CN115685067A - Normal-mode signal blind estimation method and system for positioning and tracking of multi-rotor unmanned aerial vehicle - Google Patents
Normal-mode signal blind estimation method and system for positioning and tracking of multi-rotor unmanned aerial vehicle Download PDFInfo
- Publication number
- CN115685067A CN115685067A CN202211385592.4A CN202211385592A CN115685067A CN 115685067 A CN115685067 A CN 115685067A CN 202211385592 A CN202211385592 A CN 202211385592A CN 115685067 A CN115685067 A CN 115685067A
- Authority
- CN
- China
- Prior art keywords
- unmanned aerial
- aerial vehicle
- rotor unmanned
- signal
- angle
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 67
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 52
- 230000008569 process Effects 0.000 claims abstract description 19
- 238000012545 processing Methods 0.000 claims abstract description 18
- 238000012360 testing method Methods 0.000 claims abstract description 14
- 238000001914 filtration Methods 0.000 claims abstract description 13
- 239000011159 matrix material Substances 0.000 claims description 52
- 238000004088 simulation Methods 0.000 claims description 38
- 230000006870 function Effects 0.000 claims description 29
- 238000003491 array Methods 0.000 claims description 26
- 238000010586 diagram Methods 0.000 claims description 20
- 238000009826 distribution Methods 0.000 claims description 18
- 239000013598 vector Substances 0.000 claims description 18
- 238000004364 calculation method Methods 0.000 claims description 12
- 238000013461 design Methods 0.000 claims description 11
- 238000013178 mathematical model Methods 0.000 claims description 8
- 238000013519 translation Methods 0.000 claims description 7
- 238000001228 spectrum Methods 0.000 claims description 6
- 238000011161 development Methods 0.000 claims description 4
- 230000018109 developmental process Effects 0.000 claims description 4
- 238000001514 detection method Methods 0.000 claims description 3
- 230000008030 elimination Effects 0.000 claims description 2
- 238000003379 elimination reaction Methods 0.000 claims description 2
- 238000012163 sequencing technique Methods 0.000 claims description 2
- 208000001992 Autosomal Dominant Optic Atrophy Diseases 0.000 claims 1
- 206010011906 Death Diseases 0.000 claims 1
- 238000004458 analytical method Methods 0.000 abstract description 9
- 230000008054 signal transmission Effects 0.000 abstract description 5
- 238000005070 sampling Methods 0.000 description 17
- 238000011160 research Methods 0.000 description 10
- 238000005516 engineering process Methods 0.000 description 8
- 238000004891 communication Methods 0.000 description 7
- 230000000694 effects Effects 0.000 description 7
- 230000009466 transformation Effects 0.000 description 7
- 238000002474 experimental method Methods 0.000 description 6
- 230000005540 biological transmission Effects 0.000 description 5
- 238000000354 decomposition reaction Methods 0.000 description 4
- 238000005259 measurement Methods 0.000 description 4
- 230000003595 spectral effect Effects 0.000 description 3
- 150000001875 compounds Chemical class 0.000 description 2
- 238000004590 computer program Methods 0.000 description 2
- 238000004141 dimensional analysis Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 230000008092 positive effect Effects 0.000 description 2
- 230000002457 bidirectional effect Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 230000002860 competitive effect Effects 0.000 description 1
- 230000001934 delay Effects 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000005315 distribution function Methods 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 238000013401 experimental design Methods 0.000 description 1
- 239000002360 explosive Substances 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000010363 phase shift Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000012958 reprocessing Methods 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Landscapes
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention belongs to the field of unmanned aerial vehicle positioning and tracking, and discloses a method and a system for blind estimation of a normal mode signal for positioning and tracking of a multi-rotor unmanned aerial vehicle, wherein a 2D-MUSIC algorithm is used for blind estimation of a normal mode signal DOA of the multi-rotor unmanned aerial vehicle; by a curve fitting method, a relational expression between the minimum critical angle of normal mode signal incidence of the multi-rotor unmanned aerial vehicle and the number of unmanned aerial vehicle signal sources is deduced; the real-value beam forming space ESPRIT algorithm is combined with the unscented Kalman filtering algorithm and the ODAS open embedded test system, and the direction angle and the pitch angle of the multi-rotor unmanned aerial vehicle are detected in real time. The invention also provides an interface operating system for the three-dimensional space DOA blind estimation of the multi-rotor unmanned aerial vehicle, which greatly improves the position positioning, signal analysis and processing and intelligent management in the signal transmission process of the multi-rotor unmanned aerial vehicle, realizes automatic observation and operation, is safe and convenient, and can visually display the data of the real-time DOA three-dimensional space elevation angle and azimuth angle of the multi-rotor unmanned aerial vehicle.
Description
Technical Field
The invention belongs to the field of unmanned aerial vehicle positioning and tracking, and particularly relates to a normal mode signal blind estimation method and system for multi-rotor unmanned aerial vehicle positioning and tracking.
Background
At present, wireless communication technology exhibits explosive rate increase with the increasing degree of national informatization and the rapid increase of information amount. Nowadays, wireless communication technology for multi-rotor unmanned aerial vehicles has been developed and advanced unprecedentedly. A signal master base station capable of collecting signals is built, and various parameters, contents and information of the signals transmitted are received, analyzed and processed by laying array signal array elements, so that the problem of real-time positioning and tracking of the multi-rotor unmanned aerial vehicle is solved. In modern society, wireless communication technology for positioning and tracking of multi-rotor unmanned aerial vehicles has become increasingly widely recognized by the public.
With the rapid development of communication technology nowadays, a plurality of problems are gradually presented. For example, in multi-rotor unmanned aerial vehicle wireless communication, when a signal of a transmitting end is transmitted, information data such as a modulation mechanism and the like of the transmitting end are not accurately monitored at a receiving end of an operator; signals sent by a transmitting terminal are interfered and shielded by other various electronic signals in the transmission process, so that the receiving terminal of an operator receives the image transmission signals and the phenomena of signal overlapping or signal blurring occur. Therefore, how to accurately receive accurate transmission signals at the receiving end of a multi-rotor unmanned aerial vehicle manipulator, and simultaneously identify and reprocess the transmission signals has gradually become the direction of important research of people in the communication industry, and people are full of research enthusiasm for the field. In recent years, a great deal of research is carried out on the blind estimation of the normal mode signals of the positioning and tracking of the multi-rotor unmanned aerial vehicle by many scholars at home and abroad, and a great deal of research results are published at the same time.
The normal mode algorithm utilizes the normal mode characteristic of the signal, so that a reference signal is not needed, the bandwidth is liberated, the signal transmission rate is greatly improved, and the method has the advantages of constant module value, low complexity, unobvious error of a matrix array model signal and the like. Due to the advantages of the normal mode signals, in the fields of communication signal processing and digital image processing of multi-rotor unmanned aerial vehicle positioning and tracking, the normal mode algorithm is more and more emphasized by researchers and becomes a hot problem of research increasingly, and the blind estimation technology of the normal mode signals gradually becomes a direction and a trend of competitive research.
The traditional blind estimation method has the problems of low resolution, low estimation precision and low stability in the detection of the normal mode signal DOA.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a normal mode signal blind estimation method and system for multi-rotor unmanned aerial vehicle positioning and tracking.
The invention is realized in such a way that a method for blind estimation of a normal mode signal for positioning and tracking of a multi-rotor unmanned aerial vehicle comprises the following steps:
the method comprises the steps that firstly, a 2D-MUSIC algorithm is used for carrying out blind estimation on DOA of a normal mode signal of the multi-rotor unmanned aerial vehicle, angle information of the DOA is estimated blindly by using a two-dimensional angle, and an operator is used as a reference point to estimate a real-time azimuth angle and a pitch angle of the unmanned aerial vehicle;
step two, deriving a relation between a normal mode signal incidence minimum critical angle of the multi-rotor unmanned aerial vehicle and the number of the unmanned aerial vehicle signal sources through a curve fitting method according to the changed number of the different unmanned aerial vehicle signal sources and the observed real-time positions of the different multi-rotor unmanned aerial vehicles;
and step three, constructing a multi-rotor unmanned aerial vehicle positioning and tracking blind estimation device platform based on a real-value beam forming space ESPRIT algorithm and an unscented Kalman filtering algorithm, designing a dual-system circular array model, and detecting the direction angle and the pitch angle of the multi-rotor unmanned aerial vehicle in real time by combining an ODAS open embedded test system.
Further, the specific process of the first step is as follows:
establishing a mathematical model diagram, and assuming that the set antenna array is a two-way parallel antenna array distribution, wherein the distance between two adjacent array elements along the X-axis direction is d x The distance between two parallel arrays is d y The first uniform antenna array is composed of (N + 1) signal array elements arranged along a straight line, and the second uniformly distributed linear array is composed of N antenna arrays;
suppose there is a target airspaceP multiple rotor unmanned aerial vehicle position signal is from two-dimensional direction in spaceIncident on receiving sensors laid down by the base stations of the antenna array, wherein,respectively representing an azimuth angle and a pitch angle, wherein the azimuth angle represents an included angle between a projection line of an incoming wave direction of an incident position signal of the multi-rotor unmanned aerial vehicle on an xoy plane and an x axis; the pitch angle represents the included angle between the projection line of the incoming wave direction of the incident position signal of the multi-rotor unmanned aerial vehicle on the xoy plane and the incoming wave direction of the incident signal.
Further, the specific process of the second step is as follows:
(1.1) simulating azimuth angles and pitch angles of DOAs in different directions of arrival, which are incident from normal mode signal sources of multi-rotor unmanned aerial vehicles with different numbers, wherein the signal-to-noise ratio is set to be 10dB, and determining MATLAB simulation comparison results in different pitch angles by using a control variable method under the condition of the same azimuth angle;
(1.2) importing simulation comparison result data into LabVIEW software, sequencing, arranging and integrating the simulation acquisition data, and performing curve fitting processing on the measured data quantity by using a least square method and an SVD (singular value decomposition) algorithm;
(1.3) deriving a relational expression between the number of multi-rotor unmanned aerial vehicles and the normal mode signal incidence minimum critical angle obtained by blind estimation according to a polynomial coefficient derived from LabVIEW software, wherein the expression is as follows:
Y=1.51-1.38X+0.85X 2
in the formula, X represents the number of the multi-rotor unmanned aerial vehicles, and Y represents the normal mode signal incidence minimum critical angle obtained by blind estimation.
Further, the third step comprises the following steps:
(2.1) constructing a circular array model;
(2.2) carrying out DOA blind estimation based on the real-value beam forming space ESPRIT algorithm;
and (2.3) carrying out interference elimination based on an unscented Kalman filtering algorithm.
Further, the specific process of (2.1) is as follows:
assuming that the circular array model is composed of M array elements, uniformly distributed on a circular array with a radius of R and mutually independent, assuming that N target multi-rotor unmanned aerial vehicle signals are emitted into the circular array, taking an origin of coordinate O as a central reference point, an included angle alpha formed by an unmanned aerial vehicle position signal and an XOY axis plane is a pitch angle of a target unmanned aerial vehicle to be measured, and an included angle theta formed by a projection dotted line of the multi-rotor unmanned aerial vehicle position signal on a horizontal plane and an X axis is defined as an azimuth angle, wherein the pitch angle alpha belongs to [0 DEG, 90 DEG ], the azimuth angle theta belongs to [0 DEG, 360 DEG ], and counterclockwise is taken as a forward direction, and the circular array model is expressed as:
writing the above equation in matrix form:
X(t)=AS(t)+N(t)
wherein, X (t) is output data vector of M × 1 dimension; s (t) is a far-field sound source signal of Nx 1 dimension; n (t) is noise data of M multiplied by 1 dimension and is Gaussian white noise, and the noise on each array element is irrelevant; a = [ a ] 1 (w 0 ),a 2 (w 0 ),…,a N (w 0 )]Is an M × N dimensional circular array matrix, a i (w 0 ) I =1,2, \8230, N denotes a steering vector, the expression is as follows:
in the formula, w 0 Is the angular frequency of the received signal, and w 0 =2πf 0 =2 π c/λ, c representing the speed of sound; tau is Mi Indicating the time delay of the ith signal received relative to the mth array element of the reference circular array element.
Further, the specific process of the step (2.2) is as follows:
the real-value beam forming space ESPRIT algorithm utilizes the rotation invariance of a signal subspace of a circular array model receiving data covariance matrix, and then calculates the azimuth angle and the pitch angle of a multi-rotor unmanned aerial vehicle position signal transmitted by a space signal source;
the circular array element array is an integral array model consisting of a plurality of sub-arrays, the sub-arrays cooperate with each other and are surrounded by a plurality of linear arrays, the number of the sub-array elements is M, the sub-array elements are arranged in a crossed manner, wherein the X (t) sub-array consists of the first 0 to M-1 arrays, and the Y (t) sub-array represents a combined array consisting of 1 to M arrays;
the ESPRIT algorithm of the real-value beam forming space receives a multi-rotor unmanned aerial vehicle position signal source of a far-field space, gaussian white noise with the same frequency is introduced, and due to the fact that two sub-arrays of a signal sub-space and a noise sub-space have relative translation phase values, the expression of phi is as follows:
in the formula, phi is a diagonal matrix formed by phase delays of two sub-arrays; β is the center frequency; d is the signal source wavelength; theta υ Is the arrival angle of the upsilon signal sources;
x (k) is a standard M × M matrix, then two sub-matrices x forming x (k) 1 (k) And x 2 (k) The same two array signals are received, denoted as:
in the formula, x 1 (k) And x 2 (k) Are identical in all respects, except that they are offset from each other by a known displacement vector; x (k) represents the output vector of the entire array; a is an m x m Van der Waals matrix, whereinThe structure of (a) is used to obtain an estimate of the diagonal element of phi without knowledge of the information of a; s (k) is the magnitude vector of the complex sinusoid;n 1 (k)、n 2 (k) Is complex white Gaussian noise with stable zero mean value, and aims to estimate the frequency and amplitude of each complex sine by observing data;
the corresponding matrix of the array of circular array elements is represented as:
in the formula, R xx Is a covariance matrix, R zz Is an autocorrelation matrix of x, is a diagonal matrix of d x d, each element corresponding to the power of a complex sinusoid, A H Is the transposed conjugate matrix of a and,is the average value of the minimum eigenvalue of the noise, I is an m × m matrix, the sub diagonal element of which is 1, and the other elements are zero;
the two corresponding sub-arrays are respectively represented as:
the overall signal subspace is assumed to be E x Comprising two signal subspaces E 1 And E 2 The column vector of the space of each sub array is the eigenvector corresponding to the maximum eigenvalue, a translation relation of a non-singular matrix exists between the two sub spaces, the translation relation is called as a rotation operator, and the expression is as follows:
E 1 Ψ=E 2
in the formula, psi is expressed as psi = T -1 Phi T, T is a non-singular matrix, psi and phi are similar matrices, having common eigenvalues, and since phi is a diagonal matrix, it can be said that: phi is the eigenvalue diagonal matrix of psi; e 1 Representing the signal subspace acquired by the x array;E 2 representing a signal subspace acquired by the y array;
the simplified expression of the rotation operator that exists is:
E 1 =AT
E 2 =AφT
therefore, the relationship of Ψ and φ is derived:
φ=TΨT -1
therefore, the blind estimation angle of the normal mode signal becomes a psi characteristic value solving process, phi is obtained by decomposing psi characteristic values, and the azimuth angle and the pitch angle of the multi-rotor unmanned aerial vehicle position signal in the space are estimated by using phi.
Further, the specific process of (2.3) is as follows:
(2.3.1) performing an Unknown Transform (UT) by selecting X to N (mu, sigma) 2 ) Y = g (X), in the case of N dimensions, X to N (μ, σ) 2 ) Is a Gaussian probability distribution function, g (X) represents a function obtained through nonlinear transformation, Y is an estimation function, X belongs to Gaussian distribution, mu is a mean value, and sigma is a standard deviation and is used for estimating a value of Y; selection 2n +1 Sigma points x [i] And its weight ω [i] The following conditions are satisfied:
∑ i ω [i] =1
μ=∑ i ω [i] χ [i]
∑=∑ i ω [i] (χ [i] -μ)(χ [i] -μ) T
wherein i =1,n; sigma i Cholesky decomposition representing a covariance matrix; omega [i] A weight value representing X;
(2.3.2) selecting a plurality of sampling points in the original state distribution according to a certain rule, and enabling the mean value and the covariance of the sampling points to be equal to the mean value and the covariance of the original state distribution: and substituting the points into a nonlinear function for summarizing to obtain a corresponding nonlinear function value point set, and solving the transformed mean value and covariance through the point sets. The nonlinear transformed mean and covariance accuracies thus obtained have a minimum of 2 nd order accuracy. For a gaussian distribution, a 3 rd order accuracy can be achieved. The choice of sampling points is based on the correlation columns of the square root of the prior mean and the prior covariance matrix, and the Sigma points are chosen according to these rules:
χ [0] =μ
in the formula (I), the compound is shown in the specification,represents the square root of the matrix; λ denotes a scaling parameter, λ = α 2 (n + k) -n; represents how far from the mean, n represents a dimension; sigma represents a covariance matrix; i represents the ith column vector, and i-n represents the ith-n column of the matrix;
(2.3.3) solving the square root of the correlation matrix to obtain the Sigma point, defining the covariance matrix as S, and simultaneously satisfying Sigma = S T S, according to the principle of diagonalization, the following transformation is carried out:
Σ=VDV -1
(2.3.4) assigning a Sigma point weight:
in the formula (I), the compound is shown in the specification,for the purpose of calculating the mean value,for calculating the covariance; alpha is used to adjust the influence of higher-order terms on the model, and generally takes the value of alpha =0.01. For a gaussian distribution, the effect is optimal with β = 2. The subscript m is expressed as a mean value, c is expressed as covariance, the superscript i is expressed as the ith sampling point, the selection of alpha controls the distribution state of the sampling point, and the parameter beta to be selected is a non-negative weight coefficient which can combine errors of higher-order terms in the equation, so that the influence of the higher-order terms can be included;
(2.3.5) performing Gaussian estimation on the calculated mapping value of the Sigma point to obtain the following estimation result:
where Σ' denotes the calculation of the covariance matrix of the measured variables and the observed variables,weight value of X χ [i] Representing the number of the selected sampling points, and mu' represents the mean value of data distribution;
(2.3.6) substituting the Unscented Transform (UT) into Kalman filtering to obtain Unscented Kalman filtering, and obtaining a mean value and a variance approximately through sampling and weight calculation of Unscented transformation;
the unscented transformation selects some sampling points in the original state distribution according to a certain rule, so that the mean value and covariance of the sampling points are equal to the mean value and covariance of the original state distribution: and substituting the points into a nonlinear function for summarizing to obtain a corresponding nonlinear function value point set, and solving the mean value and covariance after transformation through the point sets.
Further, the unscented kalman filter includes the steps of:
(3.1) use of χ [0] 、χ [i] 、Andobtaining a group of sampling points Sigma points and corresponding weights;
(3.2) calculate the prediction of 2n +1 Sigma point set, i =1,2,2n +1, then the state at the next time t:
(3.3) calculating the prior prediction mean and covariance by using the Sigma points which are obtained by the unscented transformation and are subjected to state transition, and calculating the one-step prediction of the system state quantity, wherein the prior estimation value is as follows:
(3.4) calculating the prior predicted value of the error covariance matrix:
(3.5) generating a new Sigma point set using the unscented transform sampling again according to the one-step predicted value:
(3.6) substituting the new Sigma point set obtained in (3.5) into the measurement equation h (x) to obtain the predicted measurement:
another object of the present invention is to provide a blind estimation system for multi-rotor drone positioning and tracking of a blind estimation method for multi-rotor drone positioning and tracking of a blind estimation system for multi-rotor drone positioning and tracking of a normal mode signal, comprising:
the 2D-MUSIC module is used for carrying out blind estimation on DOA of the normal mode signal of the multi-rotor unmanned aerial vehicle and carrying out blind estimation on angle information of the DOA by utilizing a two-dimensional angle;
the curve fitting module is used for deducing a relational expression between the minimum critical angle of normal mode signal incidence of the multi-rotor unmanned aerial vehicle and the number of unmanned aerial vehicle signal sources;
and the detection module is used for constructing a multi-rotor unmanned aerial vehicle positioning and tracking blind estimation device platform based on a real-value beam forming space ESPRIT algorithm and an unscented Kalman filtering algorithm, designing a dual-system circular array model, and detecting the direction angle and the pitch angle of the multi-rotor unmanned aerial vehicle in real time by combining an ODAS open embedded test system.
Another object of the present invention is to provide a blind estimation software for multi-rotor drone positioning and tracking that implements the blind estimation system for multi-rotor drone positioning and tracking normal mode signal, the blind estimation software for multi-rotor drone positioning and tracking normal mode signal comprising:
the LabVIEW virtual instrument is used for establishing a mathematical model and interface design of normal mode signal blind estimation, and on the basis of hardware with complete functions in three-dimensional space DOA blind estimation system software design, the virtual instrument technology is combined with software to complete various developments and simulation
Further, the normal mode signal blind estimation software for multi-rotor unmanned aerial vehicle positioning and tracking comprises:
a system login interface, wherein after a correct user name and a login password are input in the login interface, a login button is clicked, and a main function page of the DOA blind estimation system in a three-dimensional space is entered;
and the system function operation interface comprises a front panel diagram and a program block diagram, and is used for realizing the system function of the normal mode signal DOA blind estimation for the positioning and tracking of the multi-rotor unmanned aerial vehicle by using LabVIEW simulation.
Further, the specific process of the system login interface comprises the following steps:
inputting a user name and a password which are set; if the input is wrong, the system prompts that the user name and the password are input wrongly and cannot enter the main system;
the system login interface adopts a tiled sequential structure, and programming is executed through a three-frame sequence; writing a While loop in a first frame to play a role of circularly inputting a user name and a password, and when the input user name and the input password are correct, outputting a true signal instruction to a next frame at the first time by clicking login, otherwise, outputting the true signal instruction; compared with the second frame, the third frame has the function of judging output, and if the output is true, the sub VI corresponding to the system function is opened;
the input part of DOA blind estimation in the front panel image is the scanning accuracy of a covariance matrix, an azimuth angle and a pitch angle of a signal, and the output content is a spatial spectrum related parameter, a specific value parameter of the azimuth angle and the pitch angle of a received signal;
the program block diagram is a place for programming a virtual instrument, various modularized devices are adopted for connecting line programming, the whole program block is divided into two modules, and one module has the function of decomposing the characteristic value of a covariance matrix; the other part is vector calculation determined by an azimuth angle and a pitch angle, and the signal flow is processed by the two parts of signals, and finally the calculation of spatial spectrum blind estimation is carried out.
The device herein refers to: many controls, similar in appearance to traditional instruments (e.g., oscilloscopes, multimeters), can be used to conveniently create a user interface. The user interface is referred to as the front panel in LabVIEW. Using icons and wiring, objects on the front panel can be programmatically controlled.
It is a further object of the present invention to provide a computer arrangement comprising a memory and a processor, said memory storing a computer program which, when executed by said processor, causes said processor to perform the steps of said method for constant modulus signal blind estimation for multi-rotor drone position tracking.
Another object of the present invention is to provide a computer-readable storage medium, storing a computer program which, when executed by a processor, causes the processor to perform the steps of the method for normal mode signal blind estimation for multi-rotor drone position tracking.
By combining the technical scheme and the technical problem to be solved, the technical scheme to be protected by the invention has the advantages and positive effects that:
compared with the traditional Extended Kalman Filter (EKF), the average accuracy error is reduced by 51.91 percent, the purpose of high-precision positioning and tracking is achieved, the effectiveness of the method is verified, and the diversity of blind estimation of the normal mode signal for positioning and tracking of the multi-rotor unmanned aerial vehicle is better enriched; the last interface operating system who designs the blind estimation of three-dimensional space DOA to many rotor unmanned aerial vehicle, greatly promoted the position location to many rotor unmanned aerial vehicle signal transmission in-process, signal analysis handles and intelligent management, realized observing and operating with automizing, it is safe and convenient, on the digital screen, the data of the real-time DOA three-dimensional space angle of elevation and the azimuth of the many rotor unmanned aerial vehicle of audio-visual demonstration, better human-computer interaction function has.
The DOA of the normal mode signal of the multi-rotor unmanned aerial vehicle is estimated blindly by using a 2D-MUSIC (two-dimensional space MUSIC) algorithm, and the angle information of the DOA is estimated blindly by using a two-dimensional angle, namely, the real-time azimuth angle and the pitch angle of the multi-rotor unmanned aerial vehicle are estimated by using an operator as a reference point;
the method changes the number of signal sources of different multi-rotor unmanned aerial vehicles, observes the obtained real-time positions of the different multi-rotor unmanned aerial vehicles, and firstly deduces a relational expression between the minimum critical angle of the normal mode signal incidence of the multi-rotor unmanned aerial vehicle and the number of the signal sources of the unmanned aerial vehicles through a curve fitting method;
the invention builds a set of positioning and tracking blind estimation device platform based on real-value beam forming space ESPRIT algorithm combined with unscented Kalman filtering algorithm to obtain the multi-rotor unmanned aerial vehicle, designs a dual-system circular array model, combines ODAS open embedded test system, can detect the direction angle and the pitch angle of the multi-rotor unmanned aerial vehicle in real time, and compared with the traditional Extended Kalman Filter (EKF), the average accuracy error is reduced by 51.91 percent, thereby achieving the purpose of high-precision positioning and tracking, verifying the effectiveness of the device, and better enriching the diversity of blind estimation of the normal mode signals for positioning and tracking the multi-rotor unmanned aerial vehicle;
the interface operating system for the three-dimensional space DOA blind estimation of the multi-rotor unmanned aerial vehicle greatly improves the position positioning, signal analysis and processing and intelligent management in the signal transmission process of the multi-rotor unmanned aerial vehicle, realizes automatic observation and operation, is safe and convenient, and visually displays the data of the elevation angle and the azimuth angle of the three-dimensional space of the multi-rotor unmanned aerial vehicle in real time on a digital screen.
Drawings
Fig. 1 is a flowchart of a normal mode signal blind estimation method for positioning and tracking of a multi-rotor unmanned aerial vehicle according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a bi-directional parallel array simulation provided by an embodiment of the present invention;
FIG. 3 is a schematic view of an azimuth and a pitch provided by an embodiment of the present invention;
FIG. 4 is a graph of a curve fit provided by an embodiment of the present invention;
FIG. 5 is a mathematical model of a circular array provided by an embodiment of the present invention;
FIG. 6 is a circular array model provided by an embodiment of the present invention;
FIG. 7 is a model of a linear array element array provided by an embodiment of the present invention;
FIG. 8 is a result analysis graph, (a), a two-dimensional result graph, and (b) a three-dimensional result graph, provided by an embodiment of the present invention;
FIG. 9 is a comparative graph of tests provided by an embodiment of the present invention;
FIG. 10 is a comparison of UFK and EKF tests provided by embodiments of the present invention;
fig. 11 is a diagram illustrating the interference cancellation effect provided by the embodiment of the present invention;
FIG. 12 is a block diagram of a system software design provided by an embodiment of the invention;
FIG. 13 is a diagram of a login interface provided by an embodiment of the invention;
FIG. 14 is a user name and password entry correct prompt provided by an embodiment of the present invention;
FIG. 15 is a block diagram of a login interface routine provided by an embodiment of the invention;
FIG. 16 is a diagram illustrating simulation results provided by an embodiment of the present invention;
FIG. 17 is a two-dimensional DOA angle of arrival intensity dispersion plot provided by an embodiment of the present invention;
FIG. 18 is a simulation comparison diagram of two signal sources provided by an embodiment of the present invention;
FIG. 19 is a simulation comparison diagram of three signal sources provided by an embodiment of the present invention;
FIG. 20 is a simulated comparison of four signal sources provided by an embodiment of the present invention;
FIG. 21 is a graph of a curve fit interface provided by an embodiment of the present invention;
FIG. 22 is a schematic diagram of a dual-system circular array model provided by an embodiment of the present invention;
fig. 23 is a schematic view of DOA information of a multi-rotor drone provided by an embodiment of the present invention, where (a) is an azimuth angle, and (b) is a pitch angle.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not limit the invention.
This section is an illustrative example developed to explain the claims in order to enable those skilled in the art to fully understand how to implement the present invention.
As shown in fig. 1, a method for blind estimation of a normal mode signal for positioning and tracking of a multi-rotor drone provided by an embodiment of the present invention includes:
s101, blind estimation is carried out on the DOA of the normal mode signal of the multi-rotor unmanned aerial vehicle by using a 2D-MUSIC algorithm, angle information of the DOA is estimated by using a two-dimensional angle in a blind mode, and the real-time azimuth angle and the pitch angle of the unmanned aerial vehicle are estimated by using an operator as a reference point;
s102, observing the obtained real-time positions of different multi-rotor unmanned aerial vehicles according to the changed number of signal sources of different unmanned aerial vehicles, and deducing a relation between a normal mode signal incidence minimum critical angle of the multi-rotor unmanned aerial vehicles and the number of the signal sources of the unmanned aerial vehicles for the first time through a curve fitting method;
s103, a multi-rotor unmanned aerial vehicle positioning and tracking blind estimation device platform based on a real-value beam forming space ESPRIT algorithm and an unscented Kalman filtering algorithm is built, a dual-system circular array model is designed, and an ODAS open embedded test system is combined to detect the direction angle and the pitch angle of the multi-rotor unmanned aerial vehicle in real time.
The specific process of multi-rotor unmanned aerial vehicle DOA blind estimation based on the 2D-MUSIC algorithm comprises the following steps:
in an actual scene, more information transmission established in a three-dimensional space is performed on the aspect of positioning and tracking of the multi-rotor unmanned aerial vehicle, so that in order to improve the comprehensiveness and the integrity of DOA blind estimation, the DOA blind estimation of a 2D-MUSIC (two-dimensional space MUSIC) algorithm of a space array is adopted. Compared with the traditional blind estimation of DOA in one-dimensional space, the method is limited to determining one angle of the incident angle of the position signal source of the multi-rotor unmanned aerial vehicle, namely, only the target signal can be determined to be in one plane, and the accurate directional positioning cannot be realized. Therefore, utilize two-dimensional space angle to estimate many rotor unmanned aerial vehicle current position DOA's angular information, included the estimation of azimuth and pitch angle promptly, improved the degree of discernment more effectively, richened better to the blind variety of estimating of many rotor unmanned aerial vehicle position tracking's normal mode signal.
The invention designs a bidirectional parallel antenna array, provides an algorithm for two-dimensional space DOA blind estimation for pairing parameters, improves diversity of direction of arrival blind estimation (namely, blind estimation of a space azimuth angle and a pitch angle is increased), and reduces complexity of calculation on the original basis.
A mathematical model of the system is created, and as shown in fig. 2, the antenna array is assumed to be a bi-directional parallel antenna array distribution. Wherein the distance between two adjacent array elements along the X-axis direction is d y First uniform antennaThe array is composed of (N + 1) signal array elements arranged along a straight line, and the second uniformly distributed linear array is composed of N antenna arrays.
Now assume that there are P multiple rotorcraft position signals in the target airspace from the two-dimensional direction of space, as shown in FIG. 3Incident on receiving sensors laid by the base stations of the antenna array. WhereinAnd the azimuth angle and the pitch angle are respectively represented (namely, the azimuth angle represents an included angle between a projection line of the incoming wave direction of the incident position signal of the multi-rotor unmanned aerial vehicle on the xoy plane and the x axis, and the pitch angle represents an included angle between the projection line of the incoming wave direction of the incident position signal of the multi-rotor unmanned aerial vehicle on the xoy plane and the incoming wave direction of the incident signal).
The specific process of deducing the relation between the minimum critical angle of normal mode signal incidence of the multi-rotor unmanned aerial vehicle and the number of signal sources comprises the following steps:
the invention respectively simulates azimuth angles and pitch angles of DOAs in different directions of arrival, which are incident from the normal mode signal sources of the multi-rotor unmanned aerial vehicle with different numbers, and the signal-to-noise ratio is set to be 10dB. Determining MATLAB simulation comparison results in different pitch angles under the condition of the same azimuth angle by using a control variable method, and deducing a relational expression of the number of the multi-rotor unmanned aerial vehicles and the normal mode signal incidence minimum critical angle obtained by blind estimation for the first time.
Experiment one: when there are two many rotor unmanned aerial vehicle signal sources, obtain four simulation results, carry out contrastive analysis and verify the back, it is with easy to obtain: when there are two multi-rotor drone signal sources, the minimum critical pitch angle is 2 ° at the same azimuth.
Experiment two: when there are three many rotor unmanned aerial vehicle signal sources, obtain three simulation results, carry out contrastive analysis and verify the back, it is with easy to obtain: when there are three multi-rotor drone signal sources, the minimum critical pitch angle is 5 ° at the same azimuth.
Experiment three: when there are four many rotor unmanned aerial vehicle signal sources, obtain three simulation result, carry out contrastive analysis and verify the back, it is with difficult to obtain: when there are four multi-rotor drone signal sources, the minimum critical pitch angle is 10 ° for the same azimuth.
The invention is subjected to multiple times of experimental simulation, the obtained data is led into LabVIEW software, 12 groups of data acquired by simulation are sequenced, sorted and integrated by carrying out curve fitting processing on the result data, and the measured data quantity is subjected to curve fitting processing by using a least square method and an SVD algorithm, as shown in figure 4. According to a polynomial coefficient derived from LabVIEW, a relational expression of the number of the multi-rotor unmanned aerial vehicles and the normal mode signal minimum incidence critical angle obtained through blind estimation is derived for the first time, as shown in formula 1, X represents the number of the multi-rotor unmanned aerial vehicles, and Y represents the normal mode signal minimum incidence critical angle obtained through blind estimation.
Y=1.51-1.38X+0.85X 2 (1)
The specific process of the multi-rotor unmanned aerial vehicle positioning and tracking DOA blind estimation method comprises the following steps:
according to the invention, a circular array model is designed for the first time to realize the positioning of the real-time position of the multi-rotor unmanned aerial vehicle, and the azimuth angle and the pitch angle of the multi-rotor unmanned aerial vehicle normal mode signal in the space are calculated by utilizing the rotation invariance of the signal subspace of the array signal receiving data covariance matrix based on the ESPRIT algorithm of the real-value beam forming space. And then, an Unscented Kalman Filter (UKF) algorithm with higher precision and smaller calculated amount is adopted, a dual-system circular array model is designed, an ODAS open embedded test platform is combined, a three-dimensional dynamic model is constructed, the position information of the multi-rotor unmanned aerial vehicle is visually displayed in a space coordinate system, and position positioning and target track tracking are realized.
The invention designs a circular array model to realize the real-time positioning of a multi-rotor unmanned aerial vehicle, the model is supposed to be composed of M array elements, the M array elements are uniformly distributed on a circular array with the radius of R and are mutually independent, the model is shown in figure 5, N target multi-rotor unmanned aerial vehicle signals are supposed to be emitted onto the circular array, the coordinate origin O is taken as a central reference point, an included angle alpha formed by the unmanned aerial vehicle position signals and an XOY axis plane is taken as a pitch angle of a target unmanned aerial vehicle to be detected, an included angle theta formed by a projection dotted line of the multi-rotor unmanned aerial vehicle position signals on a horizontal plane and an X axis is defined as an azimuth angle, wherein the pitch angle alpha belongs to [0 degrees ], 90 degrees ], the azimuth angle theta belongs to [0 degrees ], 360 degrees ], and the anticlockwise direction is taken as the forward direction, so that a formula 2 of the circular array model can be expressed as follows:
writing equation (2) in matrix form as:
X(t)=AS(t)+N(t) (3)
x (t) is an output data vector of dimension M multiplied by 1; s (t) is a far-field sound source signal with dimension Nx 1; n (t) is noise data of M multiplied by 1 dimension and is Gaussian white noise, and the noise on each array element is irrelevant; a = [ a ] 1 (w 0 ),a 2 (w 0 ),…,a N (w 0 )]Is an M × N dimensional circular array matrix, and a i (w 0 ) I =1,2, \ 8230, N is a steering vector, whose expression is as follows
w 0 Is the angular frequency of the received signal, and w 0 =2πf 0 =2 π c/λ, c representing the speed of sound; tau. Mi Indicating the time delay of the ith signal received relative to the mth array element of the reference circular array element. Fig. 6 shows a circular array model composed of four receiving array elements.
The real-value beam forming space ESPRIT algorithm utilizes the rotation invariance of a signal subspace of a circular array signal receiving data covariance matrix to further calculate the azimuth angle and the pitch angle of a multi-rotor unmanned aerial vehicle position signal transmitted by a space signal source. Compared with the traditional MUSIC algorithm, the method has the advantages of small calculation amount, no need of continuously searching spectral peaks in space and the like, and is used for expanding functions and simplifying calculation of the traditional MUSIC algorithm.
The real-valued beamforming space ESPRIT algorithm is based on the fact that: in a rotated vector, the signal on one element is derived from the phase shift of the earlier element signal. The core idea is that a circular array is regarded as an integral model, an integral array model is composed of a plurality of sub-arrays, the sub-arrays cooperate with each other, as shown in fig. 7, the circular array element array is formed by surrounding a plurality of linear arrays, the number of the array elements is M, the array elements are arranged in a crossed manner, wherein X (t) sub-arrays are composed of front M arrays, and Y (t) represents a combined array composed of rear M arrays.
The precondition of the conventional MUSIC algorithm is consistent, the real-valued beamforming space ESPRIT algorithm is also a multi-rotor unmanned aerial vehicle position signal source for receiving far-field space, gaussian white noise with the same frequency is introduced, and because two sub-arrays of a signal sub-space and a noise sub-space have a phase value of relative translation, the expression of phi is as follows:
assuming that the matrix is an M × M standard square matrix, the two matrix signals received by the two matrices can be expressed as:
the resulting correspondence matrix for the array is represented as:
the two corresponding sub-arrays may be represented as:
let the overall signal subspace assume to be E x Comprising two signal subspaces E 1 And E 2 The column vectors of the space of each decomposed sub-array are all eigenvectors corresponding to the maximum eigenvalue, and the translation relation of a non-singular matrix exists between the two sub-spaces is called as: and the rotation operator is expressed by a relational expression as follows:
E 1 Ψ=E 2 (9)
the simplified expression of the rotation operator that exists is:
E 1 =AT (10)
E 2 =AφT (11)
therefore, it is not difficult to derive the relationship between:
φ=TΨT -1 (12)
therefore, the blind estimation angle of the normal mode signal becomes a psi characteristic solving process, and the azimuth angle and the pitch angle of the position signal of the multi-rotor unmanned aerial vehicle in the space are estimated through the sampling signal.
According to the invention, firstly, a multi-rotor unmanned aerial vehicle normal mode signal source is acquired and analyzed, a circular array is arranged at equal intervals, the universality and the reliability of an experiment are improved by setting the number of sampling points, the positions where the multi-rotor unmanned aerial vehicle is planned to be located are x =13m, y =12m and z =11m, the incident wavelength transmitted by the normal mode signal is lambda, the distance d between every two adjacent array elements is set to be lambda/2, and the experimental result is shown in figure 8. In two-dimensional analysis chart and three-dimensional analysis chart, can accurate demonstration target many rotor unmanned aerial vehicle's specific position.
In the case of wanting pain and fast beat number, signal-to-noise ratio and array element number, a comparison experiment is performed with the MUSIC algorithm and various improved ESPRIT algorithms, as shown in fig. 9, the result shows that the array measurement accuracy of the actual value beam forming space ESPRIT algorithm is relatively highest.
The invention designs a dual-system circular array model based on Unscented Kalman Filter (UKF) with higher precision and smaller calculated amount, combines ODAS to open an embedded test platform to construct a three-dimensional dynamic model, visually displays the position information of the multi-rotor unmanned aerial vehicle in a space coordinate system to perform position positioning and target trajectory tracking, and reduces the average precision error by 51.91% by comparing with the traditional Extended Kalman Filter (EKF).
Unscented Kalman Filter (UKF) is based on the statistical conclusion of an Unscented Transform (UT) on a random vector transformed by a nonlinear function, using finite Sigma points to describe the statistical properties of the vector to be transformed, to select a particular set of Sigma samples to project completely the first-order second-order matrix of the original distribution, and then non-linearly transform these points from the UT and distribute them into a new space where it transforms the statistical information of the distribution, to compute the statistical information used in the output.
First an Unshunted Transform (UT) is performed, now taking X-N (μ, σ) as the starting point 2 ) Y = g (X), here n-dimensional, X belonging to a gaussian distribution, the objective being to estimate the value of Y, in the n-dimensional case 2n +1 Sigma points are chosen: chi-type food processing machine [i] And its weight: omega [i] The following conditions need to be satisfied:
∑ i ω [i] =1 (13)
μ=∑ i ω [i] χ [i]
∑=∑ i ω [i] (χ [i] -μ)(χ [i] -μ) T
there are many different ways of assigning the samples and weights, and the Sigma point is selected according to these rules:
to solve the Sigma point mentioned above, the square root of the correlation matrix needs to be solved next, where S is defined as the matrix, and Σ = S is satisfied T S, according to diagonalizationThe principle is transformed as follows:
Σ=VDV -1 (15)
for the assignment of the Sigma point weights,for the purpose of calculating the mean value,for calculating covariance
Next, further gaussian estimation is performed on the calculated mapping value of the Sigma point, and the following estimation result is obtained:
finally, the Unscented kalman filter can be obtained by substituting the Unscented Transform (UT) into the kalman filter:
13:returnμ t ,Σ t 。
the Unscented Kalman Filter (Unscented Kalman Filter, UKF) directly finds a gaussian distribution that is similar to the true distribution without using linear characterization, whereas the traditional Extended Kalman Filter (EKF) finds a linear model by solving the first-order full derivative to approximate a nonlinear model. The traditional EKF utilizes Taylor decomposition to linearize the model, and then utilizes Gaussian hypothesis to solve the problem of difficult probability calculation, but the introduction of linear error reduces the model precision. For a nonlinear model, it is difficult to solve a Bayesian recursion formula directly by using an analytic method, and the mean and variance of each probability distribution are not easy to obtain, but the problem can be solved well by using an Unknown Transform (UT) method, and the mean and variance can be obtained approximately by sampling and weighting according to a certain rule. Moreover, because the approximation precision of the Unscented Transform (UT) method to the matrix is higher, the simulation comparison of the UKF and the EKF shows that the average precision error is reduced by 51.91 percent as shown in FIG. 10, and the effect of the UFK can reach the effect of a second-order EKF.
According to the invention, a dual-system circular array model is designed, an ODAS open embedded test platform is combined to construct a three-dimensional dynamic mathematical model, and the position information of the multi-rotor unmanned aerial vehicle is visually displayed in a space coordinate system to perform position positioning and target track tracking. As shown in fig. 11, in the three-dimensional space, the moving green big sphere represents the real-time position and flight path of the multi-rotor unmanned aerial vehicle, and the rest blue small spheres represent the noise signal source interference caused by the rotation of the multiple rotors of the unmanned aerial vehicle. Through experimental verification, the UKF adopted by the invention is obviously superior to the traditional EKF in the aspect of noise filtering effect.
The embodiment of the invention also provides a normal mode signal blind estimation system for implementing the normal mode signal blind estimation of the multi-rotor unmanned aerial vehicle positioning and tracking, and the software design of the three-dimensional space DOA blind estimation interface operating system comprises:
in order to continuously and deeply research blind estimation of a DOA pitch angle and an azimuth angle of a multi-rotor unmanned aerial vehicle in a three-dimensional space, and by applying LabVIEW software simulation, an interface operating system for the DOA blind estimation in the three-dimensional space is designed, so that the blind estimation research of a normal mode signal of the multi-rotor unmanned aerial vehicle can be better realized.
By using a LabVIEW virtual instrument to establish a mathematical model and an interface design of a normal mode signal blind estimation, on the basis of hardware with complete functions in a three-dimensional DOA blind estimation system software design, a virtual instrument technology is combined with extremely high software efficiency to complete various developments and simulation simulations. The software design block diagram of the operating interface system is shown in fig. 12.
When the software is opened, the system initial page of the software is firstly entered into a login interface, some production information of the software is displayed in the login interface, and after a correct user name and a login password are input, a login button is clicked, so that the main function page of the three-dimensional DOA blind estimation system can be entered, as shown in FIG. 13.
The correct user name and login password are input, and the login button is clicked, so that the host system can be accessed. However, the premise must be to input the user name and password that have already been set, that is: the user name "administrator", password "666" is entered.
If the correct user name and password are input, the user can enter the password successfully; otherwise, the system may prompt "user name and password entered incorrectly" to fail to access the host system, as shown in FIG. 14.
In order to design the login interface and realize related functions, the login interface adopts a tiled sequential structure and executes programming through a three-frame sequence. And writing a While loop in the first frame to play a role of circularly inputting the user name and the password, and when the input user name and the input password are correct, clicking the login can output a signal instruction of 'true' to the next frame for the first time, otherwise, the output is false. For the second frame, the third frame serves as a judgment output, and if the output is "true", the sub VI corresponding to the system function is turned on, as shown in fig. 15.
The invention utilizes LabVIEW simulation to realize the system function of normal mode signal DOA blind estimation for positioning and tracking of the multi-rotor unmanned aerial vehicle, and the realized overall system effect diagram is shown below.
The input part of the DOA blind estimation is the covariance matrix of the signals, the scanning precision of the azimuth angle and the pitch angle, and the output content is parameters such as space spectrum related parameters, specific values of the azimuth angle and the pitch angle of the received signals and the like.
The program diagram of LabVIEW is the place where the virtual instrument is programmed, and the wiring programming is performed by adopting various modularized devices. As can be observed from the figure, the whole is divided into two parts of modules, and one part of the modules is a function of carrying out eigenvalue decomposition on the covariance matrix; the other part is vector calculation determined by an azimuth angle and a pitch angle, and the signal flow is processed by the two parts of signals, and finally the calculation of spatial spectrum blind estimation is carried out.
Application examples. In order to prove the creativity and the technical value of the technical scheme of the invention, the part is the application example of the technical scheme of the claims on specific products or related technologies.
The method comprises the steps of firstly selecting a receiving end sensor with the number of the array elements of the wireless antenna array being 3 as a simulation base station, arranging the array at equal intervals, setting the number of sampling points to improve the universality and reliability of the experiment, selecting 3 far-field narrow-band signals which are drawn to be relatively independent, setting the incident wavelength transmitted by a signal source to be lambda, and setting the distance d between every two adjacent array elements to be lambda/2, namely setting the adjacent interval to be one half of the wavelength of the incident signal of the signal source.
The simulation assumes 3 signal source incident signals, and the set DOA angles in the direction of arrival are respectively: azimuth (30 °,50 °, 70 °); pitch angles (20 °,40 °,60 °), signal-to-noise ratios were all set to 10db, and a diagram of matlab simulation experiment results is shown in fig. 16. Through the simulation image, the positions of 3 wave crests in a oscillogram and the information of azimuth angles and pitch angles corresponding to the spectral peaks respectively can be found out according to the principle of spectral peak search of a spatial two-dimensional space MUSIC algorithm, 3 estimated angles can be obviously observed from the image, and the angles respectively corresponding to DOA angles are about: azimuth (30 °,50 °, 70 °); the pitch angles (20 degrees, 40 degrees and 60 degrees) are completely consistent with the actually pre-assumed incident angle signals, and the experimental study of blind estimation of the DOA of the normal mode signals by applying the two-dimensional space MUSIC algorithm is successfully realized and is consistent with the expected result. Next, the performance of MATLAB image simulation is utilized to perform reprocessing simulation on the set azimuth angle and the set pitch angle, the simulation result is shown in fig. 17, and the accuracy and the reliability of the research of the invention are verified again.
The invention respectively simulates azimuth angles and pitch angles of DOAs (direction of arrival) in different directions of arrival, which are incident from the normal mode signal sources of the multi-rotor unmanned aerial vehicle with different numbers, and the signal-to-noise ratio is set to be 10dB. And determining MATLAB simulation comparison results in different pitch angles by using a control variable method under the condition of the same azimuth angle, and deducing a relational expression of the number of the multi-rotor unmanned aerial vehicles and the normal mode signal incidence minimum critical angle obtained by blind estimation for the first time.
Example 1: (two multi-rotor unmanned signal)
(1) Setting DOA: azimuth angle (50 ° ); the pitch angle is (40 degrees, 60 degrees)
And (4) analyzing results: the wave crest is independent, and the angle is clear.
(2) Setting DOA: azimuth angle (50 ° ); the pitch angle is (50 degrees, 60 degrees)
And (4) analyzing results: the wave crest is independent, and the angle is clear.
(3) Setting DOA: azimuth angle is (50 ° ); the pitch angle is (50 degrees, 55 degrees)
And (4) analyzing results: the wave crests are independent, the angle can be observed, and the wave bottom is interfered.
(4) Setting DOA: azimuth angle (50 ° ); the pitch angle is (50 degrees, 52 degrees)
And (4) analyzing results: the wave crests are fuzzy, fused with each other and the angle is unknown.
And (4) conclusion: comparing and verifying the four simulation results, the following conclusion can be easily obtained, namely: when there are two multi-rotor drone signal sources, the minimum critical pitch angle is 2 ° at the same azimuth as shown in fig. 18.
Example 2: (three multi rotor unmanned signal)
(1) Setting DOA: the azimuth angle is (50 °,50 °,50 °); the pitch angle is (20 degrees, 40 degrees, 60 degrees)
And (4) analyzing results: the wave crest is independent, and the angle is clear.
(2) Setting DOA: the azimuth angle is (50 °,50 °,50 °); the pitch angle is (40 degrees, 50 degrees, 60 degrees)
And (4) analyzing results: the wave crest is relatively independent, the angle can be observed, and the wave bottom generates interference.
(3) Setting DOA: the azimuth angle is (50 °,50 °,50 °); the pitch angle is (45 degrees, 50 degrees, 55 degrees)
And (4) analyzing results: the wave crests are fuzzy, fused with each other and the angle is unknown.
And (4) conclusion: comparing and verifying the three simulation results, and easily obtaining the following conclusion that: when there are three multi-rotor drone signal sources, the minimum critical pitch angle is 5 ° at the same azimuth as shown in fig. 19.
Example 3: (four multi-rotor unmanned signal)
(1) Setting DOA: the azimuth angle is (50 °,50 °,50 °,50 °); the pitch angle is (20 degrees, 40 degrees, 60 degrees, 80 degrees)
And (4) analyzing results: the wave crest is independent, and the angle is clear.
(2) Setting DOA: the azimuth angle is (50 °,50 °,50 °,50 °); the pitch angle is (20 degrees, 35 degrees, 50 degrees, 65 degrees)
And (4) analyzing results: the interference is obvious, the peak intensities are different, and the angle can be roughly estimated.
(3) Setting DOA: the azimuth angle is (50 °,50 °,50 °,50 °); the pitch angle is (30 degrees, 40 degrees, 50 degrees, 60 degrees)
And (4) analyzing results: the peak fusion is severe and the angle is not available.
And (4) conclusion: comparing and verifying the three simulation results, and easily obtaining the following conclusion that: when there are four multi-rotor drone signal sources, the minimum critical pitch angle is 10 ° at the same azimuth as shown in fig. 20.
The invention is subjected to multiple times of experimental simulation, the obtained data is led into LabVIEW software, 12 groups of data acquired by simulation are sequenced, sorted and integrated by carrying out curve fitting processing on the result data, and the measured data quantity is subjected to curve fitting processing by using a least square method and an SVD algorithm, as shown in figure 21. According to a polynomial coefficient derived from LabVIEW, a relational expression between the number of the multi-rotor unmanned aerial vehicles and the normal mode signal minimum critical angle obtained through blind estimation is derived for the first time, as shown in a formula 22, X represents the number of the multi-rotor unmanned aerial vehicles, and Y represents the normal mode signal minimum critical angle obtained through blind estimation.
Y=1.51-1.38X+0.85X 2 (22)
Evidence of the relevant effects of the examples. The embodiment of the invention has some positive effects in the process of research and development or use, and indeed has great advantages compared with the prior art, and the following contents are described by combining data, graphs and the like in the experimental process.
According to the invention, a circular array model is designed for the first time to realize the positioning of the real-time position of the multi-rotor unmanned aerial vehicle, and the azimuth angle and the pitch angle of the multi-rotor unmanned aerial vehicle in the space are calculated by utilizing the rotation invariance of the signal subspace of the array signal receiving data covariance matrix based on the ESPRIT algorithm of the real-value beam forming space. Subsequently, an Unscented Kalman Filter (Unscented Kalman Filter, UKF) with higher precision and smaller calculation amount is adopted to design a circular array model of the dual system, as shown in fig. 22. The method is characterized in that a three-dimensional dynamic model is constructed by combining an ODAS open embedded test platform, the position information of the multi-rotor unmanned aerial vehicle is visually displayed in a space coordinate system, the position location and the target track tracking are realized, and compared with the traditional Extended Kalman Filter (EKF), the average accuracy error is reduced by 51.91%. As shown in fig. 23, the two graphs represent the azimuth angle of the target multi-rotor drone in the X, Y-axis horizontal plane direction and the pitch angle in the Z-axis direction in real time. Obviously, in the experimental design of the invention, the target multi-rotor unmanned aerial vehicle moves in the range of the Z-axis positive half shaft with the pitch angle of 30-60 degrees and the X-axis and Y-axis positive half shafts with the azimuth angle of 0-100 degrees, so that the aim of high-precision positioning and tracking is realized, the effectiveness of the device is verified, and the diversity of blind estimation of the normal mode signals for positioning and tracking the multi-rotor unmanned aerial vehicle is better enriched
Finally, the invention designs an interface operating system aiming at the blind estimation of the DOA in the three-dimensional space of the multi-rotor unmanned aerial vehicle, continuously and deeply researches the blind estimation of the pitch angle and the azimuth angle of the DOA in the three-dimensional space, and applies LabVIEW software simulation, thereby greatly promoting the position positioning, signal analysis processing and intelligent management in the normal-mode signal transmission process of the multi-rotor unmanned aerial vehicle, realizing automatic observation and operation, being safe and convenient, intuitively displaying the data of the real-time DOA in the three-dimensional space of the multi-rotor unmanned aerial vehicle on a digital screen, and having better human-computer interaction function.
The above description is only for the specific embodiment of the present invention, but the scope of the present invention is not limited thereto, and any modification, equivalent replacement, and improvement made by those skilled in the art within the technical scope of the present invention disclosed in the present invention should be covered within the scope of the present invention.
Claims (10)
1. A method for blind estimation of a normal mode signal for positioning and tracking of a multi-rotor unmanned aerial vehicle is characterized by comprising the following steps:
blind estimation is carried out on DOA of a normal mode signal of the multi-rotor unmanned aerial vehicle by using a 2D-MUSIC algorithm, angle information of the DOA is blind estimated by using a two-dimensional angle, and a real-time azimuth angle and a pitch angle of the unmanned aerial vehicle are estimated; observing the obtained real-time positions of different multi-rotor unmanned aerial vehicles according to the change of the number of signal sources of different unmanned aerial vehicles; and detecting the direction angle and the pitch angle of the multi-rotor unmanned aerial vehicle in real time by adopting a real-value beam forming space ESPRIT algorithm, an unscented Kalman filtering algorithm and an ODAS open embedded test system.
2. The method for blind estimation of a normal mode signal for multi-rotor unmanned aerial vehicle position tracking according to claim 1, comprising in particular:
the method comprises the steps that firstly, a 2D-MUSIC algorithm is used for carrying out blind estimation on DOA of a normal mode signal of the multi-rotor unmanned aerial vehicle, angle information of the DOA is estimated blindly by using a two-dimensional angle, and an operator is used as a reference point to estimate a real-time azimuth angle and a pitch angle of the unmanned aerial vehicle;
step two, deriving a relation between a normal mode signal incidence minimum critical angle of the multi-rotor unmanned aerial vehicle and the number of the unmanned aerial vehicle signal sources through a curve fitting method according to the number of the different unmanned aerial vehicle signal sources and the observed real-time positions of the different multi-rotor unmanned aerial vehicles;
and step three, constructing a platform of a multi-rotor unmanned aerial vehicle positioning and tracking blind estimation device based on a real-value beam forming space ESPRIT algorithm and combined with an unscented Kalman filtering algorithm, designing a dual-system circular array model, and detecting the direction angle and the pitch angle of the multi-rotor unmanned aerial vehicle by combining with an ODAS open embedded test system.
3. The method for blind estimation of the normal mode signal for multi-rotor unmanned aerial vehicle positioning and tracking according to claim 2, wherein the specific process of the first step is as follows:
establishing a mathematical model diagram, and assuming that the set antenna array is a two-way parallel antenna array distribution, wherein the distance between two adjacent array elements along the X-axis direction is d x The distance between two parallel arrays is d y The first uniform antenna array is composed of (N + 1) signal array elements arranged in a straight line, and the second uniformly distributed linear array is composed of N antenna arrays;
suppose there are P multi-rotor unmanned aerial vehicle position signals in the target airspace from the two-dimensional direction in spaceIncident on receiving sensors laid down by the base stations of the antenna array, wherein,respectively representing an azimuth angle and a pitch angle, wherein the azimuth angle represents an included angle between a projection line of an incoming wave direction of an incident position signal of the multi-rotor unmanned aerial vehicle on a xoy plane and an x axis; pitch angle represents multi-rotor drone incident position signalAnd the included angle between the projection line of the wave direction on the xoy plane and the incoming wave direction of the incident signal.
4. The method for blind estimation of the normal mode signal for multi-rotor unmanned aerial vehicle positioning and tracking according to claim 2, wherein the specific process of the second step is as follows:
(1.1) simulating azimuth angles and pitch angles of DOAs in different arrival directions, which are incident from normal mode signal sources of multi-rotor unmanned aerial vehicles with different numbers, setting signal-to-noise ratios to be 10dB, and determining MATLAB simulation comparison results at different pitch angles under the condition of the same azimuth angle by using a control variable method;
(1.2) importing simulation comparison result data into LabVIEW software, sequencing, arranging and integrating the simulation acquisition data, and performing curve fitting processing on the measured data quantity by using a least square method and an SVD algorithm;
(1.3) deriving a relational expression between the number of the multi-rotor unmanned aerial vehicles and the normal-mode signal minimum incidence critical angle obtained by blind estimation according to polynomial coefficients derived by LabVIEW software, wherein the expression is as follows:
Y=1.51-1.38X+0.85X 2
in the formula, X represents the number of the multi-rotor unmanned aerial vehicles, and Y represents the normal mode signal incidence minimum critical angle obtained by blind estimation.
5. The method for blind estimation of normal mode signals for multi-rotor drone position tracking according to claim 2, characterized in that said step three comprises the following steps:
(2.1) constructing a circular array model;
(2.2) carrying out DOA blind estimation based on the real-value beam forming space ESPRIT algorithm;
and (2.3) carrying out interference elimination based on an unscented Kalman filtering algorithm.
6. The method of blind estimation of normal mode signals for multi-rotor drone position tracking according to claim 5, characterized in that said circular array model is:
written in matrix form as:
X(t)=AS(t)+N(t)
wherein, X (t) is output data vector of M × 1 dimension; s (t) is a far-field sound source signal with dimension Nx 1; n (t) is noise data of M multiplied by 1 dimension and is Gaussian white noise, and the noise on each array element is irrelevant; a = [ a ] 1 (w 0 ),a 2 (w 0 ),…,a N (w 0 )]Is an M × N dimensional circular array matrix, a i (w 0 ) I =1,2, \8230, N denotes a steering vector, the expression is as follows:
in the formula, w 0 Is the angular frequency of the received signal, and w 0 =2πf 0 =2 π c/λ, c representing the speed of sound; tau is Mi Indicating the time delay of the ith signal received relative to the mth array element of the reference circular array element.
7. The method for blind estimation of normal mode signals for multi-rotor unmanned aerial vehicle position tracking according to claim 5, wherein the specific process of step (2.2) is:
the real-value beam forming space ESPRIT algorithm utilizes the rotation invariance of a signal subspace of a circular array element array receiving data covariance matrix to further calculate the azimuth angle and the pitch angle of a multi-rotor unmanned aerial vehicle position signal transmitted by a space signal source;
the circular array element array is composed of a plurality of sub-arrays, a plurality of linear arrays surround the sub-arrays, the number of the sub-array elements is arranged in a crossed manner, the sub-array elements are M, wherein the X (t) sub-array represents a combined array composed of 0 to M-1 arrays, and the Y (t) sub-array represents a combined array composed of 1 to M arrays;
if the overall signal subspace is assumed to be E x ,E x Comprising two signal subspaces E 1 And E 2 A translation relation of a non-singular matrix exists between the two subspaces, the translation relation is called as a rotation operator, and the expression is as follows:
E 1 Ψ=E 2
in the formula, Ψ represents Ψ = T -1 Phi T, T is a nonsingular matrix, and phi is a characteristic value diagonal matrix of psi; e 1 Representing the signal subspace acquired by the X (t) subarrays; e 2 Representing the signal subspace acquired by the Y (t) subarray;
the simplified expression of the rotation operator is as follows:
E1=AT
E 2 =AφT
deriving the relationship Ψ and φ:
φ=TΨT -1
and (4) decomposing psi through a characteristic value to obtain phi, and estimating the azimuth angle and the pitch angle of the position signal of the multi-rotor unmanned aerial vehicle in the space by using the phi.
8. A blind estimation system of constant modulus signals for multi-rotor drone position tracking implementing the method of blind estimation of constant modulus signals for multi-rotor drone position tracking according to any one of claims 1 to 7, characterized in that it comprises:
the 2D-MUSIC module is used for carrying out blind estimation on the DOA of the normal mode signal of the multi-rotor unmanned aerial vehicle and carrying out blind estimation on the angle information of the DOA by utilizing a two-dimensional angle;
the curve fitting module is used for deducing a relational expression between the minimum critical angle of normal mode signal incidence of the multi-rotor unmanned aerial vehicle and the number of unmanned aerial vehicle signal sources;
and the detection module is used for constructing a multi-rotor unmanned aerial vehicle positioning and tracking blind estimation device platform based on a real-value beam forming space ESPRIT algorithm and an unscented Kalman filtering algorithm, designing a dual-system circular array model, and detecting the direction angle and the pitch angle of the multi-rotor unmanned aerial vehicle in real time by combining an ODAS open embedded test system.
9. A blind estimation processing terminal of a normal mode signal for positioning and tracking of a multi-rotor unmanned aerial vehicle, which implements the blind estimation system of a normal mode signal according to claim 8, the specific processing procedures of the processing terminal are as follows:
the method comprises the steps of establishing a mathematical model and an interface design of normal mode signal blind estimation by using a LabVIEW virtual instrument, and completing various developments and simulation simulations by combining a software system on the basis of hardware with complete functions on the software design of a three-dimensional DOA blind estimation system.
10. The normal mode signal blind estimation processing terminal for multi-rotor drone position tracking according to claim 9, characterized in that said processing terminal comprises:
a system login interface, wherein after a correct user name and a login password are input in the login interface, a login button is clicked, and a main function page of the DOA blind estimation system in a three-dimensional space is entered; inputting a user name and a password which are set; if the input is wrong, the system prompts that the user name and the password are input wrongly and cannot enter the main system; the system login interface adopts a tiled sequential structure, and programming is executed through a three-frame sequence; writing a While loop in a first frame to play a role of circularly inputting a user name and a password, and when the input user name and the input password are correct, outputting a true signal instruction to a next frame at the first time by clicking login, otherwise, outputting the true signal instruction; compared with the second frame, the third frame plays a role in judging output, and if the output is true, the sub VI corresponding to the system function is opened; the input part of DOA blind estimation in the front panel image is frequency points, a covariance matrix of signals, the scanning precision of an azimuth angle and a pitch angle, and the output content is a space spectrum, the azimuth angle of a received signal, the pitch angle and error parameters; the program block diagram is a place where a virtual instrument is programmed, the virtual instrument is connected with a line for programming by adopting a modularized device, the whole program block is divided into two parts of modules, and one part of the module has the function of decomposing the characteristic value of the covariance matrix; the other part is vector calculation determined by an azimuth angle and a pitch angle, and signal flow is subjected to signal processing of the two parts, and finally spatial spectrum blind estimation is calculated;
and the system function operation interface comprises a front panel diagram and a program block diagram, and is used for realizing the system function of the normal mode signal DOA blind estimation for the positioning and tracking of the multi-rotor unmanned aerial vehicle by using LabVIEW simulation.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211385592.4A CN115685067B (en) | 2022-11-07 | 2022-11-07 | Constant modulus signal blind estimation method and system for multi-rotor unmanned aerial vehicle positioning tracking |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211385592.4A CN115685067B (en) | 2022-11-07 | 2022-11-07 | Constant modulus signal blind estimation method and system for multi-rotor unmanned aerial vehicle positioning tracking |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115685067A true CN115685067A (en) | 2023-02-03 |
CN115685067B CN115685067B (en) | 2023-06-06 |
Family
ID=85049542
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211385592.4A Active CN115685067B (en) | 2022-11-07 | 2022-11-07 | Constant modulus signal blind estimation method and system for multi-rotor unmanned aerial vehicle positioning tracking |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115685067B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116359836A (en) * | 2023-05-31 | 2023-06-30 | 成都金支点科技有限公司 | Unmanned aerial vehicle target tracking method and system based on super-resolution direction finding |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106950529A (en) * | 2017-03-21 | 2017-07-14 | 陕西理工大学 | Acoustic vector near field sources ESPRIT and MUSIC method for parameter estimation |
CN108645404A (en) * | 2018-03-30 | 2018-10-12 | 西安建筑科技大学 | A kind of small-sized multi-rotor unmanned aerial vehicle attitude algorithm method |
CN110492911A (en) * | 2019-07-10 | 2019-11-22 | 鹰潭泰尔物联网研究中心 | A kind of beam tracking method and system for UAV Communication |
CN113447919A (en) * | 2021-06-29 | 2021-09-28 | 重庆大学 | Extended Kalman prediction angle tracking method |
CN113630164A (en) * | 2021-08-09 | 2021-11-09 | 南京航空航天大学 | Three-dimensional UPF beam tracking method for millimeter wave communication platform of unmanned aerial vehicle |
CN114584192A (en) * | 2022-02-25 | 2022-06-03 | 南京邮电大学 | Anti-jitter robust beamforming optimization method for millimeter wave unmanned aerial vehicle communication system |
-
2022
- 2022-11-07 CN CN202211385592.4A patent/CN115685067B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106950529A (en) * | 2017-03-21 | 2017-07-14 | 陕西理工大学 | Acoustic vector near field sources ESPRIT and MUSIC method for parameter estimation |
CN108645404A (en) * | 2018-03-30 | 2018-10-12 | 西安建筑科技大学 | A kind of small-sized multi-rotor unmanned aerial vehicle attitude algorithm method |
CN110492911A (en) * | 2019-07-10 | 2019-11-22 | 鹰潭泰尔物联网研究中心 | A kind of beam tracking method and system for UAV Communication |
CN113447919A (en) * | 2021-06-29 | 2021-09-28 | 重庆大学 | Extended Kalman prediction angle tracking method |
CN113630164A (en) * | 2021-08-09 | 2021-11-09 | 南京航空航天大学 | Three-dimensional UPF beam tracking method for millimeter wave communication platform of unmanned aerial vehicle |
CN114584192A (en) * | 2022-02-25 | 2022-06-03 | 南京邮电大学 | Anti-jitter robust beamforming optimization method for millimeter wave unmanned aerial vehicle communication system |
Non-Patent Citations (1)
Title |
---|
LIU NING 等: "A modified virtual array ESPRIT algorithm based on real signal subspace" * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116359836A (en) * | 2023-05-31 | 2023-06-30 | 成都金支点科技有限公司 | Unmanned aerial vehicle target tracking method and system based on super-resolution direction finding |
CN116359836B (en) * | 2023-05-31 | 2023-08-15 | 成都金支点科技有限公司 | Unmanned aerial vehicle target tracking method and system based on super-resolution direction finding |
Also Published As
Publication number | Publication date |
---|---|
CN115685067B (en) | 2023-06-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109444810B (en) | Mutual-prime array non-grid DOA estimation method under nonnegative sparse Bayesian learning framework | |
Swindlehurst et al. | A performance analysis of subspace-based methods in the presence of model error. II. Multidimensional algorithms | |
Fuchs | On the application of the global matched filter to DOA estimation with uniform circular arrays | |
CN110261841A (en) | MIMO radar list based on iteration weighting proximal end projection measures vector DOA estimation method | |
CN106125041B (en) | The wideband source localization method of sparse recovery is weighted based on subspace | |
CN110082708A (en) | Nonuniform noise design and Wave arrival direction estimating method | |
CN108872926A (en) | A kind of amplitude and phase error correction and DOA estimation method based on convex optimization | |
CN112904272B (en) | Three-dimensional co-prime cubic array direction of arrival estimation method based on cross-correlation tensor | |
CN108802669B (en) | Two-dimensional direction of arrival estimation method, two-dimensional direction of arrival estimation device and terminal | |
CN115685067A (en) | Normal-mode signal blind estimation method and system for positioning and tracking of multi-rotor unmanned aerial vehicle | |
Tseng et al. | Steering vector estimation in uncalibrated arrays | |
CN111352063A (en) | Two-dimensional direction finding estimation method based on polynomial root finding in uniform area array | |
CN113567913A (en) | Two-dimensional plane DOA estimation method based on iteration reweighting dimension reduction | |
CN116224219A (en) | Array error self-correction atomic norm minimization DOA estimation method | |
CN105572629B (en) | A kind of two-dimentional direction-finding method of low computational complexity suitable for General Cell structure | |
CN105242236B (en) | Sensor position uncertainties bearing calibration in broadband signal super-resolution direction finding | |
Fu et al. | Off-grid error calibration for DOA estimation based on sparse Bayesian learning with weighted Sinc interpolation | |
CN110095750A (en) | Quick two-dimensional underdetermined angle measurement method based on quasi-stationary signal sparse reconstruction | |
CN114460531A (en) | Uniform linear array MUSIC spatial spectrum estimation method | |
CN112485755A (en) | Angle estimation method based on 2 q-order nested MIMO array structure | |
CN104391268B (en) | A kind of ripple reaches azimuthal detection method | |
Shu et al. | Passive direction finding with a pair of acoustic vector sensors using fourth-order cumulants | |
CN106872933A (en) | A kind of electromagnetic vector sensor array element posture position 6 DOF error self-correcting method | |
Zhang et al. | Two-dimensional underdetermined doa estimation of quasi-stationary signals via sparse bayesian learning | |
Rajani et al. | Direction of arrival estimation by using artificial neural networks |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right | ||
TR01 | Transfer of patent right |
Effective date of registration: 20240430 Address after: Room 501-01, Building 3, No. 99 Fuda Road, Ludu Street, Taicang City, Suzhou City, Jiangsu Province, 215412 Patentee after: Jiangsu Jinjing Intelligent Control Technology Co.,Ltd. Country or region after: China Address before: 341000 No. 86 Hongqi Avenue, Jiangxi, Ganzhou Patentee before: Jiangxi University of Science and Technology Country or region before: China |