A system and method for an acoustic-optic directional denial strategy for airport security

Abstract

The application discloses an acoustic-optic directional denial strategy system and method for airport safety, relates to the field of bird repelling technology, and is characterized in that the acoustic-optic cooperative directional denial system is constituted, different types of sensors are installed in the system, the scale of birds is classified, different acoustic-optic frequency ranges are set for different bird scales, and dangerous grades are marked, and a four-dimensional repelling strategy library is constructed; different weights are given to different state data, bird stress resistance indexes are obtained by fusing different bird state data according to the weights; the bird stress resistance indexes are judged by using a repelling threshold value, and different repelling strategies are selected; in the frequency range of the selected repelling strategy, an unpredictable time sequence modulation mechanism is set, and the acoustic-optic emission time sequence and the acoustic wave frequency are dynamically adjusted; the bird repelling failure rate is counted, the bird escape time at each repelling time is measured, the resistance inhibition coefficient is calculated, the resistance inhibition coefficient is judged, and the four-dimensional repelling strategy library is updated.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of bird repelling technology, in particular to a sound-light directional denial strategy system and method for airport safety. BACKGROUND

[0002] Traditional airport bird repelling methods include physical deterrents (such as scarecrows, protective nets, etc.), chemical agents, ultrasonic bird repelling, etc., but these methods have low efficiency, strong bird adaptability, environmental pollution, etc. For example, ultrasonic bird repelling has limited range, strong sound bird repelling lacks monitoring and evaluation means, and laser bird repelling has poor daytime effect, etc. With the continuous progress of technology, sound-light directional denial systems for airport safety have gradually improved. For example, some bird repelling systems integrate directional sound wave bird repelling devices and mobile detection systems to effectively drive and concentrate the prevention and control of bird flocks. At the same time, laser technology has also been applied to airport bird repelling, which deters and repels birds by emitting laser beams of specific wavelengths.

[0003] However, when using sound waves and lasers for bird repelling, due to the different sizes of birds, their resistance to sound and light is also different, so targeted bird repelling strategies are needed for targeted repelling. However, birds move quickly, resulting in rapid changes in real time, so it is crucial to quickly adjust the repelling strategy according to changes. Birds have natural adaptability to the environment, and the sound and light in the static repelling strategy will greatly reduce the repelling effect after being used for a period of time due to the adaptation of birds. SUMMARY

[0004] The purpose of the present application is to provide a sound-light directional denial strategy system and method for airport safety to solve the problems in the prior art.

[0005] To achieve the above-mentioned purpose, the present application provides the following technical scheme:

[0006] A sound-light directional denial strategy method for airport safety, the method comprising the following steps:

[0007] S100, setting sound wave emitters and laser emitters in the airport respectively to form a sound-light cooperative directional denial system, and installing different types of sensors in the system to collect bird state data when the system repels birds;

[0008] Further, the specific steps of collecting bird state data when the system repels birds are:

[0009] S101, set up a sound and light emission tower in the airport, install a sound wave emitter and a laser emitter in the sound and light emission tower, additionally install a monitoring radar, the monitoring radar detects birds in the airport, when a bird is detected, a response signal and bird location information are sent to the sound wave emitter and the laser emitter, after the sound wave emitter and the laser emitter receive the response signal, sound waves and laser are emitted according to the bird location information respectively;

[0010] The sound wave emitter, the laser emitter and the monitoring radar are set up at the same time, the monitoring radar can quickly detect birds and send location information, so that the sound wave emitter and the laser emitter can accurately aim at the birds, avoiding the problems of blind emission and low efficiency of traditional driving devices, ensuring that the driving effect acts on the target birds and reducing the interference on the non-target area.

[0011] An infrared thermal imager and a microphone array are installed in the sound and light emission tower, the infrared thermal imager is used to detect bird temperature data, and the microphone array is used to detect bird calling frequency data; all sensors and radars are time and space calibrated, and different types of data collected at the same time and space are output;

[0012] S102, extract the effective radius of the sound wave emitter and the laser emitter in each sound and light emission tower, set the distance between the sound and light emission towers, the formula is: D=3 1 / 2 ×r, D represents the distance between the sound and light emission towers, r represents the effective radius of the sound wave emitter and the laser emitter, the effective radius is the minimum value of the two emission radii of the sound wave emitter and the laser emitter; according to the distance, set up the sound and light emission towers in the airport, use the honeycomb grid coverage model to regard the effective coverage range of each sound and light emission tower as a hexagon, and use all the sound and light emission towers to build an airspace wall in the airport.

[0013] Regarding the effective range of each emission tower as a hexagon can maximize the use of the coverage range of the equipment, eliminate the driving blind area, ensure that the airspace of the airport is protected in all directions, and effectively prevent birds from entering the key area of the airport from the uncovered area.

[0014] S200, collect bird scale in historical bird driving records of the airport, classify the bird scale, extract sound and light frequency in the driving records, set different sound and light frequency ranges for different bird scales, mark the danger level, and build a four-dimensional driving strategy library;

[0015] Further, the specific steps of building the four-dimensional driving strategy library are as follows:

[0016] S201, collect the bird size in the historical bird driving record of the airport, classify the bird size, collect the size of all birds in the driving record, calculate the average value of the size of all birds, divide the birds larger than the average value into large birds, and divide the birds smaller than the average value into small birds, extract the minimum value of the number of birds during migration as the migration group threshold value, and determine that it is a migration group when the number of birds appearing at the same time in the airport is greater than the migration group threshold value; finally, the bird size in the airport is divided into large birds, small birds and migration groups;

[0017] According to the average value of the size, the birds are divided into large and small, and the migration group is determined in combination with the migration group threshold value, so that the classification of the bird size is more in line with the actual situation. Different sizes of birds have great differences in response to driving means, for example, large birds may need stronger driving intensity. This classification method provides a basis for formulating differentiated driving strategies, avoids the "one-size-fits-all" driving method, and improves the pertinence of driving.

[0018] S202, extract the sound wave frequency and laser frequency emitted by the sound wave emitter and the laser emitter in the driving record for different bird sizes, extract the maximum and minimum values to form a frequency interval [p min , p max ], set a danger level for different bird sizes, specifically: small birds are class 1, large birds are class 2, and migration groups are class 3; construct a four-dimensional driving strategy {bird size, frequency interval, danger level}, and integrate the driving strategies of the three bird sizes to obtain a four-dimensional driving strategy library.

[0019] S300, standardize the bird state data during bird driving collected by the sensor in real time, give different weights to different state data, and fuse different bird state data according to the weights to obtain a bird stress resistance index;

[0020] Further, the specific steps of fusing different bird state data according to the weights to obtain the bird stress resistance index are:

[0021] S301, use the monitoring radar to track the motion data of the birds during driving, obtain the escape acceleration a of the birds, use the infrared thermal imager to detect the body temperature change value AT of the birds during escape, use the microphone array to detect the call frequency offset Af of the birds, and set the weights of the three kinds of bird state data as w1, w2 and w3 respectively; the weights are set by the airport staff according to experience;

[0022] Standardize the three kinds of detected bird state data, fuse the standardized bird state data according to the weights, and the formula is:

[0023] ;

[0024] In the formula, R represents the bird stress resistance index, and a0 represents the initial escape acceleration of the bird before being driven away.

[0025] S400, according to the different safety levels in the four-dimensional driving strategy library, the driving threshold is calculated, the bird stress resistance index is judged by using the driving threshold, and different driving strategies are selected;

[0026] Further, the specific steps of selecting different driving strategies are:

[0027] S401, the resistance index of each driving of the three bird sizes at the initial stage of operation of the sound-light cooperative directional denial system is calculated, the average value and the standard deviation of the resistance index are calculated, the upper control limit value UCL is obtained by using the average value plus the standard deviation, and the resistance index of the driving is calculated during the operation of the sound-light cooperative directional denial system thereafter, when the resistance index is less than the upper control limit value, the average value and the standard deviation of the initial operation of the resistance index are optimized by using the resistance index, and the formula is:

[0028] ;

[0029] In the formula, μ(t) represents the average value of the resistance index at the tth driving, R t represents the resistance index at the tth driving, μ(t-1) represents the average value of the resistance index at the (t-1)th driving, represents the variance of the resistance index at the tth driving, represents the standard deviation at the tth driving, represents the variance of the resistance index at the (t-1)th driving, and λ represents a smoothing factor; the optimized standard deviation is obtained by taking the arithmetic square root of the optimized variance of the resistance index;

[0030] S402, for the driving strategy of each bird size, the driving threshold is calculated by using the optimized average value plus the standard deviation, the driving threshold of the small bird is In1, the driving threshold of the large bird is In2, and the driving threshold of the migratory bird is In3, when the bird is driven in real time, the real-time resistance index is judged by using the three driving thresholds, and the real-time driving strategy is dynamically changed, specifically: when the real-time resistance index is less than In1, the class1 strategy of the danger level is selected, when the real-time resistance index is less than In2 but greater than In1, the class2 strategy of the danger level is selected, and when the real-time resistance index is less than In3 but greater than In2, the class3 strategy of the danger level is selected.

[0031] According to the optimized average value plus standard deviation, the threshold values (In1, In2, In3) of different bird sizes are calculated, and corresponding strategies are selected according to the relationship between the real-time resistance index and the threshold value. The bird driving strategy is flexibly adjusted according to the real-time stress resistance of the birds. For example, when the bird resistance index rises, a stronger driving strategy is automatically selected to avoid driving failure caused by lagging driving strategy, thereby improving the flexibility and effectiveness of driving.

[0032] S500, in the selected driving strategy frequency range, an unpredictable timing modulation mechanism is set to dynamically adjust the sound and light emission timing and the sound wave frequency;

[0033] Further, the specific steps of dynamically adjusting the sound and light emission timing and the sound wave frequency are:

[0034] S501, after selecting the driving strategy, an unpredictable timing modulation mechanism is set in the sound and light frequency interval of the driving strategy, specifically:

[0035] ;

[0036] In the formula, g n+1 represents the sound and light emission time interval of the n+1th time, g n represents the sound and light emission time interval of the nth time; The unpredictable sound and light emission time interval breaks the regularity of fixed timing. Birds have learning and adaptation ability, and fixed timing driving means can easily adapt to birds, reducing the driving effect. Unpredictable timing can effectively avoid this problem and maintain the deterrent effect of the driving means.

[0037] A sound and light frequency random walk mechanism is constructed, and the formula is:

[0038] ;

[0039] In the formula, f s (g) the sound and light frequency at the time interval g, rand(g) represents a pseudo-random number based on the time interval; According to the above unpredictable timing modulation mechanism and sound and light frequency random walk mechanism, sound waves and laser are emitted in the frequency interval of the selected driving strategy according to the random emission time interval and random sound and light frequency to drive birds. The frequency random walk mechanism is constructed to make the sound and light frequency randomly change in the effective interval. The randomly changing frequency can stimulate birds from multiple aspects, increase the discomfort of birds, and further improve the driving effect, avoiding the tolerance of birds to a single frequency.

[0040] S600, the bird driving failure rate is counted, the bird escape time at each driving is measured, the resistance inhibition coefficient is calculated, the resistance inhibition coefficient is judged, and the four-dimensional driving strategy library is updated.

[0041] Further, the specific steps of updating the four-dimensional driving strategy library are:

[0042] S601, the sound-light cooperative directional rejection system records each time of driving away, calculates the driving failure rate, and the formula is: pf=N fail / N total , wherein pf represents the driving failure rate, N fail represents the number of driving failure records, N total represents the total number of driving; the bird escape time tr is measured each time of driving, the resistance inhibition coefficient is calculated, and the formula is:

[0043] ;

[0044] In the formula, , wherein α represents the attenuation coefficient, tr0 represents the initial escape time, and pf0 represents the initial driving failure rate; the driving failure rate is calculated, the bird escape time is measured, the driving effect is presented in the form of specific data, the effectiveness of the current strategy is intuitively understood by the staff, and an objective basis is provided for judging whether the strategy needs to be updated.

[0045] S602, when <0.9, it is judged that the four-dimensional driving strategy library needs to be updated, and the deep learning algorithm is started to update the four-dimensional driving strategy library.

[0046] An acoustic-light directional rejection strategy system for airport safety, the acoustic-light directional rejection strategy system comprises a sound-light cooperative directional rejection module, a four-dimensional driving strategy library module, a resistance index calculation module, a strategy selection module, a random mechanism emission module and a strategy updating module;

[0047] The sound-light cooperative directional rejection module is used for setting sound wave emitters and laser emitters in the airport respectively, forming a sound-light cooperative directional rejection system, and installing different types of sensors in the system to collect state data of birds when the system drives away the birds;

[0048] The four-dimensional driving strategy library module is used for setting different sound-light frequency ranges for different bird scales, and marking the danger level to construct a four-dimensional driving strategy library;

[0049] The resistance index calculation module is used for giving different weights to different state data, and fusing different bird state data according to the weights to obtain a bird stress resistance index;

[0050] The strategy selection module is used for calculating a driving threshold according to different safety levels in the four-dimensional driving strategy library, judging the bird stress resistance index by using the driving threshold, and selecting different driving strategies;

[0051] The random mechanism emitting module is used for setting an unpredictable timing modulation mechanism in a selected drive strategy frequency range, dynamically adjusting the sound-light emitting timing and the sound wave frequency.

[0052] The strategy updating module is used for measuring the bird escape time at each time of driving, calculating the resistance inhibition coefficient, judging the resistance inhibition coefficient, and updating the four-dimensional drive strategy library.

[0053] The strategy selection module comprises an upper control limit value updating unit, a drive threshold calculation unit and a strategy selection unit.

[0054] The upper control limit value updating unit is used for optimizing the average value and the standard deviation of the resistance index at the initial operation by using the resistance index when the resistance index is less than the upper control limit value.

[0055] The drive threshold calculation unit is used for calculating the drive threshold by using the optimized average value plus the standard deviation for each bird size drive strategy.

[0056] The strategy selection unit is used for judging the real-time resistance index by using the three drive thresholds when driving the birds in real time, and dynamically changing the real-time drive strategy.

[0057] The random mechanism emitting module comprises an unpredictable timing modulation mechanism and a sound-light frequency random walk mechanism.

[0058] The unpredictable timing modulation mechanism is used for randomly calculating the sound-light emitting time interval.

[0059] The sound-light frequency random walk mechanism is used for randomly calculating the sound-light frequency.

[0060] Compared with the prior art, the beneficial effects of the present application are:

[0061] 1、The present application converts the "resistance degree of birds to driving" into a quantifiable value by standardizing the bird escape acceleration, body temperature change value and calling frequency offset three types of core data, combining with the airport experience to set the weight to calculate the bird stress resistance index R, providing an objective basis for strategy adjustment, and avoiding the subjective error of traditional manual observation and judgment.

[0062] 2、In the present application, the average value and the standard deviation of the resistance index are calculated at the initial operation of the system, and are dynamically optimized through a smoothing factor in the subsequent period, and the drive threshold is determined based on the optimized "average value + standard deviation". When the bird resistance index changes due to the adaptation to the driving means, the threshold can be adjusted synchronously, ensuring that the strategy always matches the current state of the bird, and avoiding the problem of "quick failure" of the traditional fixed threshold.

[0063] 3. Traditional deterrent devices, due to their "fixed timing and single frequency," easily allow birds to adapt (e.g., they become accustomed to sound waves at fixed intervals and no longer flee). This invention avoids bird adaptation from a dual dimension of "timing + frequency" by using unpredictable timing modulation and random walks of sound and light frequencies, thus maintaining long-term deterrent effectiveness. Attached Figure Description

[0064] Figure 1 This is a schematic diagram of a module of an acoustic-optical directional denial strategy system for airport security according to the present invention. Detailed Implementation

[0065] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0066] Example: Figure 1 As shown, the present invention provides a technical solution.

[0067] A method for acoustic and optical directional denial strategy for airport security, the method comprising the following steps:

[0068] S100. Sound wave emitters and laser emitters are set up at the airport to form an acoustic-optical coordinated directional denial system. Different types of sensors are installed in the system to collect data on the status of birds when the system drives them away.

[0069] The specific steps for the data acquisition system to collect data on the birds' status during bird removal are as follows:

[0070] S101. An audio-visual transmission tower is set up in the airport. The audio-visual transmission tower is equipped with a sound wave transmitter and a laser transmitter. An additional monitoring radar is installed. The monitoring radar detects birds in the airport. When a bird is detected, a response signal and the bird's location information are sent to the sound wave transmitter and the laser transmitter. After receiving the response signal, the sound wave transmitter and the laser transmitter emit sound waves and lasers respectively according to the bird's location information.

[0071] Simultaneously equipped with a sonic emitter, a laser emitter, and a monitoring radar, the monitoring radar can quickly detect birds and send their location information, allowing the sonic and laser emitters to accurately target the birds. This avoids the problems of blind firing and low efficiency of traditional deterrent equipment, ensuring that the deterrent effect is on the target birds and reducing interference with non-target areas.

[0072] In the sound and light emission tower, an infrared thermal imager and a microphone array are installed, the infrared thermal imager is used to detect bird body temperature data, and the microphone array is used to detect bird calling frequency data; all sensors and radars are calibrated in time and space, and different types of data collected at the same time and space are output;

[0073] S102, the effective radius of the sound wave emitter and the laser emitter in each sound and light emission tower is extracted, and the spacing of the sound and light emission tower is set, and the formula is: D=3 1 / 2 ×r, D represents the spacing of the sound and light emission tower, r represents the effective radius of the sound wave emitter and the laser emitter, and the effective radius is the minimum value of the two emission radii of the sound wave emitter and the laser emitter; according to the spacing, the sound and light emission tower is set in the airport, a hexagon is regarded as an effective coverage range of each sound and light emission tower by using a honeycomb grid coverage model, and an airspace wall is constructed in the airport by using all sound and light emission towers.

[0074] The effective range of each emission tower is regarded as a hexagon, which can maximize the use of the coverage range of the equipment, eliminate the drive-off blind area, ensure that the airport airspace is protected in all directions, and effectively prevent birds from entering the key area of the airport from the uncovered area.

[0075] S200, bird scales in historical bird drive-off records of the airport are collected, the bird scales are classified, sound and light frequencies in the drive-off records are extracted, different sound and light frequency ranges are set for different bird scales, and a dangerous level is marked to construct a four-dimensional drive-off strategy library;

[0076] The specific steps of constructing the four-dimensional drive-off strategy library are as follows:

[0077] S201, bird scales in historical bird drive-off records of the airport are collected, the bird scales are classified, the sizes of all birds in the drive-off records are collected, the average value of the sizes of all birds is calculated, birds larger than the average value are classified as large birds, and birds smaller than the average value are classified as small birds; according to the number of birds during migration, the minimum value of the number of birds during migration of all birds is extracted as a migration group threshold value, when the number of birds appearing in the airport at the same time is greater than the migration group threshold value, it is judged as a migration group; finally, the bird scales in the airport are divided into large birds, small birds and migration groups;

[0078] According to the average value of the size, the birds are divided into large and small, and the migration group threshold value is combined to determine the migration group, so that the classification of bird scales is more in line with the actual situation. Different scales of birds have great differences in response to drive-off means, for example, large birds may need stronger drive-off intensity. This classification method provides a basis for subsequent differentiated drive-off strategy, avoids the "one-size-fits-all" drive-off method, and improves the pertinence of drive-off.

[0079] S202, extract the sound wave frequency and laser frequency emitted by the sound wave emitter and the laser emitter in the repelling record for different bird sizes, extract the maximum and minimum to form a frequency interval [p min , p max ], set the danger level for different bird sizes, specifically: small birds are class 1, large birds are class 2, and migrating groups are class 3; construct a four-dimensional repelling strategy {bird size, frequency interval, danger level}, and integrate the repelling strategies of the three bird sizes to obtain a four-dimensional repelling strategy library.

[0080] S300, standardize the bird state data collected by the sensor in real time during bird repelling, assign different weights to different state data, and fuse different bird state data according to the weights to obtain a bird stress resistance index;

[0081] The specific steps of fusing different bird state data according to the weights to obtain the bird stress resistance index are:

[0082] S301, use the monitoring radar to track the motion data of the bird during repelling, obtain the escape acceleration a of the bird, use the infrared thermal imager to detect the body temperature change value AT of the bird during escape, use the microphone array to detect the calling frequency offset Af of the bird, and set the weights of the three bird state data as w1, w2, and w3, respectively; the weights are set by the airport staff according to experience;

[0083] Standardize the detected three bird state data, and fuse the standardized bird state data according to the weights, the formula is:

[0084]

[0085] In the formula, R represents the bird stress resistance index, and a0 represents the initial escape acceleration of the bird before repelling.

[0086] S400, calculate the repelling threshold according to the different safety levels in the four-dimensional repelling strategy library, use the repelling threshold to judge the bird stress resistance index, and select different repelling strategies;

[0087] The specific steps of selecting different repelling strategies are:

[0088] S401, calculate the resistance index of each repelling of the three bird sizes at the initial stage of operation of the sound-light cooperative directional rejection system, calculate the average value and standard deviation of the resistance index, use the average value plus the standard deviation to obtain the upper control limit value UCL, and calculate the resistance index during repelling when the sound-light cooperative directional rejection system is running thereafter, when the resistance index is less than the upper control limit value, use the resistance index to optimize the average value and standard deviation of the initial running resistance index, the formula is:

[0089] ​ ;

[0090] In the formula, μ(t) represents the average of the resistance index at the tth time of driving, R t represents the resistance index at the tth time of driving, μ(t-1) represents the average of the resistance index at the (t-1)th time of driving, represents the variance of the resistance index at the tth time of driving, represents the standard deviation at the tth time of driving, represents the variance of the resistance index at the (t-1)th time of driving, λ represents a smoothing factor; the optimized standard deviation is obtained by taking the arithmetic square root of the optimized variance of the resistance index;

[0091] S402, for each bird size driving strategy, the driving threshold is calculated by using the optimized average value plus the standard deviation, the driving threshold of small birds is In1, the driving threshold of large birds is In2, and the driving threshold of migratory birds is In3. When driving birds in real time, the real-time resistance index is judged by using the three driving thresholds, and the real-time driving strategy is dynamically changed. Specifically, when the real-time resistance index is less than In1, the class1 strategy is selected, when the real-time resistance index is less than In2 but greater than In1, the class2 strategy is selected, and when the real-time resistance index is less than In3 but greater than In2, the class3 strategy is selected.

[0092] According to the optimized average value plus the standard deviation, the driving thresholds (In1, In2, In3) of different bird sizes are calculated, and the corresponding strategy is selected according to the relationship between the real-time resistance index and the threshold. According to the real-time stress resistance of birds, the driving strategy is flexibly adjusted, for example, when the bird resistance index increases, a stronger driving strategy is automatically selected, which avoids the lag of the driving strategy leading to driving failure, and improves the flexibility and effectiveness of the driving.

[0093] S500, in the frequency range of the selected driving strategy, an unpredictable timing modulation mechanism is set to dynamically adjust the sound and light emission timing and the sound wave frequency;

[0094] The specific steps of dynamically adjusting the sound and light emission timing and the sound wave frequency are as follows:

[0095] S501, after selecting the driving strategy, an unpredictable timing modulation mechanism is set in the sound and light frequency range of the driving strategy, specifically:

[0096] ;

[0097] In the formula, g n+1 represents the (n+1)th sound and light emission time interval, g nThe nth time interval of sound and light emission is represented; the unpredictable time interval of sound and light emission breaks the rule of fixed timing. Birds have learning and adaptation ability, and fixed timing of the repelling means can easily make birds adapt, reducing the repelling effect. Unpredictable timing can effectively avoid this problem and maintain the deterrent of the repelling means.

[0098] The sound and light frequency random walk mechanism is constructed, and the formula is:

[0099] ;

[0100] In the formula, f s (g) The sound and light frequency at time interval g, rand(g) represents a pseudo-random number based on time interval; according to the above unpredictable timing modulation mechanism and sound and light frequency random walk mechanism, sound waves and laser beams are emitted for bird repelling in the frequency interval of the selected repelling strategy according to the random emission time interval and the random sound and light frequency. The frequency random walk mechanism is constructed to make the sound and light frequency randomly change in the effective interval. The randomly changing frequency can stimulate birds from multiple aspects, increase the discomfort of birds, and further improve the repelling effect to avoid birds' resistance to a single frequency.

[0101] S600, the bird repelling failure rate is counted, the bird escape time during each repelling is measured, the resistance inhibition coefficient is calculated, the resistance inhibition coefficient is judged, and the four-dimensional repelling strategy library is updated.

[0102] The specific steps of updating the four-dimensional repelling strategy library are:

[0103] S601, the sound and light cooperative directional rejection system records each repelling, calculates the repelling failure rate, and the formula is: pf=N fail / N total , wherein pf represents the repelling failure rate, N fail represents the number of repelling failure records, and N total represents the total number of repellings; the bird escape time tr during each repelling is measured, the resistance inhibition coefficient is calculated, and the formula is:

[0104] ;

[0105] In the formula, represents the resistance inhibition coefficient, a represents the attenuation coefficient, tr0 represents the initial escape time, and pf0 represents the initial repelling failure rate; the repelling failure rate is calculated, and the bird escape time is measured, the repelling effect is presented in the form of specific data, which is convenient for workers to intuitively understand the effectiveness of the current strategy, and provides an objective basis for judging whether the strategy needs to be updated.

[0106] S602, when When the value is less than 0.9, it is determined that the four-dimensional repulsion strategy library needs to be updated, and a deep learning algorithm is started to update the four-dimensional repulsion strategy library.

[0107] An acoustic-optical directional rejection strategy system for airport safety, the acoustic-optical directional rejection strategy system comprising an acoustic-optical cooperative directional rejection module, a four-dimensional repulsion strategy library module, a resistance index calculation module, a strategy selection module, a random mechanism emission module, and a strategy updating module.

[0108] The acoustic-optical cooperative directional rejection module is configured to set acoustic wave emitters and laser emitters in the airport respectively, to form an acoustic-optical cooperative directional rejection system, and to install different types of sensors in the system to collect state data of birds during bird repulsion.

[0109] The four-dimensional repulsion strategy library module is configured to set different acoustic-optical frequency ranges for different bird sizes, and to mark a danger level, to build a four-dimensional repulsion strategy library.

[0110] The resistance index calculation module is configured to give different weights to different state data, and to fuse different bird state data according to the weights to obtain a bird stress resistance index.

[0111] The strategy selection module is configured to calculate a repulsion threshold according to different safety levels in the four-dimensional repulsion strategy library, to judge the bird stress resistance index using the repulsion threshold, and to select different repulsion strategies.

[0112] The random mechanism emission module is configured to set an unpredictable timing modulation mechanism within the frequency range of the selected repulsion strategy, and to dynamically adjust the acoustic-optical emission timing and the acoustic wave frequency.

[0113] The strategy updating module is configured to measure the bird escape time during each repulsion, to calculate a resistance suppression coefficient, to judge the resistance suppression coefficient, and to update the four-dimensional repulsion strategy library.

[0114] The strategy selection module comprises an upper control limit value updating unit, a repulsion threshold calculation unit, and a strategy selection unit.

[0115] The upper control limit value updating unit is configured to optimize the initial running average value and standard deviation of the resistance index using the resistance index when the resistance index is less than the upper control limit value.

[0116] The repulsion threshold calculation unit is configured to calculate the repulsion threshold using the optimized average value plus standard deviation for each bird size repulsion strategy.

[0117] The strategy selection unit is configured to judge the real-time resistance index using the three repulsion thresholds when performing real-time repulsion on the birds, and to dynamically change the real-time repulsion strategy.

[0118] The random mechanism emitting module includes an unpredictable timing modulation mechanism and an acousto-optic frequency random walk mechanism;

[0119] The unpredictable timing modulation mechanism is used for randomly calculating acousto-optic emitting time intervals;

[0120] The acousto-optic frequency random walk mechanism is used for randomly calculating acousto-optic frequencies.

[0121] Embodiment: Deployment of an acousto-optic directional countermeasure system in an international airport;

[0122] Scenario setting: An international airport needs to protect a runway and surrounding key areas, which are roughly a rectangle with a length of 2000m and a width of 1000m.

[0123] Device parameters: The effective radius of the selected sound wave emitter is r_sound=300m, and the effective radius of the laser emitter is r_laser=400m;

[0124] Calculate the installation spacing and number:

[0125] Take the minimum effective radius: r=min(r_sound, r_laser)=300m.

[0126] According to the formula D=√3×r, calculate the acousto-optic emitting tower spacing: D=1.732*300m≈519.6m. To cover a length of 2000m, the number of towers needed in each row is: N_length=2000 / 519.6≈3.85, rounded up to 4. To cover a width of 1000m, the number of rows needed is: N_width=1000 / (519.6*√3 / 2)≈2.22, rounded up to 3 rows.

[0127] Total number of towers: 4×3=12. According to the honeycomb grid layout, it realizes the all-around coverage of the target airspace, forming the "airspace wall" basic framework.

[0128] Extract historical effective repelling frequencies and construct the following four-dimensional strategy library:

[0129] |Bird size|Sound frequency range (Hz)|Laser frequency range (Hz)|Danger level|

[0130] |Small birds (sparrows)|[8000, 16000]|[5, 10]|Class1|;

[0131] |Large birds (seagulls)|[2000, 8000]|[10, 20]|Class2|;

[0132] |Migratory flocks (flocks of geese)|[500, 2000]|[20, 50]|Class3|;

[0133] One day, the radar detects an invasion of a flock of seagulls (large birds).

[0134] The monitoring radar measures the escape acceleration a = 8.5 m / s2.

[0135] The infrared thermal imager measures the temperature change AT = 1.2 °C.

[0136] The microphone array measures the call frequency shift Af = 350 Hz.

[0137] Assume that the system has an initial escape acceleration reference for seagulls a_0 = 5.0 m / s2. Calculate the real-time resistance index = 1.56; calculate the upper control limit = 1.478, this time R = 1.56 > UCL (1.478), the system immediately determines that the bird has developed resistance;

[0138] The large bird drive-off threshold In2 = μ_0 + σ_0 = 1.25 + 0.114 = 1.364;

[0139] Since R (1.56) > In2 (1.364), the system automatically upgrades the drive-off strategy from the current Class2 to the higher intensity Class3 (migrating flock) strategy.

[0140] It is apparent to those skilled in the art that the present application is not limited to the details of the foregoing exemplary embodiments, and that the present application can be implemented in other particular forms without departing from the spirit or essential characteristics of the present application. Therefore, the embodiments should be considered in all respects as illustrative and not restrictive, the scope of the present application being defined by the appended claims rather than the foregoing description, and it is intended that all changes that come within the meaning and range of equivalency of the claims are embraced therein. Any reference signs in the claims should not be construed as limiting the claims to the figures in which the reference signs are used.

Claims

1. A method for an acoustic-optic directional denial strategy for airport security, characterized by: The method comprises the following steps: S100, respectively setting a sound wave emitter and a laser emitter in an airport to form a sound-light cooperative directional denial system, and installing different types of sensors in the system to collect bird state data when the system drives away the birds; The specific steps of collecting bird state data when the system drives away the birds are: S101, setting a sound-light emission tower in the airport, installing a sound wave emitter and a laser emitter in the sound-light emission tower, additionally installing a monitoring radar, detecting birds in the airport by the monitoring radar, sending a response signal and bird position information to the sound wave emitter and the laser emitter after detecting the birds, and respectively emitting sound waves and laser according to the bird position information after the sound wave emitter and the laser emitter receive the response signal; In the sound-light emission tower, an infrared thermal imager and a microphone array are installed, the infrared thermal imager is used to detect bird temperature data, the microphone array is used to detect bird calling frequency data, all sensors and radars are subjected to space-time calibration, and different types of data collected at the same time and space are outputted; S102、extracting the effective radius of the sound wave transmitter and the laser transmitter in each sound and light emission tower, setting the spacing of the sound and light emission tower, the formula is: D=3 1 / 2 ×r, D represents the spacing of the sound and light emission tower, r represents the effective radius of the sound wave transmitter and the laser transmitter, the effective radius takes the minimum value of the two emission radii of the sound wave transmitter and the laser transmitter; according to the spacing, the sound and light emission tower is set in the airport, the effective coverage range of each sound and light emission tower is regarded as a hexagon by adopting a honeycomb grid coverage model, and an airspace wall is constructed in the airport by using all the sound and light emission towers. S200, collecting bird scales in historical bird driving records of the airport, classifying the bird scales, extracting sound-light frequencies in the driving records, setting different sound-light frequency ranges for different bird scales, marking danger levels, and constructing a four-dimensional driving strategy library; The specific steps of constructing the four-dimensional driving strategy library are: S201, collecting bird scales in historical bird driving records of the airport, classifying the bird scales, collecting the body types of all birds in the driving records, calculating the average value of all bird body types, classifying birds larger than the average value as large birds, classifying birds smaller than the average value as small birds, extracting the minimum value of the number of birds during migration as a migration group threshold value, judging that it is a migration group when the number of birds appearing in the airport at the same time is greater than the migration group threshold value, and finally classifying the bird scales in the airport into large birds, small birds and migration groups; S202、extract the sound wave frequency and laser frequency emitted by the sound wave emitter and the laser emitter in the driving away record for different bird sizes, extract the maximum and minimum to form a frequency interval[p min , p max ];set a danger level for different bird sizes, specifically: small birds are class 1, large birds are class 2, and migrating groups are class 3; construct a four-dimensional driving away strategy as {bird size, frequency interval, danger level}, and integrate the driving away strategies of the three bird sizes to obtain a four-dimensional driving away strategy library; S300, standardizing bird state data collected by the sensors in real time when the birds are driven away, giving different weights to different state data, fusing different bird state data according to the weights to obtain a bird stress resistance index; S400, calculating a driving threshold value according to different safety levels in the four-dimensional driving strategy library, judging the bird stress resistance index by using the driving threshold value, and selecting different driving strategies; S500, setting an unpredictable time sequence modulation mechanism in the frequency range of the selected driving strategy, and dynamically adjusting the sound-light emission time sequence and the sound wave frequency; The specific steps of dynamically adjusting the sound-light emission time sequence and the sound wave frequency are: S501, after selecting a driving strategy, setting an unpredictable time sequence modulation mechanism in the sound-light frequency range of the driving strategy, specifically: ； In the formula, g n+1 represents the (n+1)th time interval of the sound-light emission, g n represents the nth time interval of the sound-light emission; Constructing a sound-light frequency random walk mechanism, the formula is: ； In the formula, f s (g) the sound and light frequency at the time interval g, and rand(g) represents a pseudo-random number based on the time interval; according to the above unpredictable timing modulation mechanism and the sound and light frequency random walk mechanism, sound waves and laser beams are emitted according to the random emission time interval and the random sound and light frequency in the frequency interval selected in the driving strategy to drive away birds. S600, counting the bird driving failure rate, measuring the bird escape time during each driving, calculating a resistance inhibition coefficient, judging the resistance inhibition coefficient, and updating the four-dimensional driving strategy library.

2. The method of acousto-optic directional denial strategy for airport security according to claim 1, characterized in that: The specific steps of S300 are: S301, the motion data of the bird during the bird repelling is tracked by using the monitoring radar, the escape acceleration a of the bird is obtained, the body temperature change value AT of the bird during the bird escaping is detected by using the infrared thermal imager, the calling frequency offset Af of the bird is detected by using the microphone array, and the weights w1, w2 and w3 are set for the three kinds of bird state data respectively; the weight is set by the airport staff according to experience; The three kinds of bird state data detected are standardized, and the standardized bird state data are fused by using the weight, and the formula is: ； In the formula, R represents the bird stress resistance index, and a0 represents the initial escape acceleration of the bird before repelling.

3. The method of claim 2, wherein the method is used for airport security. The specific steps of selecting different repelling strategies in S400 are: S401, the resistance index of each repelling of the three bird sizes in the initial operation of the sound-light cooperative directional rejection system is calculated, the average value and the standard deviation of the resistance index are calculated, the upper control limit value UCL is obtained by using the average value plus the standard deviation, and the resistance index during the repelling is calculated when the sound-light cooperative directional rejection system is operated, when the resistance index is less than the upper control limit value, the average value and the standard deviation of the initial operation of the resistance index are optimized by using the resistance index, and the formula is: ； In the formula, μ(t) represents the average of the resistance index at the tth time of expelling, R t represents the resistance index at the tth time of expelling, μ(t-1) represents the average of the resistance index at the (t-1)th time of expelling, represents the variance of the resistance index at the tth time of expelling, represents the standard deviation at the tth time of expelling, represents the variance of the resistance index at the (t-1)th time of expelling, and λ represents a smoothing factor; the optimized standard deviation is obtained by taking the arithmetic square root of the optimized variance of the resistance index; S402, for the repelling strategy of each bird size, the repelling threshold is calculated by using the optimized average value plus the standard deviation, the repelling threshold of the small bird is In1, the repelling threshold of the large bird is In2, and the repelling threshold of the migration group is In3, when the bird is repelled in real time, the real-time resistance index is judged by using the three repelling thresholds, and the real-time repelling strategy is dynamically changed, and specifically, when the real-time resistance index is less than In1, the class1 strategy is selected, when the real-time resistance index is less than In2 but greater than In1, the class2 strategy is selected, and when the real-time resistance index is less than In3 but greater than In2, the class3 strategy is selected.

4. The method of claim 3, wherein the method is used for airport security. The specific steps of updating the four-dimensional repelling strategy library in S600 are: S601、The sound-light cooperative directional rejection system records each time of repelling, calculates the repelling failure rate, and the formula is: pf=N fail / N total , wherein pf represents the repelling failure rate, N fail represents the number of repelling failure records, N total represents the total number of repelling; measures the bird escape time tr when each time of repelling, calculates the resistance inhibition coefficient, and the formula is: ； In the formula, represents the resistance inhibition coefficient, a represents the attenuation coefficient, tr0 represents the initial escape time, and pf0 represents the initial expulsion failure rate. S602, when When the value is less than 0.9, it is determined that the four-dimensional expulsion strategy library needs to be updated, and the deep learning algorithm is started to update the four-dimensional expulsion strategy library.

5. An acoustic-optic directed denial strategy system for airport security, according to the method of any one of claims 1-4, characterized by: The sound-light directional rejection strategy system includes a sound-light cooperative directional rejection module, a four-dimensional repelling strategy library module, a resistance index calculation module, a strategy selection module, a random mechanism emission module and a strategy updating module; The sound-light cooperative directional rejection module is used for setting sound wave emitters and laser emitters in the airport respectively, forming a sound-light cooperative directional rejection system, and installing different types of sensors in the system to collect the bird state data during the bird repelling by the system; The four-dimensional repelling strategy library module is used for setting different sound-light frequency ranges for different bird sizes, and marking the danger levels to construct a four-dimensional repelling strategy library; The resistance index calculation module is used for giving different weights to different state data, and fusing different bird state data to obtain the bird stress resistance index according to the weight; The strategy selection module is used for calculating the repelling threshold according to different safety levels in the four-dimensional repelling strategy library, judging the bird stress resistance index by using the repelling threshold, and selecting different repelling strategies; The random mechanism emission module is used for setting an unpredictable timing modulation mechanism in the frequency range of the selected repelling strategy, and dynamically adjusting the sound-light emission timing and the sound wave frequency. The strategy updating module is used for measuring bird escape time at each time of driving, calculating a resistance inhibition coefficient, judging the resistance inhibition coefficient, and updating a four-dimensional driving strategy library.

6. The system as claimed in claim 5, wherein the system is used for airport security. The strategy selection module comprises an upper control limit value updating unit, a driving threshold value calculation unit, and a strategy selection unit. The upper control limit value updating unit is used for optimizing an average value and a standard deviation of an initial running resistance index by using the resistance index when the resistance index is less than an upper control limit value. The driving threshold value calculation unit is used for calculating a driving threshold value by using the optimized average value plus the standard deviation for each bird size driving strategy. The strategy selection unit is used for judging a real-time resistance index by using three driving threshold values when driving birds in real time, and dynamically changing a real-time driving strategy.

7. The system as claimed in claim 5, wherein the system is used for airport security. The random mechanism emitting module comprises an unpredictable timing modulation mechanism and an acousto-optic frequency random walk mechanism. The unpredictable timing modulation mechanism is used for randomly calculating acousto-optic emission time intervals. The acousto-optic frequency random walk mechanism is used for randomly calculating acousto-optic frequencies.

Images