Unmanned aerial vehicle and manned aerial vehicle fusion airspace monitoring method based on ADS-B

Abstract

The invention discloses an ADS-B (Automatic Dependent Surveillance-Broadcast)-based unmanned aerial vehicle and manned aerial vehicle fusion airspace monitoring method, and belongs to the technical field of air traffic control. According to the method, attitude vectors of a manned aerial vehicle are collected through a ground ADS-B receiver and an unmanned aerial vehicle broadcast double channel, and relative measurement vectors of a self-sensing sensor of the unmanned aerial vehicle are combined and converted into absolute attitude vectors under a global coordinate system; carrying out weighted fusion on the signal credibility based on the three types of acquisition channels to obtain an accurate attitude vector; extracting position, speed and course angle change sequence features, and calculating various flight intention probabilities; predicting a mixed future position and a covariance based on the intention probability; and finally, constructing a space kernel function and introducing a directional amplification factor to generate an intention risk field. The attitude vector precision is improved through multi-source information fusion, modeling is optimized in combination with the intention probability and the directivity risk amplification mechanism, and fusion airspace monitoring accuracy and risk assessment reliability are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of air traffic control, in particular to a method for monitoring airspace by integrating unmanned aerial vehicles (UAVs) and manned aerial vehicles (MAVs) based on automatic dependent surveillance - broadcast (ADS-B). BACKGROUND

[0002] With the acceleration of low-altitude airspace opening process, the demand for UAVs and MAVs to operate in the same airspace is growing, and the traffic density in the airspace is significantly improved, which puts forward higher real-time and safety requirements for the airspace monitoring system. The existing monitoring system mainly relies on ground ADS-B receivers to obtain the position information and motion state of MAVs, and at the same time, in some areas, self-sensing sensors carried by UAVs are used to make up for the shielding and attenuation of ADS-B signals in low-altitude environment. Although the multi-source monitoring technology improves the monitoring coverage ability to some extent, the accuracy, time delay and noise characteristics of various data are significantly different, and the fusion difficulty is still high.

[0003] In airspace safety management, predicting the future flight state of MAVs and generating a spatial risk distribution is the key to supporting UAV collision avoidance and airspace control. However, the existing risk assessment methods usually only construct static or semi-static risk regions based on the current position, speed or heading information, which cannot effectively utilize the continuous flight state change characteristics of MAVs, and lack the ability to depict the multi-modal future motion trend corresponding to different flight intentions. Since the intention probability and directional characteristics are not introduced in the risk calculation, the risk field generated by the existing technology has limited accuracy and cannot accurately reflect the risk distribution of complex airspace. SUMMARY

[0004] In view of the above problems in the prior art, the present application provides a method for monitoring airspace by integrating UAVs and MAVs based on ADS-B, which solves the problem of low accuracy of the risk field generated by the prior art.

[0005] In order to achieve the above-mentioned application purpose, the technical scheme adopted by the present application is as follows: a method for monitoring airspace by integrating UAVs and MAVs based on ADS-B, comprising the following steps:

[0006] S1, collecting the attitude vector of MAVs through ground ADS-B receivers and UAV broadcast dual channels, and simultaneously collecting the relative measurement vector using the self-sensing sensor of the UAV, and converting the relative measurement vector into the absolute attitude vector of the MAV in the global coordinate system;

[0007] S2, obtaining the signal reliability of the ground ADS-B receiver, the UAV and the self-sensing sensor carried by the UAV respectively, weighting the attitude vectors of the three based on the reliability of the three, and obtaining the accurate attitude vector of the MAV;

[0008] S3, subtracting the accurate manned aircraft attitude vectors of adjacent time to obtain the position change sequence, the speed change sequence and the heading angle change sequence, extracting features from the three change sequences, and obtaining the intention probability of each intention category based on the features;

[0009] S4, predicting the future position under each intention category, and obtaining the mixed prediction position and the mixed prediction covariance based on the intention probability of each intention category;

[0010] S5, constructing a kernel function for any point in space according to the mixed prediction position and the mixed prediction covariance, and introducing a directional amplification factor to generate an intention risk field.

[0011] Further, the attitude vector of the manned aircraft collected by the ground ADS-B receiver in S1 includes the ground decoded position of the manned aircraft, the ground decoded speed of the manned aircraft and the ground decoded heading angle;

[0012] The attitude vector of the manned aircraft collected by the unmanned aerial vehicle includes the unmanned aerial vehicle end decoded position of the manned aircraft, the unmanned aerial vehicle end decoded speed of the manned aircraft and the unmanned aerial vehicle end decoded heading angle;

[0013] The relative measurement vector of the unmanned aerial vehicle collected by the self-sensing sensor carried by the unmanned aerial vehicle includes the relative distance from the unmanned aerial vehicle to the manned aircraft, the azimuth angle and the elevation angle.

[0014] Further, S2 includes the following steps:

[0015] S21, obtaining the signal reliability of the ground ADS-B receiver, the unmanned aerial vehicle and the self-sensing sensor carried by the unmanned aerial vehicle respectively;

[0016] S22, adding the signal reliabilities of the three to obtain the total signal reliability;

[0017] S23, taking the ratio of each signal reliability to the total signal reliability as the signal reliability ratio of each signal;

[0018] S24, multiplying each signal reliability ratio with the corresponding attitude vector to obtain a local attitude vector, and adding the three local attitude vectors to obtain the accurate attitude vector of the manned aircraft.

[0019] Further, the formula for obtaining the signal reliability is:

[0020]

[0021] wherein, is the ith signal reliability, is the reliability level of the ith signal source, the value range is 0-1, is the continuity of the ith signal, Value range 0~1, is the normalized value of the inverse of the update time interval of the i-th signal, is the signal source reliability level weight, is the signal continuity weight, is the update time interval weight, i value range is 1~3, i equals 1 corresponds to the signal of the ground ADS-B receiver acquiring data, i equals 2 corresponds to the signal of the unmanned aerial vehicle acquiring data, and i equals 3 corresponds to the signal of the self-sensing sensor carried by the unmanned aerial vehicle acquiring data.

[0022] Further, S3 includes the following sub-steps:

[0023] S31, subtract the accurate attitude vectors of manned aircraft at adjacent time points to obtain an accurate attitude change sequence of manned aircraft;

[0024] S32, separate the data belonging to position, velocity and heading angle in the accurate attitude change sequence of manned aircraft to form a position change sequence, a velocity change sequence and a heading angle change sequence;

[0025] S33, use a sliding window to slide on the position change sequence, the velocity change sequence and the heading angle change sequence respectively, and extract the mean and standard deviation under each sliding window to obtain a position change feature matrix, a velocity change feature matrix and a heading angle change feature matrix, wherein the first row of the matrix is the mean and the second row is the standard deviation;

[0026] S34, configure corresponding weights and biases for the set of intention categories, and weight the three feature matrices respectively to obtain the linear discriminant score of each intention category;

[0027] S35, process the linear discriminant score using a softmax function to obtain the intention probability of each intention category at each time point.

[0028] Further, S4 includes the following sub-steps:

[0029] S41, predict the future position of straight driving using a uniform straight line model;

[0030] S42, predict the future position of turning using a coordinated turning model;

[0031] S43, predict the future position of climbing / descending using a vertical speed model;

[0032] S44, multiply the intention probability of each intention category with the future position of the corresponding intention category, and add all the multiplication results to obtain a mixed prediction position;

[0033] S45, obtain the mixed prediction covariance according to the difference between the future position of each intention category and the mixed prediction position.

[0034] Furthermore, the formula for the mixed prediction covariance in S45 is:

[0035]

[0036] in, For the first Mixed forecast covariance at the forecast time For the first Predicted time number The prediction covariance matrix for each intent category, For the first Hybrid prediction location at the predicted time For the first Predicted time number Future location of each intent category For the first Time of the first The intent probability of each intent category, where I is the set of intent categories. Numbering for future moments. As a number for historical moments, The number representing the intent category. For transpose operation, for The transpose of .

[0037] Furthermore, S5 includes the following sub-steps:

[0038] S51. Construct a kernel function for any point in space based on the mixed prediction location and the mixed prediction covariance;

[0039] S52. When turning, set the directional magnification factor;

[0040] S53. Multiply the directional amplification factor by the kernel function to obtain the risk of any point in space in the future;

[0041] S54. Take the average risk of all points in any grid in the three-dimensional spatial grid in the future to obtain the risk of that grid, thus forming the intentional risk field.

[0042] Furthermore, the formula for the kernel function is:

[0043]

[0044] in, For any point in space In the Kernel function for predicting time, For the first Hybrid prediction location at the predicted time For the first Mixed forecast covariance at the forecast time det(Cov) is the determinant of the covariance matrix, is the coordinate of any point in space, is the exponential function, is the number of future time, is the number of historical time.

[0045] Further, the formula of the directional amplification factor is:

[0046]

[0047] wherein, is the directional amplification factor when turning, is the forward weighting factor, is the indicator function, is the coordinate of any point in space, is the first is the mixed prediction position of the prediction time, is the unit direction vector corresponding to the manned aircraft heading angle, is the angle between the "vector from any point in space to manned aircraft" and the "heading direction", and max is the maximum of the two.

[0048] The beneficial effects of the present application are:

[0049] 1. The present application calculates the signal credibility of the ground ADS-B receiver, unmanned aerial vehicle broadcast signal and unmanned aerial vehicle self-sensing sensor through step S2, and performs weighted fusion based on the credibility, effectively balances the advantages and defects of data from different sources (such as the wide coverage of ADS-B and the low-altitude blind filling ability of self-sensing sensor), reduces the fusion error caused by data heterogeneity, and finally outputs a high-precision manned aircraft accurate attitude vector.

[0050] 2. The present application extracts the position, speed and heading angle change sequence characteristics of adjacent time through step S3, calculates the probability of different flight intentions (such as straight flight, turning, climbing, etc.) based on continuous motion characteristics, and breaks through the limitation of "only relying on the current state"; step S4 further fuses the future position prediction results under multiple intentions based on the intention probability, obtains the mixed prediction position and mixed prediction covariance, not only covers all possible motion trends, but also quantifies the prediction uncertainty, which is more suitable for the actual scene of manned aircraft dynamic flight compared with the static prediction method.

[0051] 3、The application is based on mixed prediction position and covariance, combined with directional amplification factor to construct kernel function, so that the generated risk field is not only related to the future position probability distribution of manned aircraft, but also reflects the influence of flight direction on risk propagation range. Compared with the existing technology which only generates symmetric and static risk area, the application can present dynamic change and obvious directional risk distribution in airspace over time, which is more in line with the actual unmanned aerial vehicle collision avoidance logic and airspace operation rules, and improves the accuracy of generated risk field.

[0052] 4、Since the risk field generated by the application can reflect the future possible position and uncertainty of manned aircraft in real time, the unmanned aerial vehicle can realize more accurate path planning and risk avoidance on this basis. The application can significantly reduce the probability of air conflict, and has important application value for flight path management, conflict warning and cooperative scheduling in complex mixed airspace. BRIEF DESCRIPTION OF DRAWINGS

[0053] Figure 1 A flowchart of a method for unmanned aerial vehicle and manned aircraft integrated airspace surveillance based on ADS-B. DETAILED DESCRIPTION

[0054] The specific embodiments of the application are described below to facilitate understanding of the application by those skilled in the art, but it should be clear that the application is not limited to the scope of the specific embodiments, and for those skilled in the art, it is obvious that various changes are within the spirit and scope of the application defined and determined by the appended claims, and all applications utilizing the concept of the application are within the scope of protection.

[0055] As shown in Figure 1 A method for unmanned aerial vehicle and manned aircraft integrated airspace surveillance based on ADS-B, comprising the following steps:

[0056] S1, collecting the attitude vector of manned aircraft through ground ADS-B receiver and unmanned aerial vehicle broadcast double channel, and converting the relative measurement vector into absolute attitude vector of manned aircraft in global coordinate system by using unmanned aerial vehicle self-sensing sensor;

[0057] S2, obtaining the signal reliability of ground ADS-B receiver, unmanned aerial vehicle and self-sensing sensor carried by unmanned aerial vehicle respectively, weighting the attitude vectors of the three based on the reliability of the three, and obtaining the accurate attitude vector of manned aircraft;

[0058] S3, subtracting the accurate attitude vectors of manned aircraft at adjacent time to obtain the position change sequence, velocity change sequence and heading angle change sequence, extracting the features of the three change sequences, and obtaining the intention probability of each intention category based on the features;

[0059] S4. Predict the future location for each intent category, and based on the intent probability for each intent category, obtain the mixed predicted location and the mixed predicted covariance.

[0060] S5. Based on the mixed prediction location and mixed prediction covariance, construct a kernel function for any point in space and introduce a directional amplification factor to generate the intentional risk field.

[0061] In this embodiment, the attitude vector of the manned aircraft collected by the ground ADS-B receiver in S1 includes: the manned aircraft position decoded by the ground, the speed of the manned aircraft decoded by the ground, and the heading angle decoded by the ground.

[0062] The attitude vector of the manned aircraft collected by the UAV includes: the manned aircraft position decoded by the UAV, the manned aircraft speed decoded by the UAV, and the heading angle decoded by the UAV.

[0063] The relative measurement vectors collected by the self-sensing sensors on the drone include: the relative distance between the drone and the manned aircraft, the azimuth angle, and the pitch angle.

[0064] S1 specifically involves: acquiring the manned aircraft's attitude vector via a ground-based ADS-B receiver, acquiring the manned aircraft's attitude vector via UAV broadcast, and acquiring the UAV's relative measurement vector via its onboard self-sensing sensors; then converting the relative measurement vector into the manned aircraft's absolute attitude vector in a global coordinate system. ,in, For the first The position of the manned aircraft in the absolute attitude vector at a given moment. It is a homogeneous transformation matrix. For the first The position and attitude of the drone in the global coordinate system at a given moment. To convert the relative measurement vector into a three-dimensional position vector in the UAV coordinate system based on the sensor model. , The relative distance between the drone and the manned aircraft. For pitch angle, It is the azimuth angle. The homogeneous transformation matrix for transforming the UAV coordinate system to the global coordinate system is calculated from the UAV's position and attitude in the global coordinate system.

[0065] The heading angle of the UAV in the absolute attitude vector of the manned aircraft is equal to The manned aircraft's velocity in its absolute attitude vector can be obtained from the position changes at adjacent time points.

[0066] Self-sensing sensors are modules that can measure relative distance, azimuth, and elevation angles, such as millimeter-wave radar.

[0067] In the embodiment, S2 comprises the following sub-steps:

[0068] S21, the ground ADS-B receiver, the unmanned aerial vehicle and the self-sensing sensor carried by the unmanned aerial vehicle respectively acquire signal credibility;

[0069] S22, the signal credibility of the three is added to obtain total signal credibility;

[0070] S23, the ratio of each signal credibility to the total signal credibility is taken as each signal credibility ratio;

[0071] S24, each signal credibility ratio is multiplied by the corresponding attitude vector to obtain a local attitude vector, and the three local attitude vectors are added to obtain the accurate attitude vector of the manned aircraft.

[0072] In the embodiment, the formula for acquiring signal credibility is:

[0073]

[0074] wherein, is the i-th signal credibility, is the i-th signal source reliability level, the value range is 0-1, is the i-th signal continuity, the value range is 0-1, is the normalized value of the inverse of the i-th signal update time interval, is the signal source reliability level weight, is the signal continuity weight, is the update time interval weight, i takes the value range of 1-3, i equals 1 corresponding to the signal of the ground ADS-B receiver acquiring data, i equals 2 corresponding to the signal of the unmanned aerial vehicle acquiring data, and i equals 3 corresponding to the signal of the self-sensing sensor carried by the unmanned aerial vehicle acquiring data.

[0075] In the embodiment, the source reliability level is: ground ADS-B receiver > unmanned aerial vehicle > self-sensing sensor carried by unmanned aerial vehicle.

[0076] The normalization formula is:

[0077]

[0078] wherein, is the inverse of the i-th signal update time interval, is the minimum value of the inverse of the signal update time interval, is the maximum value of the inverse of the signal update time interval.

[0079] The application avoids one-dimension (such as only looking at the source) by calculating the reliability in multiple dimensions (signal source reliability level + continuity + update frequency) to avoid one-dimension (such as only looking at the source); meanwhile, the ground ADS-B, unmanned aerial vehicle broadcast and self-sensing sensor are differentially weighted according to their characteristics (such as the ground ADS-B source level is higher) to make the integrated manned aircraft accurate attitude vector closer to the real state.

[0080] In the embodiment, according to the data source characteristics of the low-altitude monitoring scene, the example value range is: the ground ADS-B receiver R1 is set to 0.8, the unmanned aerial vehicle broadcast channel R2 is set to 0.6; and the unmanned aerial vehicle self-sensing sensor R3 is set to 0.4.

[0081] The setting rule of signal continuity is to take “successfully obtaining signals at continuous time points” as the judgment basis, and the example value is: continuously obtaining signals at ≥5 time points: S i set to 0.9 (signal stable and continuous); continuously obtaining signals at 3-4 time points: S i set to 0.6; continuously obtaining signals at 1-2 time points: S i set to 0.3; only obtaining signals at 1 time point: S i set to 0.1.

[0082] In the embodiment, the formula of the manned aircraft accurate attitude vector is:

[0083]

[0084] wherein, is the manned aircraft accurate attitude vector, is the i-th attitude vector, is the i-th signal reliability ratio.

[0085] In the embodiment, S3 includes the following steps:

[0086] S31, subtracting the adjacent time points of the manned aircraft accurate attitude vector to obtain the manned aircraft accurate attitude change sequence, that is, subtracting the manned aircraft accurate attitude vector at the t+1 time point from the t time point;

[0087] S32, separating the data belonging to the position, speed and heading angle in the manned aircraft accurate attitude change sequence to form the position change sequence, the speed change sequence and the heading angle change sequence;

[0088] S33, sliding the sliding window on the position change sequence, the speed change sequence and the heading angle change sequence respectively, and extracting the mean and standard deviation under each sliding window to obtain the position change feature matrix, the speed change feature matrix and the heading angle change feature matrix, wherein the first row of each matrix is the mean of each window, and the second row is the standard deviation of each window;

[0089] S34. Assign corresponding weights and biases to the intent category set, and then weight the three feature matrices to obtain the linear discrimination score for each intent category:

[0090]

[0091] in, For the first The first moment Linear discriminant scores for each intent category, For the first The weight matrix of each intent category to the k-th feature matrix. For the first The bias matrix of each intent category with respect to the k-th feature matrix. k takes values ​​from 1 to 3, corresponding to the th... The feature matrices for position changes, velocity changes, and heading angle changes at any given moment are included. The intent categories in the intent category set include: straight line, turn, climb, and descent.

[0092] S35. The softmax function is used to process the linear discriminant scores to obtain the intent probability of each intent category at each time step:

[0093]

[0094] in, For the first Time of the first The probability of intent for each intent category. Let I be an exponential function, and let I be the set of intent categories.

[0095] This invention effectively captures the changing patterns of position, speed, and heading angle through feature engineering using a sliding window to extract the mean and standard deviation (e.g., the standard deviation of the heading angle increases significantly when turning). Combined with the logic of "multi-feature matrix weighting + softmax probability transformation", it can accurately map different motion features to intention categories, reduce the possibility of misjudgment of a single feature, and make the recognition results of intentions such as straight lines and turns more reliable.

[0096] In this embodiment, S4 includes the following sub-steps:

[0097] S41. Predict the future position of straight-line travel using a uniform linear model:

[0098]

[0099]

[0100] in, For the first Predict the future x, y positions of the line moving straight ahead at a given moment. is the x, y position of the accurate pose vector of the manned vehicle at the t th moment, is the horizontal velocity magnitude, is the number of future moments, is the unit direction vector corresponding to the heading angle of the manned vehicle, is the height of the accurate pose vector of the manned vehicle at the t th moment, is the x, y position of the accurate pose vector of the manned vehicle at the t th moment, is the future height position of the straight-ahead at the prediction moment, the future x, y position and height together constitute the future position of the straight-ahead;

[0101] S42, predict the future position of the turn using the coordinated turn model:

[0102]

[0103]

[0104]

[0105] wherein, is the heading angle at the t th moment, is the heading angle at the prediction moment, is the heading angle at the t th moment, is the turn angular velocity, is the x, y position of the accurate pose vector of the manned vehicle at the t th moment, is the future x, y position of the turn at the prediction moment, is the future height position of the turn at the prediction moment, the future x, y position and height together constitute the future position of the turn; S43, predict the future position of the climb / descent using the vertical velocity model:

[0106]

[0107]

[0108] wherein, is the future position of the climb / descent at the prediction moment, is the vertical velocity at the t th moment, is the vertical acceleration; S44, multiply the intention probability of each intention category with the future position of the corresponding intention category, and add all the multiplication results to obtain the mixed prediction position:

[0109]

[0110] ​​​​

[0111] in, For the first Hybrid prediction location at the predicted time For the first Future location of each intent category For the first Time of the first The intent probability of each intent category, where I is the set of intent categories. Numbering for future moments. As a number for historical moments, The number representing the intent category;

[0112] S45. Obtain the mixed prediction covariance based on the difference between the future location of each intent category and the mixed prediction location.

[0113] In this embodiment, the formula for the mixed prediction covariance in S45 is:

[0114]

[0115] in, For the first Mixed forecast covariance at the forecast time For the first Predicted time number The prediction covariance matrix for each intent category, For the first Hybrid prediction location at the predicted time For the first Predicted time number Future location of each intent category For the first Time of the first The intent probability of each intent category, where I is the set of intent categories. Numbering for future moments. As a number for historical moments, The number representing the intent category. For transpose operation, for The transpose of .

[0116] For the straight line intention: covariance matrix This primarily covers horizontal velocity measurement errors and uncertainties in x and y position predictions; regarding turning intentions: covariance matrix. This primarily covers the uncertainty of angular velocity and the uncertainty of horizontal position (x, y) prediction; regarding climb / descent intentions: covariance matrix. It mainly covers vertical velocity measurement errors, acceleration prediction uncertainties, and height position prediction uncertainties.

[0117] The present application can precisely match the laws of different motion states by adopting professional models (uniform straight line / coordinated turning / vertical speed model) suitable for three typical motion intentions of "straight, turning, climbing / descending", avoid the prediction deviation of a single model for complex motion, and make the position prediction under each intention more in line with the actual motion characteristics. The present application obtains a hybrid prediction position through "intention probability weighted fusion", and calculates a hybrid covariance combining the deviation between the intention position and the hybrid position, which comprehensively integrates the prediction information of all possible intentions and quantifies the uncertainty brought by different intentions, so that the final prediction result can adapt to the dynamic scene of "multi-intention switching" and reduce the prediction failure risk caused by single intention misjudgment.

[0118] In the present embodiment, S5 comprises the following sub-steps:

[0119] S51, constructing a kernel function for an arbitrary point in space according to the hybrid prediction position and the hybrid prediction covariance;

[0120] S52, setting a directional amplification factor when turning;

[0121] S53, multiplying the directional amplification factor with the kernel function to obtain the risk of the arbitrary point in space in the future;

[0122] S54, taking the average of the risks of all points in any grid in the three-dimensional grid in space in the future to obtain the risk of the grid, arranging the risks of each grid according to the grid positions to form an intention risk field.

[0123] In the present embodiment, the formula of the kernel function is:

[0124]

[0125] wherein, is an arbitrary point in space is the kernel function at the prediction moment, is the hybrid prediction position at the prediction moment, is the hybrid prediction covariance at the prediction moment, is the determinant of the covariance matrix, is the coordinate of the arbitrary point in space, is an exponential function, is the number of the future moment, is the number of the historical moment, is the inverse of

[0126] In the present embodiment, the formula of the directional amplification factor is:

[0127] ​​​​

[0128] wherein, is the directional amplification factor in the turning state, is the forward weighting factor, is an indicator function, which is 1 when the intention of the UAV is "turning", and 0 otherwise, is the coordinate of the spatial arbitrary point, is the first is the mixed prediction position at the prediction moment, is the unit direction vector corresponding to the heading angle of the UAV, is the angle between the "vector from the spatial arbitrary point to the UAV" and the "heading direction", and max is the maximum of the two.

[0129] The Gaussian kernel function constructed by "mixed prediction position + covariance" (S51) quantifies the spatial uncertainty of the position prediction - the more dispersed the probability distribution of the future position of the UAV (the larger the covariance), the higher the kernel function value (the basic risk) of the corresponding region, which conforms to the actual logic that "the risk of uncertain regions is higher". The directional amplification factor in the turning state (S52-S53) specifically strengthens the risk weight of the front region of the heading direction, matches the actual scene that "the front is the direction of motion trend during the turning process, and the risk needs to be paid more attention to", and avoids the deviation of the non-discriminatory evaluation of the whole region.

[0130] The above is only a preferred embodiment of the present application and is not used to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for ADS-B based UAV integration into manned airspace surveillance, the method comprising: The method comprises the following steps: S1, collecting the manned aircraft attitude vector through the ground ADS-B receiver, the unmanned aircraft broadcasting dual channel, collecting the relative measurement vector by using the self-sensing sensor of the unmanned aircraft, and converting the relative measurement vector into the absolute attitude vector of the manned aircraft in the global coordinate system; S2, obtaining the signal reliability of the ground ADS-B receiver, the unmanned aircraft and the self-sensing sensor carried by the unmanned aircraft respectively, weighting the attitude vectors of the three based on the reliabilities to obtain the accurate attitude vector of the manned aircraft; S3, subtracting the accurate attitude vectors of the manned aircraft at adjacent time points to obtain the position change sequence, the speed change sequence and the heading angle change sequence, extracting the features of the three change sequences, and obtaining the intention probability of each intention category based on the features; S4, predicting the future position under each intention category, and obtaining the mixed prediction position and the mixed prediction covariance based on the intention probability of each intention category; S5, constructing the kernel function for any point in space according to the mixed prediction position and the mixed prediction covariance, and introducing the directional amplification factor to generate the intention risk field. 2.The ADS-B based UAV integration airspace surveillance method of claim 1, wherein, The attitude vector of the manned aircraft collected by the ground ADS-B receiver comprises the ground decoded position of the manned aircraft, the ground decoded speed of the manned aircraft and the ground decoded heading angle; The attitude vector of the manned aircraft collected by the unmanned aircraft comprises the unmanned aircraft end decoded position of the manned aircraft, the unmanned aircraft end decoded speed of the manned aircraft and the unmanned aircraft end decoded heading angle; The relative measurement vector of the unmanned aircraft collected by the self-sensing sensor carried by the unmanned aircraft comprises the relative distance from the unmanned aircraft to the manned aircraft, the azimuth angle and the elevation angle. 3.The ADS-B based UAV integration airspace surveillance method of claim 1, wherein, S2 comprises the following steps: S21, obtaining the signal reliability of the ground ADS-B receiver, the unmanned aircraft and the self-sensing sensor carried by the unmanned aircraft respectively; S22, adding the signal reliabilities of the three to obtain the total signal reliability; S23, taking the ratio of each signal reliability to the total signal reliability as the signal reliability ratio of each signal; S24, multiplying each signal reliability ratio with the corresponding attitude vector to obtain the local attitude vector, and adding the three local attitude vectors to obtain the accurate attitude vector of the manned aircraft.

4. The ADS-B based UAV integration airspace surveillance method according to claim 1 or 3, characterized in that, The formula for obtaining the signal reliability is: wherein, is the reliability of the ith signal, is the reliability level of the ith signal source, with a value range of 0~1, is the continuity of the ith signal, with a value range of 0~1, is the normalized value of the inverse of the ith signal update time interval, is the signal source reliability level weight, is the signal continuity weight, is the update time interval weight, i has a value range of 1~3, i equals 1 corresponds to the signal of the ground ADS-B receiver acquiring data, i equals 2 corresponds to the signal of the unmanned aerial vehicle acquiring data, and i equals 3 corresponds to the signal of the self-sensing sensor carried by the unmanned aerial vehicle acquiring data.

5. The ADS-B based UAV integration with manned aircraft airspace surveillance method of claim 1, wherein, S3 comprises the following steps: S31, subtracting the accurate attitude vectors of the manned aircraft at adjacent time points to obtain the accurate attitude change sequence of the manned aircraft; S32, separating the data belonging to the position, the speed and the heading angle in the accurate attitude change sequence of the manned aircraft to form the position change sequence, the speed change sequence and the heading angle change sequence; S33, sliding the sliding window on the position change sequence, the speed change sequence and the heading angle change sequence respectively, and extracting the mean and the standard deviation under each sliding window to obtain the position change feature matrix, the speed change feature matrix and the heading angle change feature matrix, wherein the first row of the matrix is the mean and the second row is the standard deviation; S34, configuring the corresponding weight and bias for the intention category set, weighting the three feature matrices respectively to obtain the linear discriminant score of each intention category; S35, processing the linear discriminant score by using the softmax function to obtain the intention probability of each intention category at each time point.

6. The ADS-B based UAV integration with manned aircraft airspace surveillance method of claim 1, wherein, S4 comprises the following sub-steps: S41, predicting the future position of straight running by using uniform linear model; S42, predicting the future position of turning by using coordinated turning model; S43, predicting the future position of climbing / descending by using vertical speed model; S44, multiplying the intention probability of each intention category with the future position of the corresponding intention category, and adding all the multiplication results to obtain the mixed prediction position; S45, obtaining the mixed prediction covariance according to the difference between the future position of each intention category and the mixed prediction position.

7. The ADS-B based UAV integration with manned aircraft airspace surveillance method of claim 6, wherein, The formula of the mixed prediction covariance in S45 is: wherein, is the is the is the is the is the is the is the is the is the is the is the is the is the is the number of future time instances, is the number of history time instances, is the number of intention classes, is the transpose operation, is the transpose of . 8.The ADS-B based UAV integration with manned aircraft airspace surveillance method of claim 1, wherein, S5 comprises the following sub-steps: S51, constructing a kernel function for any point in space according to the mixed prediction position and the mixed prediction covariance; S52, setting a directional amplification factor when turning; S53, multiplying the directional amplification factor with the kernel function to obtain the risk of any point in space in the future; S54, taking the average of the risks of all points in any grid in the three-dimensional grid in space in the future to obtain the risk of the grid, thereby constituting the intention risk field. 9.The ADS-B based UAV integration airspace surveillance method of claim 8, wherein, The formula of the kernel function is: wherein, for an arbitrary point in space at the kernel function at the prediction time, for the mixed prediction position at the prediction time, for the mixed prediction covariance at the prediction time, for the determinant of the covariance matrix, for the coordinates of an arbitrary point in space, for the exponential function, for the number of future time instances, for the number of history time instances.

10. The ADS-B based UAV integration with manned aircraft airspace surveillance method of claim 8, wherein, The formula of the directional amplification factor is: wherein, is a directional amplification factor in a turning state, is a forward weighting factor, is an indicator function, is a coordinate of a spatial arbitrary point, is a first is a mixed prediction position at a prediction time, is a unit direction vector corresponding to a manned vehicle heading angle, is an angle between "a vector from a spatial arbitrary point to a manned vehicle" and "a heading forward direction", and max is a maximum value of the two.

Images