platform as claimed in claim 5, characterized in that: The communication behavior characteristic parameters include signal strength characteristics and signal frequency characteristics, and the extraction method is as follows: Calculating the signal strength mean and variance of the communication signal data as signal strength features; extracting the frequency range and center frequency of the communication signal from the communication signal spectrum diagram as signal frequency features; The target appearance characteristic parameters include the model and color of the drone, and the extraction method is as follows: Performing target detection on the drone image, marking the drone area in the drone image and cropping a fixed-size feature map containing the drone area; The feature map is input into the trained convolutional neural network model to output the model and color of the drone.

7. The method for locking and monitoring a drone based on a cloud control platform according to claim 6, characterized in that: The method for abnormal behavior detection is as follows: Collect normal drone feature data and extract behavioral feature parameters as normal samples; normalize and encode the behavioral feature parameters of normal samples to form a behavioral feature vector; the form of any behavioral feature vector is as follows: ； in, represents the behavioral feature vector of the i-th normal sample, the value range of i is 1, 2, ..., m, where m is the number of normal samples; represents the jth normalized and encoded behavior characteristic parameter of the i-th normal sample, and the value range of j is 1, 2, ..., n, where n is the total number of behavior characteristic parameters of any UAV; Construct an abnormal probability model based on Gaussian distribution, the equation is as follows: ； in, express The abnormal probability density function of represents the weight coefficient of the kth Gaussian distribution, the value range of k is 1, 2, ..., K, K is the number of Gaussian distributions in the abnormal probability model; represents the covariance matrix of the kth Gaussian distribution; express The determinant of ; express The inverse matrix of represents the mean vector of the kth Gaussian distribution; Indicates the transpose of the matrix in brackets; The abnormal probability model is trained as follows: EM algorithm is used for parameter estimation to determine the weight coefficient of each Gaussian distribution , covariance matrix , mean vector The value of and the number K of Gaussian distribution are determined by AIC criterion; the trained abnormal probability model is saved; The behavioral characteristic parameters of the drone to be detected for abnormal behavior are normalized and encoded to form a behavioral characteristic vector, denoted as , and the behavioral feature vector Input the trained anomaly probability model to obtain the abnormal behavior detection results.

8. The method for locking and monitoring a drone based on a cloud control platform according to claim 7, characterized in that: The method of obtaining the abnormal behavior detection result is specifically as follows: Equation calculation based on abnormal probability model The abnormal probability density function is denoted as ; calculate The logarithmic probability function , the formula is as follows: ； like , then the drone does not have abnormal behavior; Otherwise, the drone has abnormal behavior; in, Represents the first abnormal threshold, and the calculation formula is as follows: ； in, Represents the behavioral feature vector of the i-th normal sample The logarithmic probability function is calculated as follows: ； Represents the mean of the logarithmic probability function of the behavioral feature vectors of m normal samples, Represents the variance of the logarithmic probability function of the behavior feature vector of m normal samples; Represents the second abnormal threshold, and the calculation formula is as follows: 。 9. The method for locking and monitoring a drone based on a cloud control platform as claimed in claim 8, characterized in that: The cloud control platform is configured with a behavior pattern database of drones; the behavior pattern database contains M reference samples, and the reference samples include legal samples and illegal samples; each reference sample corresponds to a behavior pattern vector, and the behavior pattern vector is composed of a behavior feature vector and an environment feature vector; the environment feature vector of each reference sample is composed of the environment data corresponding to the reference sample after normalization and encoding processing; the environment data includes wind speed, wind direction, temperature, relative humidity, visibility, cloud thickness, and flight environment; the flight environment is the geographical environment corresponding to the area where the drone flies, including mountains, hills, plains, plateaus, basins, waters, and urban areas.

10. The method for locking and monitoring a drone based on a cloud control platform according to claim 9, characterized in that: The behavior pattern matching method is as follows: Collect the environmental data of the drone to be matched, normalize and encode it to form an environmental feature vector, and then combine it with the behavioral feature vector to form the behavioral pattern vector of the drone to be matched; Calculating the similarity between the behavior pattern vector of the drone to be matched and the behavior pattern vector of each reference sample in the behavior pattern database; Selecting the N reference samples with the highest similarity as similar samples of the drone to be matched; Calculate the legal matching degree of the drone to be matched, the formula is as follows: ； Among them, R represents the legal matching degree; Indicates the similarity between the behavior pattern vector of the drone to be matched and the behavior pattern vector of the pth similar sample, and the value range of p is 1, 2, ..., N; represents the similarity between the behavior pattern vector of the drone to be matched and the behavior pattern vector of the qth similar sample whose sample type is a legal sample, and the value range of q is 1, 2, ..., Q, where Q represents the number of legal samples in the similar samples; If the legal matching degree R is higher than the legal threshold, the matching result of the drone to be matched is legal; otherwise, the matching result of the drone to be matched is illegal.

11. A drone locking monitoring system based on a cloud control platform, which is used to implement the drone locking monitoring method based on a cloud control platform as described in any one of claims 1 to 10, characterized in that: It includes monitoring and locking module, preprocessing module, feature extraction module, anomaly detection module, data collection module, behavior matching module and control module; among which: The monitoring and locking module is used to locate the UAV and continuously collect the characteristic data of the UAV; the preprocessing module is used to preprocess the characteristic data of the UAV; The feature extraction module is used to extract the behavior feature parameters of the UAV based on the preprocessed feature data; The abnormality detection module is used to detect abnormal behavior of the drone based on the behavior characteristic parameters; The data acquisition module is used to collect environmental data of the flight environment of the UAV; The behavior matching module is used to match the behavior patterns of UAVs with abnormal behaviors; The control module is used to take illegal drone response and management measures for drones that are matched as illegal.

12. An electronic device, characterized in that: include: A memory for storing instructions; A processor is used to execute the instruction so that the device performs the operation of implementing the drone locking and monitoring method based on the cloud control platform as described in any one of claims 1-10.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the drone locking and monitoring method based on a cloud control platform as described in any one of claims 1 to 10 is implemented.

Images