Detection and positioning method and device for urban low-altitude micro UAVs

Abstract

The present invention provides a detection and positioning method for urban low-altitude micro UAVs, including: controlling monitoring devices to perform scans to obtain first signals to form a data sample set; establishing a behavior fingerprint feature library according to the data sample set and the simulated data completion method; controlling each monitoring device to perform scans and obtain first scan signals and warning signals, and preliminarily determining whether a UAV exists according to the first scan signals and the behavior fingerprint database; if it is determined that a UAV exists, determining a number of collaborative monitoring devices according to the first monitoring position information, and controlling them to monitor a first area to obtain corresponding first monitoring signals; and obtaining a first monitoring result according to the first monitoring signals and a deep neural network; wherein the monitoring devices include radio devices, optical devices, and acoustic devices. This technical solution solves the problems of how to monitor and position weak-signal UAVs and the scheduling of collaborative monitoring devices; and improves the accuracy of monitoring and positioning of UAVs.

Description

Technical Field

[0001] The present invention relates to the field of detection and positioning of unmanned aerial vehicles, and particularly to a method and device for detecting and positioning low-altitude micro unmanned aerial vehicles in cities. Background Art

[0002] The detection and positioning of low-altitude unmanned aerial vehicles in cities is a current hot issue. However, there is currently little work on the detection and positioning of low-altitude micro unmanned aerial vehicles. In fact, micro unmanned aerial vehicles are rapidly moving from military use to civilian use, quickly entering the market and being sought after by many individual players. At the same time, since micro unmanned aerial vehicles do not need to be registered for identity, to a certain extent, they avoid the constraints of unmanned aerial vehicle regulations, and the potential hazards to residents' safety, personal privacy, etc. cannot be ignored.

[0003] However, due to the extremely small size, extremely weak radio signals, and continuous conversion of the activity range around urban buildings and indoors, micro unmanned aerial vehicles pose great challenges to existing unmanned aerial vehicle detection and positioning technologies. It is very difficult to identify micro unmanned aerial vehicles by optoelectronic detection, and the field of view of optoelectronic detection is very limited around buildings and in indoor environments, so it is not practical; radar detection is likely to misjudge micro unmanned aerial vehicles as other things such as birds, and the clutter is too strong and the radiation is too large to be used in crowded places such as around urban buildings and indoors. Traditional radio detection technologies (such as TOA, TDOA, AOA, RSS, etc.) are also very difficult to detect and position micro unmanned aerial vehicles due to limitations such as weak signal strength, strong multipath effect, unstable time difference accuracy, and instantaneous switching of signal fading models. Therefore, existing single technologies are difficult to solve the problem of detecting and positioning micro unmanned aerial vehicles in the area around urban buildings.

[0004] For this reason, the present invention proposes a method for detecting and positioning low-altitude micro unmanned aerial vehicles in cities. Based on the multi-dimensional data analysis and judgment of micro unmanned aerial vehicles, by comprehensively scheduling various unmanned aerial vehicle monitoring means and dynamically allocating tracking and monitoring resources, it breaks through the traditional means of traditional mathematical analysis of single-dimensional data, and performs intelligent interpretation of multi-dimensional data based on a deep neural network to achieve the detection and positioning of low-altitude micro unmanned aerial vehicles in cities, and solve the problems of continuous monitoring and scheduling scheme and data analysis of weak-signal unmanned aerial vehicles in high-frequency dynamic scenarios. Summary of the Invention

[0005] The present invention provides a method and device for detecting and positioning low-altitude micro unmanned aerial vehicles in cities to solve the problems of relatively accurate monitoring and positioning of weak-signal unmanned aerial vehicles (in high-frequency dynamic scenarios) and the scheduling of collaborative monitoring devices.

[0006] According to a first aspect of the present invention, there is provided a method for detecting and positioning low-altitude micro unmanned aerial vehicles in cities, including:

[0007] Control the monitoring device to perform a scan and obtain a first signal to form a data sample set; the first signal characterizes the signal of the drone detected when the monitoring device performs a scan;

[0008] Process the data sample set using the simulated data completion method to establish a behavior fingerprint feature library; the behavior fingerprint feature library includes the data sample set;

[0009] Control each monitoring device to scan at a first scan frequency point and obtain a first scan signal and a warning signal until each monitoring device has exhausted all first scan frequency points; the warning signal characterizes the signal sent by the corresponding monitoring device when the intensity of the first scan signal is greater than the set intensity threshold; the first scan frequency point characterizes any data in the behavior fingerprint feature library; the warning signal includes first monitoring position information; the first monitoring position information characterizes the position information of the monitoring device that sends the warning signal; the warning signal includes the corresponding first scan signal;

[0010] Preliminarily determine whether a drone exists based on the first scan signal and the behavior fingerprint database; if it is determined that a drone exists, determine a number of collaborative monitoring devices based on the first monitoring position information, and control the number of collaborative monitoring devices to monitor a first area and obtain corresponding first monitoring signals; the first area characterizes the area where the monitoring device that sends the warning signal is located; and

[0011] Process the first monitoring signal using a deep neural network to obtain a first monitoring result from the deep neural network; wherein, the monitoring device includes a radio device, an optical device, and an acoustic device.

[0012] Optionally, the first signal includes corresponding radio signals, optical signals, and acoustic signals.

[0013] Optionally, the simulated data completion method includes a generative adversarial network method, a sample feature clustering analysis method, and an interpolation method.

[0014] Optionally, using the simulated data completion method to process the data sample set to establish a behavior fingerprint feature library specifically includes:

[0015] Use the generative adversarial network method to expand the data sample set to generate a number of generated values;

[0016] Use the sample clustering analysis method to analyze the data sample set and eliminate abnormal first signals;

[0017] Filling the data sample set by using the interpolation method to form filled values; the behavior fingerprint feature library includes fingerprint feature values; the behavior fingerprint feature values include the generated values, the remaining first signals after removing the abnormal first signals, and the filled values.

[0018] Optionally, the first scanning signal includes a first acoustic scanning signal and a first radio scanning signal.

[0019] Optionally, controlling each of the monitoring devices to scan at a first scanning frequency point and obtain a first scanning signal and a warning signal includes:

[0020] Determining a first number of the first scanning frequency points for each of the monitoring devices; wherein, the frequency point values of each of the first scanning frequency points of any one of the monitoring devices are different from each other;

[0021] Controlling each of the monitoring devices to sequentially scan at the first number of the first scanning frequency points to obtain corresponding several first scanning signals until each of the monitoring devices has exhausted all of the first scanning frequency points.

[0022] Optionally, the scanning period when controlling each of the monitoring devices to sequentially scan at the first number of the first scanning frequency points is a first scanning period.

[0023] Optionally, specifically determining a first number of the first scanning frequency points for each of the monitoring devices includes:

[0024] Determining the first number of the first scanning frequency points of a first monitoring device; the first monitoring device represents any one of the monitoring devices;

[0025] Determining the first number of the first scanning frequency points of a second monitoring device; the second monitoring device represents the monitoring devices within a first distance around the first monitoring device; the frequency point values of the first scanning frequency points of the first monitoring device are different from those of the first scanning frequency points of the second monitoring device;

[0026] Determining the first number of the first scanning frequency points of a third monitoring device; the third monitoring device represents the monitoring devices within the first distance around the second monitoring device; wherein, the frequency point values of the first scanning frequency points of the third device are different from those of the first scanning frequency points of the second device;

[0027] And so on until determining the first number of the first scanning frequency points for each of the monitoring devices.

[0028] Optionally, the first distance is 100m.

[0029] Optionally, the first scanning frequency points include fixed - period scanning frequency points and non - fixed - period scanning frequency points;

[0030] The fixed - period scanning frequency points represent the fingerprint feature values with the most concentrated probability; the non - fixed - period scanning frequency points represent the other frequency points outside the fixed - period scanning frequency points.

[0031] Optionally, controlling each of the monitoring devices to sequentially scan with the first number of the first scanning frequency points to obtain corresponding several first scanning signals until each of the monitoring devices has exhausted all of the first scanning frequency points specifically includes:

[0032] Controlling the monitoring devices to scan with the first number of the fixed - period scanning frequency points to obtain the first scanning signals until each of the monitoring devices has scanned all of the fixed - period scanning frequency points;

[0033] Controlling the monitoring devices to scan with the first number of the non - fixed - period scanning frequency points to obtain the first scanning signals; until each of the monitoring devices has scanned all of the non - fixed - period scanning frequency points.

[0034] Optionally, initially determining whether a drone exists based on the first scanning signals and the behavior fingerprint database specifically includes:

[0035] Determining a first correlation degree and a second correlation degree based on the first scanning signals and the behavior fingerprint database; the first correlation degree represents the correlation degree between the first scanning signals and the corresponding first signals, and the second correlation degree represents the correlation degree between the first scanning signals and the corresponding generated values and filled values;

[0036] When the first correlation degree is greater than a first preset correlation degree, or the second correlation degree is greater than the second preset correlation degree, initially determine that a drone exists.

[0037] Optionally,

[0038] The first monitoring signals include: first optical monitoring signals, first acoustic monitoring signals, and first radio monitoring signals.

[0039] Optionally, determining several cooperative monitoring devices according to the first monitoring position information, and controlling the several cooperative monitoring devices to monitor a first area to obtain corresponding first monitoring signals, specifically includes:

[0040] Controlling the acoustic devices and the radio devices to scan with the corresponding first scanning signals and obtain the first acoustic monitoring signals and the first radio monitoring signals; and

[0041] Control the optical device to monitor the first area and obtain the first optical monitoring signal.

[0042] Optionally, the first monitoring result includes an identification result and a position result; the identification result represents a judgment result on whether a drone exists; the position result represents the positioning information of the drone.

[0043] Optionally, the first monitoring result is obtained based on the first monitoring data and a deep neural network, specifically including:

[0044] Obtain a first situation map based on the first monitoring data;

[0045] Obtain the first monitoring result based on the first situation map and the deep neural network.

[0046] Optionally, obtaining a first situation map based on the first monitoring data; specifically including:

[0047] Obtain a first situation map based on the first acoustic monitoring data and the first optical monitoring data in the first monitoring data.

[0048] Optionally, before obtaining a first situation map based on the first monitoring data, it further includes;

[0049] Perform a feasibility interpretation on the first optical monitoring signal to obtain a first interpretation feasibility;

[0050] Compare the first interpretation feasibility with a preset feasibility; if the first interpretation feasibility is not lower than the preset feasibility, then eliminate the incorrect first acoustic monitoring signal and / or the first radio monitoring signal.

[0051] According to a second aspect of the present invention, there is provided a device for a detection and positioning method of an urban low-altitude micro and small drone, the device includes:

[0052] A sample data set acquisition module; used to control a monitoring device to perform scanning and obtain a first signal to form the sample data set;

[0053] A behavior fingerprint library establishment module; used to establish a behavior fingerprint library according to the sample data set and the simulation data complement method;

[0054] A signal detection and warning module; used to control each of the control monitoring devices to perform scanning at a first scanning frequency point to obtain corresponding first scanning signals and warning signals;

[0055] Collaborative monitoring device scheduling and monitoring data acquisition module; used to determine whether a drone exists according to the first scanning signal and the behavior fingerprint database, and determine collaborative monitoring devices according to the warning signal, and control the collaborative monitoring devices to monitor a first area and obtain first monitoring data;

[0056] Identification and positioning module; used to obtain a first monitoring result according to a deep neural network and the first monitoring data.

[0057] According to a third aspect of the present invention, there is provided an electronic device including a processor and a memory, where the memory is used to store code;

[0058] The processor is used to execute the code in the memory to implement the method for detecting and positioning a low-altitude micro unmanned aerial vehicle in the city according to the first aspect of the present invention.

[0059] According to a fourth aspect of the present invention, there is provided a storage medium on which a computer program is stored, and when the program is executed by a processor, it implements the method for detecting and positioning a low-altitude micro unmanned aerial vehicle according to any one of the first aspect of the present invention.

[0060] A method for detecting and positioning a low-altitude micro unmanned aerial vehicle provided by the present invention obtains corresponding first monitoring signals by scheduling collaborative monitoring devices according to a first monitoring position; and performs data analysis based on a deep neural network on the first monitoring signals to obtain a more reliable first monitoring result; since the collaborative monitoring devices among them are various types of monitoring devices, the corresponding first monitoring signals obtained are multi-dimensional signal data; and the analysis of multi-dimensional data signals based on a deep neural network is more accurate and reliable for detecting and positioning weak-signal drones in a high-frequency dynamic scenario than using a single monitoring data in the prior art; at the same time, determining and scheduling collaborative monitoring devices to monitor a first area according to the first monitoring position in the warning signal; realizing the scheduling of collaborative monitoring devices in continuous monitoring. It can be seen that the technical solution provided by the present invention solves the problems of how to relatively accurately monitor and position weak-signal drones (in a high-frequency dynamic scenario) and the scheduling of collaborative monitoring devices; improving the accuracy of monitoring and positioning of low-altitude micro unmanned aerial vehicles. Description of the Drawings

[0061] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention, and those of ordinary skill in the art can also obtain other drawings without creative efforts based on these drawings.

[0062] Figure 1It is a grid deployment diagram of a detection and positioning method for urban low-altitude micro UAVs;

[0063] Figure 2 It is a schematic flowchart of a detection and positioning method for urban low-altitude micro UAVs in an embodiment of the present invention;

[0064] Figure 3 It is the schematic flowchart of step S12 in an embodiment of the present invention Figure 1 ;

[0065] Figure 4 It is the schematic flowchart of step S13 in an embodiment of the present invention Figure 2 ;

[0066] Figure 5 It is the schematic flowchart of step S131 in an embodiment of the present invention Figure 3 ;

[0067] Figure 6 It is a schematic diagram of program modules of a detection and positioning method for urban low-altitude micro UAVs in an embodiment of the present invention;

[0068] Figure 7 It is a schematic diagram of the structure of an electronic device in an embodiment of the present invention. Detailed implementation manners

[0069] 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.

[0070] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above accompanying drawings are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not necessarily need to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0071] The detection and positioning of low-altitude urban drones is a current hot topic. In fact, micro and small drones are rapidly transitioning from military to civilian use, quickly entering the market and being sought after by many individual users. At the same time, since micro and small drones do not require identity registration, to a certain extent, they avoid the constraints of drone regulations, and their potential hazards to residents' safety, personal privacy, etc. cannot be ignored. However, there is currently little work on the detection and positioning of low-altitude micro and small drones.

[0072] Due to the extremely small size, extremely weak radio signals, and the continuous conversion of the activity range around urban buildings and indoors of micro and small drones, it poses great challenges to existing drone detection and positioning technologies. It is very difficult to identify micro drones through optoelectronic detection, and the field of view of optoelectronic detection is very limited around buildings and in indoor environments, so it is not practical; radar detection is likely to misjudge micro and small drones as other things such as birds, and the clutter is too strong and the radiation is too large to be used in densely populated areas such as around urban buildings and indoors. Traditional radio detection technologies (such as TOA, TDOA, AOA, RSS, etc.) are also very difficult to detect and position micro and small drones due to limitations such as weak signal strength, strong multipath effect, unstable time difference accuracy, and instantaneous switching of signal fading models.

[0073] Therefore, existing single technologies are difficult to solve the problem of detecting and positioning micro and small drones in the area around urban buildings.

[0074] In view of this, the present invention proposes a method for detecting and positioning low-altitude micro and small urban drones. Based on the multi-dimensional data analysis and judgment of micro and small drones, by comprehensively scheduling a variety of drone monitoring means and dynamically allocating tracking and monitoring resources, it breaks through the traditional means of performing traditional mathematical analysis on single-dimensional data, and performs intelligent interpretation of multi-dimensional data based on a deep neural network to solve the continuous monitoring and scheduling scheme and data analysis problems of weak-signal drones in a high-frequency dynamic scenario; realizing the detection and positioning of low-altitude micro and small urban drones;

[0075] Please refer to Figure 1 , the technical solution of the method for detecting and positioning low-altitude micro and small urban drones provided by the present invention is as follows: First, deploy a monitoring device network indoors and outdoors at low altitude in the city; use a test drone to traverse each monitoring device network; the system implementing this solution includes the following modules:

[0076] A module for collecting multi-dimensional measured data of micro and small drones; used to collect multi-dimensional data of the test drone by using the monitoring devices in the monitoring device network to form a sample data set; where the monitoring devices include acoustic sensors, optical sensors, and radio sensors;

[0077] Miniature UAV actual measurement multi-dimensional data acquisition module; used to establish a behavior fingerprint feature library for miniature UAVs based on simulated data completion; the behavior fingerprint feature library of miniature UAVs includes a sample data set;

[0078] Detection and warning module for miniature UAVs; used to monitor and give warnings to UAVs, and report the possible discovery of UAVs; among them, the monitoring device that reports the discovery of UAVs is the monitoring device for warning;

[0079] Miniature UAV monitoring collaborative scheduling module based on fingerprint-related matching; used to match the monitoring signals collected by the aforementioned modules with the behavior fingerprint feature library of miniature UAVs; initially determine whether a UAV exists, and if a UAV exists, control the monitoring devices around the monitoring device that gives warnings to conduct collaborative monitoring; among them, the monitoring device indicated by the arrow is the monitoring device for collaborative monitoring;

[0080] Miniature UAV recognition and positioning module based on deep neural network situation interpretation; used to conduct interpretation based on the neural network and output the recognition result and positioning result of the miniature UAV.

[0081] The technical solution of the present invention will be described in detail below with specific embodiments. These several specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments.

[0082] Please refer to Figures 2 - 6 , according to an embodiment of the present invention, a detection and positioning method for urban low-altitude miniature UAVs is provided. A flowchart of a detection and positioning method for urban low-altitude miniature UAVs is as Figure 2 shown, and the method includes:

[0083] S11: Control the monitoring device to scan and obtain the first signal to form a data sample set; the first signal characterizes the signal of the UAV detected by the monitoring device during scanning; among them, the monitoring device includes a radio device, an optical device, and an acoustic device; in a specific implementation manner, the radio device is a radio sensor; the optical device is an optical sensor; the acoustic device is an acoustic sensor; of course, the present application is not limited thereto, and any implementation form of the corresponding monitoring device that can achieve the purpose of the present invention is within the protection scope of the present invention; in an embodiment, the first signal includes the corresponding radio signal, optical signal, and acoustic signal.

[0084] S12: Process the data sample set using the simulated data completion method to establish a behavior fingerprint feature library; the behavior fingerprint feature library includes the data sample set; in a specific embodiment, the simulated data completion method includes the generative adversarial network method, the sample feature clustering analysis method, and the interpolation method. The specific content of establishing the behavior fingerprint feature library by the simulated data completion method is prior art and will not be elaborated here.

[0085] S13: Control each monitoring device to scan at a first scanning frequency point and obtain a first scanning signal and a warning signal until each monitoring device has exhausted all the first scanning frequency points; the warning signal indicates that when the intensity of the first scanning signal is greater than the set intensity threshold, a signal sent by the corresponding monitoring device is received; the first scanning frequency point represents any data in the behavior fingerprint feature library; the warning signal includes first monitoring position information; the first monitoring position information represents the position information of the monitoring device that sends the warning signal; the warning signal includes the corresponding first scanning signal;

[0086] In an embodiment, the first scanning signal includes a first acoustic scanning signal and a first radio scanning signal.

[0087] Among them, the warning signal also includes the corresponding first scanning signal, and here the corresponding first scanning signal specifically includes: a first warning monitoring signal; the first warning monitoring signal includes: when the monitoring device that sends the warning signal monitors the drone to be warned, the radio signal and the acoustic signal set by the monitoring device.

[0088] S14: Initially determine whether a drone exists based on the first scanning signal and the behavior fingerprint database; here, the first scanning signal refers to the corresponding first scanning signal included in the warning signal.

[0089] If it is determined that a drone exists, determine a number of cooperative monitoring devices according to the first monitoring position information, and control the number of cooperative monitoring devices to monitor a first area and obtain corresponding first monitoring signals; the first area represents the area where the monitoring device that sends the warning signal is located;

[0090] Among them, the cooperative monitoring devices in the cooperative monitoring devices monitor the first area with the corresponding first warning monitoring signal; among them, the radio device monitors the first area with the optical signal in the corresponding first warning signal; the acoustic device monitors the first area with the acoustic signal in the corresponding first warning signal; in an embodiment, the first monitoring signal includes: a first optical monitoring signal, a first acoustic monitoring signal, and a first radio monitoring signal. And

[0091] S15: Process the first monitoring signal using a deep neural network to obtain a first monitoring result with the deep neural network; wherein, the first monitoring signal is the monitoring signal received by the collaborative monitoring devices at the same moment. In one embodiment, the first monitoring result includes an identification result and a location result; the identification result represents the judgment result of whether a drone exists; the location result represents the positioning information of the drone.

[0092] Among them, the processes of steps S13 - S15 are continuously cyclic, and the data sample set and the behavior fingerprint feature library in steps S11 - S12 are also updated in real time; and the first scanning frequency point in step S13 changes with the update of the behavior fingerprint feature library in step S12.

[0093] Among them, in one implementation, the acoustic signal is the sound signal collected under different indoor and outdoor distance conditions (1 meter - 100 meters) and different background noise conditions.

[0094] In one implementation, the radio signal includes: with an outdoor position division of 3 * 3 meters and an indoor position division of 1 * 1 meter, when the drone is at each position, the radio signal collected by the radio device; among them, for positions such as building corners, street intersections, and tree trunks, the position division density is increased to 1 * 1 meter.

[0095] The collected optical signal refers to: optical photography influence maps at various angles under different weather, lighting, indoor and outdoor, and distance conditions (1 meter - 100 meters).

[0096] In one implementation, in step S11, when controlling the monitoring device to scan, its scanning object includes: micro - small drones on the market that are released to the corresponding test sites in the form of experiments.

[0097] Among them, the micro - small drone is a weak - signal drone, and it is difficult for a single monitoring device to accurately judge its position information through monitoring and positioning.

[0098] In one implementation, in step S11, before controlling the monitoring device to scan, it is necessary to set parameters for the monitoring device.

[0099] In one implementation, set the parameters of the radio device as the first scanning parameters; the first scanning parameters refer to any combination of data such as the frequency and signal strength that can be set by default for the radio device; the parameter setting of the acoustic device is similar to that of the radio device and will not be elaborated here.

[0100] Since the monitoring equipment in the above steps refers to radio equipment, optical equipment and acoustic equipment, the first signal, the first scanning signal, the early warning signal and the first monitoring signal include multi-dimensional data monitored by various monitoring equipment.

[0101] The present invention provides a detection and positioning method for low-altitude micro-UAVs in cities. Due to the analysis of multi-dimensional data signals based on deep neural networks, the detection and positioning of weak-signal UAVs in high-frequency dynamic scenarios is more accurate and reliable than the existing technology that uses single monitoring data. At the same time, according to the first monitoring position in the early warning signal, the collaborative monitoring equipment is determined and dispatched to monitor the first area, thereby realizing the scheduling of collaborative monitoring equipment in continuous monitoring.

[0102] It can be seen that the technical solution provided by the present invention obtains the corresponding first monitoring signal by scheduling the collaborative monitoring equipment according to the first monitoring position; and obtains a more reliable first monitoring result by performing data analysis based on a deep neural network on the multi-dimensional first monitoring signal, thereby solving the problem of how to monitor and locate weak-signal UAVs (in high-frequency dynamic scenarios) and the scheduling of collaborative monitoring equipment; and improving the accuracy of monitoring and positioning of low-altitude micro-UAVs.

[0103] Among them, steps S12-S15 specifically include:

[0104] In one embodiment, step S12, using the simulated data completion method to process the data sample set to establish a behavior fingerprint feature library specifically includes: the flowchart of step S12 is as follows: Figure 3 As shown, steps S121-S123:

[0105] S121: using the generative adversarial network method to expand the data sample set to generate a number of generated values; in one embodiment, the total amount of data after the data sample set is expanded by using the generative adversarial network method to generate a number of generated values ​​is 4 times the number of data samples in the data sample set;

[0106] S122: Analyze the data sample set using the sample cluster analysis method, and remove abnormal first signals;

[0107] S123: Fill the data sample set using the interpolation method to form a filling value; the behavior fingerprint feature library includes fingerprint feature values; the behavior fingerprint feature values ​​include the generated value, the first signal remaining after removing the abnormal first signal, and the filling value. In one embodiment, when the data sample set is filled using the interpolation method, the filling is performed according to the first standard; the first standard refers to: when filling the data between the test sites, the geographic location accuracy is to 20cm, and the frequency accuracy is to 200HZ.

[0108] In one embodiment, step S13, controlling each of the monitoring devices to scan at a first scanning frequency point and obtaining a first scanning signal and a warning signal includes: The flowchart of step S13 is as follows Figure 4 shown, steps S131 - S132:

[0109] S131: Determine the first number of the first scanning frequency points of each of the monitoring devices; wherein, the frequency point values of each of the first scanning frequency points of any one of the monitoring devices are different; in one implementation, when the monitoring device is a radio device, the first number is 3; when the monitoring device is an acoustic device, the first number is another number;

[0110] In one embodiment, wherein, step S131, determining the first number of the first scanning frequency points of each of the monitoring devices specifically includes: The flowchart of step S131 is as follows Figure 5 shown, steps S1311 - S1313:

[0111] S1311: Determine the first number of the first scanning frequency points of the first monitoring device; the first monitoring device represents any one of the monitoring devices;

[0112] S1312: Determine the first number of the first scanning frequency points of the second monitoring device; the second monitoring device represents the monitoring devices within a first distance around the first monitoring device; the frequency point values of the first scanning frequency points of the first monitoring device and the second monitoring device are different; in one embodiment, the first distance is 100m.

[0113] S1313: Determine the first number of the first scanning frequency points of the third monitoring device; the third monitoring device represents the monitoring devices within the first distance around the second monitoring device; wherein, the frequency point values of the first scanning frequency points of the third device and the second device are different;

[0114] And so on, until the first number of the first scanning frequency points of each of the monitoring devices is determined.

[0115] In one embodiment, the first scanning frequency points include fixed-period scanning frequency points and non-fixed-period scanning frequency points; the fixed-period scanning frequency points represent the fingerprint feature values with the most concentrated probabilities; the non-fixed-period scanning frequency points represent the other frequency points outside the fixed-period scanning frequency points. When the monitoring device is a radio device, in one implementation, the fixed-period scanning frequency points are the frequency bands of the first frequency band quantity with the highest radio working frequency in the behavior fingerprint feature library, and are the frequency points of the first frequency point quantity with the highest radio working frequency in the corresponding frequency bands; in one implementation, the first frequency band quantity is 6, and the first frequency point quantity is 20; when the monitoring device is an acoustic device, the selection of the corresponding fixed period is similar to that of the radio device, and the specific frequency bands and frequency points are selected according to needs, and the present invention does not limit this.

[0116] S132: Control each of the monitoring devices to sequentially scan with the first quantity of the first scanning frequency points to obtain corresponding first scanning signals until each of the monitoring devices has exhausted all the first scanning frequency points; and obtain corresponding first scanning signals and warning signals. In one implementation, the scanning period is the first scanning period. In a specific embodiment, the first period is 800 milliseconds; it can also be other first periods, and any implementation form of the first period is within the protection scope of the present invention, and the present invention is not limited thereto.

[0117] In one embodiment, step S132, controlling each of the monitoring devices to sequentially scan with the first quantity of the first scanning frequency points to obtain corresponding first scanning signals until each of the monitoring devices has exhausted all the first scanning frequency points, specifically includes: S1321 - S1322:

[0118] S1321: Control the monitoring device to scan with the first quantity of the fixed-period scanning frequency points to obtain the first scanning signals until each of the monitoring devices has scanned all the fixed-period scanning frequency points;

[0119] S1322: Control the monitoring device to scan with the first quantity of the non-fixed-period scanning frequency points to obtain the first scanning signals; until each of the monitoring devices has scanned all the non-fixed-period scanning frequency points.

[0120] In one embodiment, in step S14, initially determining whether a drone exists according to the first scanning signals and the behavior fingerprint database specifically includes: S141 - S142:

[0121] S141: Determine a first correlation degree and a second correlation degree based on the first scanning signal and the behavior fingerprint database; the first correlation degree characterizes the correlation degree between the first scanning signal and the corresponding first signal, and the second correlation degree characterizes the correlation degree between the first scanning signal and the corresponding generated value and filled value;

[0122] S142: When the first correlation degree is greater than a first preset correlation degree, or the second correlation degree is greater than the second preset correlation degree, preliminarily determine that the drone exists. In one implementation, the first preset correlation degree is 70%; the second preset correlation degree is 80%;

[0123] In one embodiment, in step S14, determine a plurality of collaborative monitoring devices according to the first monitoring position information, and control the plurality of collaborative monitoring devices to monitor a first area to obtain corresponding first monitoring signals. Specifically, it further includes:

[0124] S143: Control the acoustic device and the radio device to scan with the corresponding first scanning signal and obtain the first acoustic monitoring signal and the first radio monitoring signal; and

[0125] Control the optical device to monitor the first area and obtain the first optical monitoring signal. Among them, the optical device in the collaborative monitoring device is an optoelectronic monitoring device in an adjacent visible area;

[0126] Among them, in step S143, controlling the optical device to monitor the first area and obtain the first optical monitoring signal specifically includes:

[0127] S1431: Control the radio device to monitor the first area with the radio signal in the first early warning monitoring signal and obtain the first radio monitoring signal; and control the acoustic device to monitor the first area with the radio signal in the first early warning monitoring signal and obtain the first radio monitoring signal; and

[0128] S1432: Control the optical device to monitor the first area and obtain the first optical monitoring signal.

[0129] In one embodiment, in step S15, obtain a first monitoring result according to the first monitoring data and a deep neural network. Specifically, it includes: S151 - S152:

[0130] S151: Obtain a first situation map according to the first monitoring data;

[0131] In one embodiment, in step S151, obtaining a first situation map according to the first monitoring data specifically includes:

[0132] Obtain a first situation map based on the first acoustic monitoring data and the first optical monitoring data in the first monitoring data.

[0133] S152: Obtain the first monitoring result based on the first situation map and the deep neural network.

[0134] In one embodiment, before obtaining the first situation map based on the first monitoring data, it further includes;

[0135] Perform a feasibility interpretation on the first optical monitoring signal to obtain a first interpretation feasibility;

[0136] Compare the first interpretation feasibility with a preset feasibility; if the first interpretation feasibility is not lower than the preset feasibility, then eliminate the incorrect first acoustic monitoring signal and / or the first radio monitoring signal. In a specific embodiment, the preset interpretation feasibility is 85%;

[0137] Secondly, according to an embodiment of the present invention, there is also provided a device 20 for a detection and positioning method of an urban low-altitude micro unmanned aerial vehicle. A schematic diagram of a program module of a detection and positioning method of an urban low-altitude micro unmanned aerial vehicle is as Figure 6 shown, and the device includes:

[0138] A sample data set acquisition module 21; used to control the monitoring device to perform scanning and obtain the first signal to form the sample data set;

[0139] A behavior fingerprint library establishment module 22; used to establish a behavior fingerprint library according to the sample data set and the simulation data completion method;

[0140] A signal detection and warning module 23; used to control each of the control monitoring devices to perform scanning at a first scanning frequency point to obtain corresponding first scanning signals and warning signals;

[0141] A collaborative monitoring device scheduling and monitoring data acquisition module 24; used to determine whether an unmanned aerial vehicle exists according to the first scanning signal and the behavior fingerprint database, and determine a collaborative monitoring device according to the warning signal, and control the collaborative monitoring device to monitor a first area and obtain first monitoring data;

[0142] An identification and positioning module 25; used to obtain a first monitoring result according to a deep neural network and the first monitoring data.

[0143] In addition, according to an embodiment of the present invention, there is also provided an electronic device, including a processor and a memory, and the memory is used to store code;

[0144] The processor is configured to execute the code in the memory to implement the detection and positioning method of the urban low-altitude micro UAV described in the foregoing embodiments of the present invention.

[0145] According to an embodiment of the present invention, there is also provided a storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the detection and positioning method of the urban low-altitude micro UAV described in any one of the foregoing embodiments of the present invention.

[0146] There is provided an electronic device 30, as shown in Figure 7 which includes a processor 31 and a memory 32. The memory is configured to store code; the processor 31 and the memory 32 communicate through a bus 33.

[0147] The processor is configured to execute the code in the memory to implement the detection method of the TDOA site described above.

[0148] The embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements a detection and positioning method of an urban low-altitude micro UAV described above.

[0149] Those of ordinary skill in the art can understand that all or part of the steps of implementing the foregoing method embodiments can be completed by hardware related to program instructions. The foregoing program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps including the foregoing method embodiments; and the foregoing storage medium includes: various media such as ROM, RAM, magnetic disk, or optical disc that can store program code.

[0150] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A detection and positioning method for a low-altitude micro UAV in the city, characterized in that, Including: Controlling the monitoring device to perform scanning and obtaining a first signal to form a data sample set; The first signal characterizes the signal of the unmanned aerial vehicle detected when the monitoring device performs scanning; Processing the data sample set by using an analog data completion method to establish a behavior fingerprint feature library; the behavior fingerprint feature library includes the data sample set; Controlling each monitoring device to scan at a first scanning frequency point and obtaining a first scanning signal and a warning signal until each monitoring device has exhausted all the first scanning frequency points; the warning signal characterizes that when the intensity of the first scanning signal is greater than a set intensity threshold, the signal sent by the corresponding monitoring device is received; the first scanning frequency point characterizes any data in the behavior fingerprint feature library; the warning signal includes first monitoring position information; the first monitoring position information characterizes the position information of the monitoring device that sends the warning signal; the warning signal includes the corresponding first scanning signal; Preliminarily determining whether there is an unmanned aerial vehicle according to the first scanning signal and the behavior fingerprint database; if it is determined that there is an unmanned aerial vehicle, determining a plurality of cooperative monitoring devices according to the first monitoring position information, and controlling the plurality of cooperative monitoring devices to monitor a first area to obtain corresponding first monitoring signals; the first area characterizes the area where the monitoring device that sends the warning signal is located; And Processing the first monitoring signal by using a deep neural network to obtain a first monitoring result; Wherein, the monitoring device includes a radio device, an optical device and an acoustic device.

2. The method for detecting and positioning an urban low-altitude micro unmanned aerial vehicle according to claim 1, characterized in that The first signal includes corresponding radio signals, optical signals and acoustic signals.

3. The method for detecting and positioning an urban low-altitude micro unmanned aerial vehicle according to claim 2, characterized in that The analog data completion method includes a generative adversarial network method, a sample feature clustering analysis method and an interpolation method.

4. The detection and positioning method of the urban low-altitude micro UAV according to claim 3, characterized in that, Processing the data sample set by using an analog data completion method to establish a behavior fingerprint feature library specifically includes: Expanding the data sample set by using the generative adversarial network method to generate a plurality of generated values; Analyzing the data sample set by using the sample clustering analysis method and removing abnormal first signals; Filling the data sample set by using the interpolation method to form filled values; the behavior fingerprint feature library includes fingerprint feature values; the behavior fingerprint feature values include the generated values, the remaining first signals after removing abnormal first signals and the filled values.

5. The detection and positioning method of the urban low-altitude micro UAV according to claim 4, characterized in that, The first scanning signal includes a first acoustic scanning signal and a first radio scanning signal.

6. The detection and positioning method of the urban low-altitude micro UAV according to claim 5, characterized in that, Controlling each monitoring device to scan at a first scanning frequency point and obtaining a first scanning signal and a warning signal includes: Determining a first number of the first scanning frequency points of each monitoring device; wherein, the frequency point values of each first scanning frequency point of any monitoring device are different; Control each of the monitoring devices to sequentially scan with the first number of the first scanning frequency points to obtain corresponding first scanning signals until each of the monitoring devices has exhausted all the first scanning frequency points.

7. The detection and positioning method of the urban low-altitude micro UAV according to claim 6, characterized in that, The scanning period when controlling each of the monitoring devices to sequentially scan with the first number of the first scanning frequency points is the first scanning period.

8. The detection and positioning method of the urban low-altitude micro UAV according to claim 7, characterized in that Determining the first number of the first scanning frequency points for each of the monitoring devices specifically includes: Determining the first number of the first scanning frequency points for the first monitoring device; the first monitoring device represents any one of the monitoring devices; Determining the first number of the first scanning frequency points for the second monitoring device; the second monitoring device represents the monitoring devices within a first distance around the first monitoring device; the frequency point values of the first scanning frequency points of the first monitoring device and the first scanning frequency points of the second monitoring device are different from each other; Determining the first number of the first scanning frequency points for the third monitoring device; the third monitoring device represents the monitoring devices within the first distance around the second monitoring device; wherein, the frequency point values of the first scanning frequency points of the third monitoring device and the first scanning frequency points of the second monitoring device are different from each other; And so on until the first number of the first scanning frequency points for each of the monitoring devices is determined.

9. The detection and positioning method of the urban low-altitude micro UAV according to claim 8, characterized in that, The first distance is 100m.

10. The detection and positioning method for an urban low-altitude micro UAV according to claim 9, characterized in that The first scanning frequency points include fixed-period scanning frequency points and non-fixed-period scanning frequency points; The fixed-period scanning frequency points represent the fingerprint feature values with the most concentrated probability; the non-fixed-period scanning frequency points represent the other frequency points outside the fixed-period scanning frequency points.

11. The detection and positioning method of the urban low-altitude micro UAV according to claim 10, characterized in that, Controlling each of the monitoring devices to sequentially scan with the first number of the first scanning frequency points to obtain corresponding first scanning signals until each of the monitoring devices has exhausted all the first scanning frequency points specifically includes: Controlling the monitoring devices to scan with the first number of the fixed-period scanning frequency points to obtain the first scanning signals until each of the monitoring devices has scanned all the fixed-period scanning frequency points; Controlling the monitoring devices to scan with the first number of the non-fixed-period scanning frequency points to obtain the first scanning signals; until each of the monitoring devices has scanned all the non-fixed-period scanning frequency points.

12. The detection and positioning method of the urban low-altitude micro UAV according to claim 11, characterized in that, Preliminarily determining whether a UAV exists according to the first scanning signals and the behavior fingerprint database specifically includes: Determining a first correlation degree and a second correlation degree according to the first scanning signals and the behavior fingerprint database; the first correlation degree represents the correlation degree between the first scanning signals and the corresponding first signals, and the second correlation degree represents the correlation degree between the first scanning signals and the corresponding generated values and filled values; When the first correlation degree is greater than a first preset correlation degree, or the second correlation degree is greater than a second preset correlation degree, it is preliminarily determined that a UAV exists.

13. The detection and positioning method of the urban low-altitude micro UAV according to claim 12, wherein the first monitoring signals include: a first optical monitoring signal, a first acoustic monitoring signal, and a first radio monitoring signal.

14. The detection and positioning method of the urban low-altitude micro UAV according to claim 13, characterized in that, Determine a number of collaborative monitoring devices according to the first monitoring position information, and control the number of collaborative monitoring devices to monitor a first area to obtain corresponding first monitoring signals, specifically including: Controlling the acoustic device and the radio device to scan with the corresponding first scan signal and obtain the first acoustic monitoring signal and the first radio monitoring signal; and Controlling the optical device to monitor the first area and obtain the first optical monitoring signal.

15. The detection and positioning method of the urban low-altitude micro UAV according to claim 14, wherein the first monitoring result includes an identification result and a position result; the identification result represents a judgment result of whether the UAV exists; the position result represents the positioning information of the UAV.

16. The detection and positioning method of the urban low-altitude micro UAV according to claim 15, characterized in that, Obtain the first monitoring result according to the first monitoring data and the deep neural network, specifically including: Obtain a first situation map according to the first monitoring data; Obtain the first monitoring result according to the first situation map and the deep neural network.

17. The detection and positioning method of the urban low-altitude micro UAV according to claim 16, characterized in that, Obtain a first situation map according to the first monitoring data; specifically including: Obtain a first situation map according to the first acoustic monitoring data and the first optical monitoring data in the first monitoring data.

18. The detection and positioning method of the urban low-altitude micro UAV according to claim 17, characterized in that, Before obtaining the first situation map according to the first monitoring data, it also includes; Perform a feasibility interpretation on the first optical monitoring signal to obtain a first interpretation feasibility; Compare the first interpretation feasibility with a preset feasibility; If the first interpretation feasibility is not lower than the preset feasibility, eliminate the incorrect first acoustic monitoring signal and / or the first radio monitoring signal.

19. An apparatus for a detection and positioning method of a low-altitude micro UAV in a city, characterized in that, The device is used to implement the detection and positioning method of the urban low-altitude micro UAV as described in claim 1, and the device includes: A sample data set acquisition module; used to control the monitoring device to scan and obtain a first signal to form the sample data set; A behavior fingerprint library establishment module; used to establish a behavior fingerprint library according to the sample data set and the simulated data complement method; A signal detection and early warning module; used to control each of the control monitoring devices to scan at a first scan frequency point to obtain a corresponding first scan signal and an early warning signal; A collaborative monitoring device scheduling and monitoring data acquisition module; used to determine whether a UAV exists according to the first scan signal and the behavior fingerprint database, and determine collaborative monitoring devices according to the early warning signal, and control the collaborative monitoring devices to monitor a first area and obtain first monitoring data; An identification and positioning module; used to obtain a first monitoring result according to the deep neural network and the first monitoring data.

20. An electronic device, characterized in that, It includes a processor and a memory, and the memory is used to store code; The processor is used to execute the code in the memory to implement the detection and positioning method of the urban low-altitude micro UAV as described in any one of claims 1 to 18.

21. A storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the detection and positioning method of the urban low-altitude micro UAV according to any one of claims 1 to 18.

Images