Unmanned aerial vehicle situation awareness method based on radio technology

Abstract

The invention discloses an unmanned aerial vehicle situation awareness method and system based on a radio technology. The method comprises the following steps that S1, a receiving node receives a radio signal transmitted by an unmanned aerial vehicle and processes the received signal into a snapshot matrix; s2, inputting the snapshot matrix into a covariance matrix, decomposing the covariance matrix to obtain a noise subspace matrix, inputting the noise subspace matrix and an array flow vector into a MUSIC spectrum, and searching a peak value of PMUSIC (theta) to obtain estimation of each incident angle; s3, obtaining an observation vector according to the incident angle, and determining the initial coordinate of the unmanned aerial vehicle through the observation vector, the initial coordinate of the receiving node and a least square method; s4, determining an initial state vector according to the initial coordinate of the unmanned aerial vehicle; s5, iteratively calculating the current position and speed of the unmanned aerial vehicle at the moment k according to the covariance matrix of the initialized state vector, and outputting the current position and speed as the current situation; the method is high in detection precision, strong in anti-interference capability and good in adaptability.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) situational awareness technology, and in particular to a UAV situational awareness method based on radio technology. Background Technology

[0002] The core of UAV situational awareness is target localization and status monitoring. Current mainstream technologies include GPS positioning, visual positioning, and radio positioning. Among these, radio positioning is widely used in complex environments due to its unobstructed and all-weather operation (citation: *Radio Positioning Principles and Applications*, Electronic Industry Press, 2020). Angle of Arrival (AOA) algorithm, as one of the core methods of radio positioning, achieves positioning by measuring the incident angle of radio signals, offering advantages such as low hardware complexity and fast response speed.

[0003] The shortcomings of existing technology are:

[0004] - Large angle measurement error: Traditional AOA algorithms rely on a single antenna or simple array, which is affected by multipath interference. The incident angle measurement error is usually greater than ±3°, resulting in insufficient positioning accuracy.

[0005] - Weak anti-interference capability: In complex electromagnetic environments, noise and multipath signals superimposed can easily lead to the failure of angle estimation;

[0006] - Simplified positioning model: Existing algorithms mostly use ideal geometric models and do not consider the impact of the UAV's motion state (speed, acceleration) on the positioning results, resulting in significant dynamic positioning errors.

[0007] A search revealed that Chinese patent application number 202010048409.6 discloses a method for integrated navigation of unmanned aerial vehicles (UAVs). This patent prevents filter divergence and adjusts the filter in a timely manner by performing positive definiteness detection on the covariance matrix of the filter in real time. Summary of the Invention

[0008] To address the technical problems existing in the background art, this invention proposes a UAV situational awareness method based on radio technology.

[0009] The present invention proposes a UAV situational awareness method based on radio technology, comprising the following steps:

[0010] S1. The receiving node receives the radio signals transmitted by the drone and processes the received signals into a snapshot matrix. ;

[0011] S2. Input the snapshot matrix into the covariance matrix and decompose the covariance matrix to obtain the noise subspace matrix. Input the noise subspace matrix and the array flow vector into the MUSIC spectrum to search for P. MUSICThe peak value of (θ) is used to estimate the incidence angles.

[0012] S3. Obtain the observation vector based on the incident angle, and determine the initial coordinates of the UAV using the observation vector, the initial coordinates of the receiving node, and the least squares method.

[0013] S4. Determine the initial state vector based on the initial coordinates of the UAV;

[0014] S5. Iteratively calculate the current position and velocity of the UAV at time k based on the covariance matrix of the initialized state vector, and output it as the current situation.

[0015] Specifically, the receiving node adopts an N-element uniform linear array with an element spacing d = λ / 2, where λ is the wavelength of the radio signal and N is an integer greater than 2.

[0016] Specifically, the covariance matrix in step S2 is

[0017] Where K is the number of sampling points, for The conjugate transpose of;

[0018] The noise subspace Un obtained from covariance matrix decomposition and the array flow vector are input into the MUSIC spectral function to extract the incident signal angles, obtaining the incident angles θ1, θ2...θ at each node. M .

[0019] Specifically, the MUSIC spectral function is:

[0020] Where: Un is the noise subspace of the array received signal covariance matrix; the superscript H denotes the conjugate transpose; α(θ) is the array flow vector; and the array manifold vector is:

[0021] α(θ)=[1,e −j2πdsinθ / λ ,e −j4πdsinθ / λ ,...,e −j2π(N−1)dsinθ / λ ] T In the formula, N is the number of array elements, d is the spacing between array elements, and λ is the signal wavelength.

[0022] Step S5 includes: using the best estimated state vector of the UAV at time k-1. Best covariance at the previous moment Predicting the state at the next moment and predict covariance matrix ;

[0023] Based on the observation function, global orientation angle observations, and predicted state Calculate the optimal state estimate for the current time step using the predicted covariance matrix. and the current best covariance matrix .

[0024] Specifically, ;

[0025] ;

[0026] F is the state transition matrix, and Q is the process noise covariance;

[0027] F is the state transition matrix, and Q is the process noise covariance;

[0028] F= T is the time step.

[0029] Specifically, the Jacobian matrix H of the observation matrix is ​​calculated. k :

[0030] ;

[0031] ;

[0032] In this embodiment: ,

[0033] x1 and y1 are the coordinates of the first receiving point, x2 and y2 are the coordinates of the second receiving point, x3 and y3 are the coordinates of the third receiving point, and x and y are the coordinates of the UAV.

[0034] Calculate the Kalman gain K k :

[0035] ;

[0036] H k It is the observation function in the predicted state The Jacobian matrix is ​​given by R, where R is the observation noise covariance.

[0037]

[0038]

[0039] get: ;

[0040] The current optimal covariance matrix is: .

[0041] A UAV situational awareness system based on radio technology includes a radio signal transmitter installed on the UAV and a receiving node deployed on the ground, wherein the information transmitted by the radio signal transmitter is received by the receiving node.

[0042] It also includes a signal processing module. The information received by the receiving node is transmitted to the signal processing module, which implements the corresponding program module in steps S1-S5 of any of the above schemes.

[0043] The UAV situational awareness method based on radio technology proposed in this invention has the following advantages:

[0044] 1. The angle measurement error of the MUSIC algorithm is ≤ ±0.5°. Combined with EKF dynamic correction, the static positioning error is ≤ 1m and the dynamic positioning error is ≤ 3m.

[0045] 2. Enhanced anti-interference capability: Array signal processing and noise subspace separation technology effectively suppress multipath interference and electromagnetic noise;

[0046] 3. Good dynamic adaptability: The EKF model adapts to the movement state of the UAV and can track high-speed moving targets in real time.

[0047] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0048] Figure 1 This describes the distribution of receiving nodes in an embodiment of the present invention.

[0049] Figure 2 This is a flowchart of the method of the present invention. Detailed Implementation

[0050] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar symbols denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0051] like Figure 1 - Figure 2 The method for UAV situational awareness based on radio technology, as shown, includes the following steps:

[0052] Implementation conditions: The UAV is equipped with a radio transmitter, and three receiving nodes are set up as receiving node A, receiving node B, and receiving node C. Each receiving node is connected to the data processing center via a wireless communication link. The receiving nodes adopt an 8-element uniform linear array (N=8), with an element spacing d=0.125m (adapted to 2.4GHz radio signals, λ=0.25m, satisfying the d=λ / 2 design); the array elements adopt omnidirectional antennas (gain ≥2dBi), and the signal processing module is equipped with an FPGA chip (model: Xilinx XC7K325T) and an ARM processor (model: STM32H743); the UAV is equipped with a 2.4GHz radio transmitter (transmit power 10dBm, signal bandwidth 2MHz);

[0053] Software environment: The signal processing algorithm is developed based on MATLAB 2023b, and the embedded software is written in C language. EKF filter parameter settings: process noise covariance Q=diag([1e-4, 1e-4, 1e-6, 1e-6]), observation noise covariance R=diag([(0.5°×π / 180)). 2 , (0.5°×π / 180) 2 , (0.5°×π / 180) 2 ]);

[0054] Deployment of receiving nodes: Three receiving nodes are deployed at coordinates A (0,0), B (200m,0), and C (100m,200m), respectively, with the array normal direction deflection angles α1=0°, α2=90°, and α3=180° for each node.

[0055] S1. Three receiving nodes simultaneously receive the radio signals emitted by the drone and convert the received radio signals into digital sampling data through an ADC module. The receiving nodes receive the radio signals emitted by the drone and process the received signals into a snapshot matrix. ;

[0056] S2. Input the snapshot matrix into the covariance matrix and decompose the covariance matrix to obtain the noise subspace matrix. Input the noise subspace matrix and the array flow vector into the MUSIC spectrum search to obtain the estimates of each incident angle.

[0057] The covariance matrix is:

[0058] ;

[0059] Where K is the number of sampling points, and in this embodiment K is 1024. Let be the vector formed by the array data received by the i-th node. for The conjugate transpose of;

[0060] ;

[0061] In the formula: m = 1, 2, 3, ..., K; Let be the i-th component of the array manifold vector α(θ); in this embodiment, there is only one signal source, therefore, For e −j2π(i−1)dsinθ / λ ; It is the signal complex envelope. It is additive white Gaussian noise.

[0062] Decomposing the covariance matrix yields eigenvalues ​​λ1≥λ2≥...≥λN and corresponding eigenvectors [v1,v2,...,v...]. N In this embodiment, N=8. Therefore, the noise subspace of this embodiment is Un=[v2,v3,...,v8];

[0063] The signal subspace and noise subspace are separated by eigenvalue decomposition and substituted into the MUSIC spectrum function to extract the incident signal angle, thereby obtaining the incident angles θ1, θ2, and θ3 of each node;

[0064] Specifically, the MUSIC spectral function is:

[0065] ;

[0066] Where: Un is the noise subspace of the array received signal covariance matrix; the superscript H denotes the conjugate transpose, α(θ) is the array flow vector, and the array flow vector is:

[0067] α(θ)=[1,e −j2πdsinθ / λ ,e −j4πdsinθ / λ ,...,e −j2π(N−1)dsinθ / λ ] T N represents the number of array elements, which is 8 in this embodiment.

[0068] θ is the signal incident angle (angle with the array normal direction), j is the imaginary unit, N is the number of array elements (N is 8 in this embodiment), d is the element spacing, and λ is the signal wavelength. Calculate the signal incident angles θ1, θ2, and θ3 using α(θ);

[0069] S3. Obtain the observation vector based on the incident angle, and determine the initial coordinates of the UAV using the observation vector, the initial coordinates of the receiving node, and the least squares method.

[0070] Specifically, calculate the global azimuth angle φ. i , φ i =θi+α i ;

[0071] Node A: φ1 = θ1 + 0;

[0072] Node B: φ2 = θ2 + 90°;

[0073] Node C: φ3 = θ3 + 180°;

[0074] The data processing center receives global azimuth angles φ1, φ2, and φ3. 3， and constitute ;

[0075] For each node i, based on its coordinates (x... i ,y i Construct the equation of the straight line:

[0076] ;

[0077] And convert the above formula to ;

[0078] Construct an overdetermined system of linear equations and solve for the initial position of the UAV using the least squares method:

[0079] ,in:

[0080] ;

[0081] ;

[0082] Thus, the initial position (x0, y0) of the drone is obtained.

[0083] S4. Determine the initial state vector based on the initial coordinates of the UAV;

[0084] The initial state vector is: ;

[0085] User-defined initial covariance:

[0086]

[0087] S5. Iteratively calculate the current position and velocity of the UAV at time k based on the covariance matrix of the initialized state vector, and output it as the current situation.

[0088] Specifically, based on the above model, the state and uncertainty of the UAV at the next moment are predicted: based on the best estimated state vector of the UAV at moment k-1. Best covariance at the previous moment Predict the predicted state and predicted covariance matrix at the next time step;

[0089] ;

[0090] ;

[0091] F is the state transition matrix, and Q is the process noise covariance;

[0092] F= Where T is the time step;

[0093] Then calculate the predicted state at the current time. and the prediction covariance matrix at the current time ;

[0094] Based on the observation function, global orientation angle observations, and predicted state Calculate the best state estimate and the current best covariance matrix at the current time step using the predicted covariance matrix:

[0095] Calculate the Jacobian matrix H of the observation matrix k :

[0096] ;

[0097] ;

[0098] In this embodiment: ,

[0099] x1 and y1 are the coordinates of the first receiving point, x2 and y2 are the coordinates of the second receiving point, x3 and y3 are the coordinates of the third receiving point, and x and y are the coordinates of the UAV.

[0100] Calculate the Kalman gain K k :

[0101] ;

[0102] H k It is the observation function in the predicted state The Jacobian matrix is ​​given by R, where R is the observation noise covariance.

[0103]

[0104]

[0105] get: ;

[0106] The current optimal covariance matrix is: .

[0107] A UAV situational awareness system based on radio technology includes a radio signal transmitter installed on the UAV and a receiving node arranged on the ground for receiving signals transmitted by the radio signal transmitter. The receiving node is a linear array with uniform array elements.

[0108] It also includes a signal processing module. The signals received by the receiving node are transmitted to the signal processing module, which implements the corresponding program modules in steps S1-S5.

[0109] Implementation results: When the actual coordinates of the UAV are (100m, 100m), the measured θ1=45.2°, θ2=134.8°, and θ3=225.1° are as follows. After EKF filtering, the angle error is ≤±0.3°. The final output situational awareness positioning coordinates are (100.8m, 99.5m), with a positioning error of 0.93m, which meets the requirements of high-precision situational awareness.

[0110] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A situational awareness method for unmanned aerial vehicles (UAVs) based on radio technology, characterized in that, Includes the following steps: S1. The receiving node receives the radio signals transmitted by the drone and processes the received signals into a snapshot matrix. ; S2. Input the snapshot matrix into the covariance matrix and decompose the covariance matrix to obtain the noise subspace matrix. Input the noise subspace matrix and the array flow vector into the MUSIC spectrum to search for P. MUSIC The peak value of (θ) is used to estimate the incidence angles. S3. Obtain the observation vector based on the incident angle, and determine the initial coordinates of the UAV using the observation vector, the initial coordinates of the receiving node, and the least squares method. S4. Determine the initial state vector based on the initial coordinates of the UAV; S5. Iteratively calculate the current position and velocity of the UAV at time k based on the covariance matrix of the initialized state vector, and output it as the current situation.

2. The UAV situational awareness method based on radio technology according to claim 1, characterized in that, The receiving node adopts an N-element uniform linear array with an element spacing d = λ / 2, where λ is the wavelength of the radio signal and N is an integer greater than 2.

3. The UAV situational awareness method based on radio technology according to claim 1, characterized in that, The covariance matrix in step S2 is ; Where K is the number of sampling points, for The conjugate transpose of; The noise subspace Un obtained from covariance matrix decomposition and the array flow vector are input into the MUSIC spectral function to extract the incident signal angles, obtaining the incident angles θ1, θ2...θ at each node. M .

4. The UAV situational awareness method based on radio technology according to claim 1, characterized in that, The MUSIC spectral function is: ; Where: Un is the noise subspace of the array received signal covariance matrix; the superscript H denotes the conjugate transpose; α(θ) is the array flow vector; and the array manifold vector is: α(θ)=[1,e −j2πdsinθ / λ ,e −j4πdsinθ / λ ,...,e −j2π(N−1)dsinθ / λ ] T In the formula, N is the number of array elements, d is the spacing between array elements, and λ is the signal wavelength.

5. The UAV situational awareness method based on radio technology according to claim 1, characterized in that, Specifically, in step S3: Calculate the global azimuth angle φ i φ i =θ i +α i ; α i Let be the deflection angle of the array normal direction at node i; For each node i, based on its coordinates (x... i ,y i Construct the equation of the straight line: ; And convert the above formula to ; Construct an overdetermined system of linear equations and solve for the initial position of the UAV using the least squares method: ,in: ; ; Thus, the initial position (x0, y0) of the drone is obtained.

6. The UAV situational awareness method based on radio technology according to claim 1, characterized in that, Step S5 includes: using the best estimated state vector of the UAV at time k-1. Best covariance at the previous moment Predicting the state at the next moment and the predicted covariance matrix ; Based on the observation function, global orientation angle observations, and predicted state Calculate the optimal state estimate for the current time step using the predicted covariance matrix. and the current best covariance matrix .

7. The UAV situational awareness method based on radio technology according to claim 6, characterized in that, ； ; F is the state transition matrix, and Q is the process noise covariance; F is the state transition matrix, and Q is the process noise covariance; F= T is the time step.

8. The UAV situational awareness method based on radio technology according to claim 6, characterized in that, Calculate the Jacobian matrix H of the observation matrix k : ; ; In this embodiment: ; x1 and y1 are the coordinates of the first receiving point; x2 and y2 are the coordinates of the second receiving point; x3 and y3 are the coordinates of the third receiving point, and x and y are the coordinates of the UAV. Calculate the Kalman gain K k : ; H k It is the observation function in the predicted state The Jacobian matrix is ​​given by R, where R is the observation noise covariance. get: ; The current optimal covariance matrix is: .

9. A situational awareness system for unmanned aerial vehicles (UAVs) based on radio technology, characterized in that, It includes a radio signal transmitter installed on a drone, and receiving nodes arranged on the ground to receive signals transmitted by the radio signal transmitter. The receiving nodes are linear arrays with uniform array elements. It also includes a signal processing module, to which the signal received by the receiving node is transmitted. The signal processing module implements the program module corresponding to steps S1-S5 as described in any one of claims 1-7.

Images