CN115494472B - Positioning method based on enhanced radar wave signal, millimeter wave radar and device - Google Patents
Positioning method based on enhanced radar wave signal, millimeter wave radar and device Download PDFInfo
- Publication number
- CN115494472B CN115494472B CN202211430352.1A CN202211430352A CN115494472B CN 115494472 B CN115494472 B CN 115494472B CN 202211430352 A CN202211430352 A CN 202211430352A CN 115494472 B CN115494472 B CN 115494472B
- Authority
- CN
- China
- Prior art keywords
- point cloud
- cloud data
- clustering
- target
- targets
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
- G01S7/417—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section involving the use of neural networks
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/66—Radar-tracking systems; Analogous systems
Abstract
The invention relates to a positioning method and a positioning device based on enhanced radar wave signals, wherein the method comprises the following steps: acquiring a transmitting signal and one or more echo signals of a continuous millimeter wave radar, and sequentially performing frequency mixing, sampling and three-dimensional Fourier transform on the transmitting signal and the echo signals to obtain point cloud data; detecting and Doppler filtering the point cloud data by using a trained depth residual shrinkage network and a constant false alarm detection method, and determining one or more targets; based on various clustering algorithms, clustering the point cloud data after Doppler filtering, and calculating the pose information of one or more targets; and predicting the track of each target according to the pose information of the target and a Kalman filtering-based matching tracking algorithm. The constant false alarm method is improved by utilizing deep learning, the target attitude information is obtained according to the constant false alarm method, the target is positioned by combining with a Kalman matching tracking algorithm, and the accuracy and the adaptability of target positioning are improved.
Description
Technical Field
The invention belongs to the technical field of radar wave positioning, and particularly relates to a positioning method, a millimeter wave radar and a device based on enhanced radar wave signals.
Background
Compared with the rapid development of the radar technology in foreign countries, the research on the millimeter wave radar starts late in China, and the research on the millimeter wave radar related technology does not start until the 80 th century. With the large amount of education investment of the country on emerging technologies since the 21 st century, some colleges and universities and enterprises in China also enter the research of millimeter wave radars, and the millimeter wave radar technology is gradually developed. In indoor people positioning systems, there are currently a number of advanced devices and solutions. For example, the Nokia research center provides an HAIP positioning technology based on a central Bluetooth device; the Duke university of colleges and universities abroad proposes to connect WiFi for positioning by using mobile phone equipment. There are also many excellent solutions in people statistics. Such as Haekwev vision-derived vision sensors, and other electronics-derived infrared detectors.
When the conventional constant-virtual early warning is adopted to remove noise in the prior art, excessive noise cannot be effectively filtered, residual noise and clutter can often influence the feature extraction of a motion attitude, larger errors can be generated, and the final accuracy is not very high.
On the other hand, the humidity is high in an indoor bathroom scene, a certain attenuation effect is achieved on millimeter waves emitted by a radar, and due to the fact that multipath interference is caused by the reflection phenomenon of the surfaces of a mirror, glass and a smooth object on the millimeter waves, the path information of people cannot be accurately identified.
Disclosure of Invention
In order to enhance the radar wave signal and improve the accuracy and adaptability of positioning, a first aspect of the present invention provides a positioning method based on enhanced radar wave signal, including: acquiring a transmitting signal and one or more echo signals of a continuous millimeter wave radar, and sequentially performing frequency mixing, sampling and three-dimensional Fourier transform on the transmitting signal and the echo signals to obtain one or more point cloud data; detecting and Doppler filtering the point cloud data by using a trained depth residual shrinkage network and a constant false alarm detection method, and determining one or more targets; based on various clustering algorithms, clustering the point cloud data after Doppler filtering, and calculating the pose information of one or more targets; and predicting the track of each target according to the pose information of the target and a Kalman filtering-based matching tracking algorithm.
In some embodiments of the present invention, the detecting and doppler filtering the point cloud data by using the trained deep residual shrinkage network and the constant false alarm detection method, and determining one or more targets includes: determining one or more threshold values of a constant false alarm detection method by using a trained deep residual shrinkage network; detecting the point cloud data based on the one or more threshold values and determining one or more targets from the point cloud data.
Further, the depth residual shrinking network comprises: at least one sub-network for extracting a set of features from an input radar wave waveform signal; and the full connection layer is used for adjusting the characteristics to be between 0 and 1 through an AReLU activation function to obtain one or more threshold values of the constant false alarm detection method.
In some embodiments of the present invention, the clustering the point cloud data after doppler filtering based on multiple clustering algorithms, and calculating pose information of one or more targets includes: clustering the point cloud data after Doppler filtering based on a DBSCAN clustering algorithm to obtain a first attitude queue of successful clustering; and clustering the first attitude queue based on a K-means clustering algorithm and a Gaussian mixture model to obtain a successfully clustered second attitude queue, wherein the second attitude queue comprises one or more targets with different pose information.
In some embodiments of the present invention, predicting the trajectory of each target according to the pose information of the target and a kalman filter-based matching tracking algorithm comprises: predicting a plurality of tracks of one or more targets by expanding a Kalman particle filter by using the pose information of the current frame and the predicted pose information of the previous frame of each target; and matching each target with a plurality of corresponding tracks by using a Hungarian algorithm to obtain the predicted track of each target.
In the above embodiment, the acquiring a transmission signal of a continuous millimeter wave radar and one or more echo signals thereof, and sequentially performing frequency mixing, sampling and three-dimensional fourier transform on the transmission signal and the echo signals to obtain one or more point cloud data includes: acquiring a transmitting signal and one or more echo signals of a continuous millimeter wave radar, and mixing the transmitting signal and the echo signals to generate one or more intermediate frequency signals; and carrying out ADC (analog to digital converter) sampling on the plurality of intermediate frequency signals, and carrying out three-dimensional Fourier transform on the sampled signals to obtain a plurality of point cloud data.
In a second aspect of the present invention, there is provided a millimeter wave radar comprising: the acquisition module is used for acquiring a transmitting signal and one or more echo signals of a continuous millimeter wave radar, and sequentially carrying out frequency mixing, sampling and three-dimensional Fourier transform on the transmitting signal and the echo signals to obtain one or more point cloud data; the determining module is used for detecting and Doppler-filtering the point cloud data by using a trained deep residual shrinkage network and a constant false alarm detection method and determining one or more targets; the clustering module is used for clustering the point cloud data after Doppler filtering based on a plurality of clustering algorithms and calculating the pose information of one or more targets; and the prediction module is used for predicting the track of each target according to the pose information of the target and a Kalman filtering-based matching tracking algorithm.
In a third aspect of the invention, a positioning device based on enhanced radar wave signals is provided, which comprises the millimeter wave radar provided in the second aspect.
In a fourth aspect of the present invention, there is provided an electronic apparatus comprising: one or more processors; a storage device, configured to store one or more programs, which when executed by the one or more processors, cause the one or more processors to implement the positioning method based on enhanced radar wave signals provided by the present invention in the first aspect.
In a fifth aspect of the present invention, a computer-readable medium is provided, on which a computer program is stored, wherein the computer program, when executed by a processor, implements the enhanced radar wave signal-based positioning method provided in the first aspect of the present invention.
The beneficial effects of the invention are:
the method improves a constant false alarm detection method by utilizing a deep learning model, and obtains the attitude information of a target according to the constant false alarm detection method; the target is positioned based on the attitude information and combined with the Kalman-based matching tracking algorithm, so that the accuracy and the adaptability of positioning, particularly indoor positioning are improved.
Drawings
Fig. 1 is a basic flow diagram of a positioning method based on enhanced radar wave signals in some embodiments of the present invention;
fig. 2 is a detailed flow chart of a positioning method based on enhanced radar wave signals according to some embodiments of the present invention;
FIG. 3 is a schematic diagram of the basic principle of determining a constant false alarm threshold for a deep residual shrinkage network in some embodiments of the present invention;
FIG. 4 is a flow chart for determining one or more targets using a clustering algorithm in some embodiments of the invention;
FIG. 5 is a flowchart of the operation of an iterative extend Kalman particle filter in some embodiments of the present invention;
FIG. 6 is a workflow diagram of the Hungarian match tracking algorithm in some embodiments of the invention;
FIG. 7 is a schematic diagram of a positioning device based on enhanced radar wave signals in some embodiments of the present invention;
fig. 8 is a schematic structural diagram of an electronic device in some embodiments of the invention.
Detailed Description
The principles and features of this invention are described below in conjunction with the following drawings, which are set forth by way of illustration only and are not intended to limit the scope of the invention.
Example 1
Referring to fig. 1 and fig. 2, in a first aspect of the present invention, there is provided a positioning method based on enhanced radar wave signals, including: s100, acquiring a transmitting signal and one or more echo signals of a continuous millimeter wave radar, and sequentially performing frequency mixing, sampling and three-dimensional Fourier transform on the transmitting signal and the echo signals to obtain one or more point cloud data; s200, detecting and Doppler filtering the point cloud data by using a trained depth residual error shrinkage network and a constant false alarm detection method, and determining one or more targets; s300, clustering the point cloud data subjected to Doppler filtering based on multiple clustering algorithms, and calculating pose information of one or more targets; s400, predicting the track of each target according to the pose information of the target and a Kalman filtering-based matching tracking algorithm.
It can be understood that, when the threshold value is calculated, the deep learning network replaces a calculation formula of the threshold value, noise in radar waves can be better removed through the threshold value trained by the network, and the later gesture recognition has higher accuracy. A deep interpolation gathering network (a deep residual shrinkage network) is a novel improvement on the deep residual network, and soft thresholding is introduced into the residual network as a nonlinear layer, so that a threshold value can be trained better.
In view of this, referring to fig. 3, in S200 according to some embodiments of the present invention, the detecting and doppler filtering the point cloud data by using the trained depth residual shrinkage network and the constant false alarm detection method, and determining one or more targets includes: s201, determining one or more threshold values of a constant false alarm detection method by using a depth residual shrinkage network after training is completed; s202, detecting the point cloud data based on the one or more threshold values, and determining one or more targets from the point cloud data.
Specifically, a Deep Residual Shrinkage Network (DRSN) is a feature learning method oriented to strong noise or highly redundant data. The depth residual shrink network merges the depth residual network, SEnet, and soft threshold function. The depth residual shrinkage network replaces the "re-weighting" in SEnet in residual mode with "soft thresholding". In SENET, the embedded small network is used for obtaining a set of weight coefficients; in the deep residual shrinkage network, the small network is used to obtain a set of thresholds. For the flow of the sub-network in the network, all elements in the input waveform signal are taken as absolute values. Then performing global mean pooling and averaging to obtain a group of features, which can be denoted as A; in the other path, the features after global mean pooling are input to a small fully connected network. This fully connected network can adjust the input to between 0 and 1, denoted as α, with an AReLU activation function as the last step. The final threshold value is alpha multiplied by A, the calculation efficiency is high, the convergence can be fast, and the expectation problem of the activated neuron can be effectively solved. Finally, certain basic modules and convolution layers, activation functions, global mean pooling and the like are stacked, and a complete deep interpolation gathering network is formed. The most suitable threshold value can be trained effectively by the deep interpolation gathering network.
The soft threshold function expresses the form:
x denotes an input characteristic, y denotes an output characteristic,
representing a threshold value. The derivative is expressed as:
it can be seen from the formula that the derivative is either 1 or 0, which can effectively prevent the gradient from exploding or disappearing. Ith sample inputxiThe AReLU activation function of (a) is as follows:
further, the depth residual shrinking network comprises: at least one sub-network for extracting a set of features from an input radar wave waveform signal; and the full connection layer is used for adjusting the characteristics to be between 0 and 1 through an AReLU activation function to obtain one or more threshold values of the constant false alarm detection method.
Referring to fig. 4, in S300 according to some embodiments of the present invention, the clustering the doppler-filtered point cloud data based on multiple clustering algorithms and calculating pose information of one or more targets includes: s301, clustering the point cloud data after Doppler filtering based on a DBSCAN clustering algorithm to obtain a first attitude queue of successful clustering; s302, clustering is carried out on the first attitude queue based on a K-means clustering algorithm and a Gaussian mixture model, and a second attitude queue which is successfully clustered is obtained, wherein the second attitude queue comprises one or more targets with different pose information.
In particular, clustering of the present disclosure refers to putting together a class of real objects having similar attributes. And detecting the attributes of the real objects, and grouping the real objects with consistent attributes into a group to find the grouping information. In the present embodiment, a moving person (object) is extracted from an environment (a plurality of objects in an image). And (3) performing cooperative and mutual assistance by adopting a plurality of clustering methods, so that data can be better clustered, clustering the point clouds roughly by adopting a large threshold DBSCAN, then clustering by adopting a small threshold, putting the point clouds into a successful queue if the clustering is successful, and storing the group of information into a pending queue if the clustering is unsuccessful. In the pending queue, because the large threshold DBSCAN clustering gives a rough class, namely a K value, the K value is used for performing K-means clustering and Gaussian mixture model clustering respectively, when the clustering effects of the two are ideal, the clustering result is stored in a successful queue, otherwise, the clustering result is discarded.
In step S400 of some embodiments of the present invention, predicting the trajectory of each target according to the pose information of the target and a kalman filter-based matching tracking algorithm includes: predicting a plurality of tracks of one or more targets by expanding a Kalman particle filter by using the pose information of the current frame and the predicted pose information of the previous frame of each target; and matching each target with a plurality of corresponding tracks by using a Hungarian algorithm to obtain the predicted track of each target.
Specifically, the position of the point cloud in the detection range of the millimeter wave radar system changes all the time. Because the system has a point cloud reflection phenomenon, the actually generated point cloud position may have deviation with a real moving object. Therefore, in this embodiment, the position and the speed of the next point cloud need to be predicted according to the point cloud of the previous frame, so as to complete the trajectory prediction of the point cloud.
Referring to fig. 5, in the embodiment, an iterative extended kalman particle filter (IEKF) is used for tracking prediction, and part of key codes are as follows:
(1) The initialization is such that k = 0 and,
for i = 0：N s from the prior probability density p (x 0 ) Sampling point is extracted at random in well
；
（2）for k = 1:N,
for i = 1:N s
By using
Calculating a Jacobian matrix:
particle renewal with IEKF:
calculating a Jacobian matrix of the observation model:
updating covariance matrix
,
Updating system state
end for
end for
for i = 1: N s
Normalizing the weight:
,
end for
end for。
referring to fig. 6, the hungarian algorithm matches each target with its corresponding plurality of tracks by matching the new observations with existing tracked objects. The algorithm of the invention is mainly applied in narrow space, and the multipath effect can be caused by the reflection of the ground surface possibly through a mirror surface, the multipath interference phenomenon can be generated, and the performance of the radar can be influenced. Assume that there are M cluster results, but N tracks may be generated due to multipath effects, where N > M, N = M + N, N being a false path generated by multipath effects. Firstly, the distance between each two clustering results and the track is calculated to obtain a cost matrix of N x M. Under the condition that the obtained matrix is not a square matrix, the Hungary algorithm cannot be directly used for distribution; it is therefore necessary to complete the matrix and fill the missing rows/columns with an infinite number. Thus, a distance matrix which can be used for Hungarian algorithm is obtained. And calculating the distance matrix by using a Hungarian algorithm to obtain the distribution result of the track and the cluster, and regarding the track-cluster pair with the distance exceeding a set threshold value, considering that the distribution is invalid, namely, the track and the cluster are not distributed.
Constructing a cost matrix
And calculating a cost matrix:
x ij in order to be an element of the cost matrix,
is the distance between one clustered result and a different track,his the height of the person and is the height of the person,
is the rate of absorption loss, and is,
is the decay in the case of drying out,
is the decay at a certain humidity level of the sample,fis the frequency of the wave(s),N''(f) Is the imaginary part of the frequency dependent complex index of refraction.
In step S100 of the foregoing embodiment, the obtaining of the emission signal of the continuous millimeter wave radar and the one or more echo signals thereof, and sequentially performing frequency mixing, sampling and three-dimensional fourier transform on the emission signal and the echo signals to obtain one or more point cloud data includes: acquiring a transmitting signal and one or more echo signals of a continuous millimeter wave radar, and mixing the transmitting signal and the echo signals to generate one or more intermediate frequency signals; and carrying out ADC (analog to digital converter) sampling on the plurality of intermediate frequency signals, and carrying out three-dimensional Fourier transform on the sampled signals to obtain a plurality of point cloud data.
In particular, the TX (transmitting) antenna continuously transmits a frequency modulated signal, part of which is reflected back to the radar RX (receiving) antenna, depending on the object surface type and shape. The mixer combines the TX and RX signals together to generate an IF (intermediate frequency) signal having a frequency that is the difference between the TX and RX frequencies and an initial phase that is the difference in phase at the current time. And then ADC (analog-digital conversion) sampling is carried out on the intermediate frequency signals, improved three-dimensional Fourier transform is carried out on the sampled signals, and the point cloud data is obtained by operation on the transformed data. And performing composite constant virtual early warning on point cloud data to filter out clutters and find a target, transmitting the filtered point cloud to a clustering algorithm, clustering the point cloud, calculating the direction (pose information) corresponding to a clustering result, transmitting corresponding data to particle filtering, predicting the track of the next frame through the particle filtering, and matching the track to different tracks through a matching algorithm.
The three-dimensional Fourier algorithm comprises the following steps: distance dimension FFT, extracting a frequency value corresponding to a peak value after Fourier transform, and converting the frequency into a distance because the frequency is in direct proportion to the distance; doppler FFT, which is used for solving a corresponding phase in each chrip (linear frequency modulation signal) after distance dimensional Fourier transform is carried out, and solving the speed according to the phase; and the angle dimension FFT is used for carrying out FFT by taking the phases at different antennas as input quantities on the basis of the Doppler FFT so as to estimate the value of the azimuth angle.
Example 2
Referring to fig. 7, in a second aspect of the present invention, there is provided a positioning apparatus based on an enhanced radar wave signal, including the millimeter wave radar 1 provided in the second aspect, including: the acquisition module 11 is configured to acquire a transmission signal of a continuous millimeter wave radar and one or more echo signals thereof, and sequentially perform frequency mixing, sampling and three-dimensional fourier transform on the transmission signal and the echo signals to obtain one or more point cloud data; the determining module 12 is configured to perform detection and doppler filtering on the point cloud data by using a trained deep residual shrinkage network and a constant false alarm detection method, and determine one or more targets; the clustering module 13 is configured to cluster the point cloud data after doppler filtering based on multiple clustering algorithms, and calculate pose information of one or more targets; and the prediction module 14 is configured to predict a track of each target according to the pose information of the target and a kalman filter-based matching tracking algorithm.
In a third aspect of the invention, a positioning device based on enhanced radar wave signals is provided, which comprises the millimeter wave radar provided in the second aspect.
Specifically, an FMCW millimeter wave radar and an antenna are used as a transceiver to realize conversion between electric energy and electromagnetic waves. The TX antenna continuously transmits a frequency modulated signal, and depending on the type and shape of the object surface, part of the electromagnetic wave is reflected back to the radar RX antenna. The mixer combines the TX and RX signals to generate an IF (intermediate frequency) signal having a frequency that is the difference between the frequencies of TX and RX, and an initial phase that is the difference in phase at the current time. And then ADC (analog-digital conversion) sampling is carried out on the intermediate frequency signal, improved three-dimensional Fourier transform is carried out on the sampled signal, and the converted data is operated to obtain point cloud data. And performing composite constant virtual early warning on point cloud data to filter out clutter and find a target, transmitting the filtered point cloud to a clustering algorithm, clustering the point cloud, calculating the direction corresponding to a clustering result, transmitting corresponding data to particle filtering, predicting the track of the next frame through the particle filtering, and matching the track to different tracks through the matching algorithm.
Example 3
Referring to fig. 8, a fourth aspect of the present invention provides an electronic apparatus comprising: one or more processors; storage means for storing one or more programs which, when executed by the one or more processors, cause the one or more processors to carry out the method of the invention in the first aspect.
The electronic device 500 may include a processing means (e.g., central processing unit, graphics processor, etc.) 501 that may perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM) 502 or a program loaded from a storage means 508 into a Random Access Memory (RAM) 503. In the RAM 503, various programs and data necessary for the operation of the electronic apparatus 500 are also stored. The processing device 501, the ROM 502, and the RAM 503 are connected to each other through a bus 504. An input/output (I/O) interface 505 is also connected to bus 504.
The following devices may be connected to the I/O interface 505 in general: input devices 506 including, for example, a touch screen, touch pad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; output devices 507 including, for example, a Liquid Crystal Display (LCD), speakers, vibrators, and the like; a storage device 508 including, for example, a hard disk; and a communication device 509. The communication means 509 may allow the electronic device 500 to communicate with other devices wirelessly or by wire to exchange data. While fig. 8 illustrates an electronic device 500 having various means, it is to be understood that not all illustrated means are required to be implemented or provided. More or fewer devices may alternatively be implemented or provided. Each block shown in fig. 8 may represent one device or may represent multiple devices as desired.
In particular, according to an embodiment of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network via the communication means 509, or installed from the storage means 508, or installed from the ROM 502. The computer program, when executed by the processing device 501, performs the above-described functions defined in the methods of embodiments of the present disclosure. It should be noted that the computer readable medium described in the embodiments of the present disclosure may be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In embodiments of the disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In embodiments of the present disclosure, however, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, optical cables, RF (radio frequency), etc., or any suitable combination of the foregoing.
The computer readable medium may be embodied in the electronic device; or may exist separately without being assembled into the electronic device. The computer readable medium carries one or more computer programs which, when executed by the electronic device, cause the electronic device to:
computer program code for carrying out operations for embodiments of the present disclosure may be written in one or more program embodiment languages, including an object oriented program embodiment language such as Java, smalltalk, C + +, python, and conventional procedural program embodiment languages, such as the "C" language or similar program embodiment languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (9)
1. A positioning method based on enhanced radar wave signals is characterized by comprising the following steps:
acquiring a transmitting signal and one or more echo signals of a continuous millimeter wave radar, and sequentially performing frequency mixing, sampling and three-dimensional Fourier transform on the transmitting signal and the echo signals to obtain one or more point cloud data;
detecting and Doppler filtering the point cloud data by using a trained depth residual error shrinkage network and a constant false alarm detection method, and determining one or more targets; detecting the point cloud data based on the one or more threshold values and determining one or more targets from the point cloud data;
based on various clustering algorithms, clustering the point cloud data after Doppler filtering, and calculating the pose information of one or more targets;
and predicting the track of each target according to the pose information of the target and a Kalman filtering-based matching tracking algorithm.
2. The enhanced radar wave signal-based positioning method according to claim 1, wherein the depth residual shrinking network comprises:
at least one sub-network for extracting a set of features from an input radar wave waveform signal;
and the full connection layer is used for adjusting the characteristics to be between 0 and 1 through an AReLU activation function to obtain one or more threshold values of the constant false alarm detection method.
3. The positioning method based on the enhanced radar wave signal according to claim 1, wherein the clustering the point cloud data after the doppler filtering based on the multiple clustering algorithms and calculating pose information of one or more targets comprises:
clustering the point cloud data after Doppler filtering based on a DBSCAN clustering algorithm to obtain a first attitude queue of successful clustering;
and clustering the first attitude queue based on a K-means clustering algorithm and a Gaussian mixture model to obtain a successfully clustered second attitude queue, wherein the second attitude queue comprises one or more targets with different pose information.
4. The positioning method based on the enhanced radar wave signal according to claim 1, wherein the predicting the track of each target according to the pose information of the target and a Kalman filtering based matching tracking algorithm comprises:
predicting a plurality of tracks of one or more targets by expanding a Kalman particle filter by using the pose information of the current frame and the predicted pose information of the previous frame of each target;
and matching each target with a plurality of corresponding tracks by using a Hungarian algorithm to obtain the predicted track of each target.
5. The positioning method based on the enhanced radar wave signal according to any one of claims 1 to 4, wherein the obtaining of the emission signal of the continuous millimeter wave radar and one or more echo signals thereof, and the sequentially performing frequency mixing, sampling and three-dimensional Fourier transform on the emission signal and the echo signal to obtain one or more point cloud data comprises:
acquiring a transmitting signal and one or more echo signals of a continuous millimeter wave radar, and mixing the transmitting signal and the echo signals to generate one or more intermediate frequency signals;
and carrying out ADC (analog to digital converter) sampling on the plurality of intermediate frequency signals, and carrying out three-dimensional Fourier transform on the sampled signals to obtain a plurality of point cloud data.
6. A millimeter wave radar, comprising:
the acquisition module is used for acquiring a transmitting signal and one or more echo signals of a continuous millimeter wave radar, and sequentially carrying out frequency mixing, sampling and three-dimensional Fourier transform on the transmitting signal and the echo signals to obtain one or more point cloud data;
the determining module is used for detecting and Doppler filtering the point cloud data by using the trained deep residual error shrinkage network and the constant false alarm detection method and determining one or more targets; detecting the point cloud data based on the one or more threshold values and determining one or more targets from the point cloud data;
the clustering module is used for clustering the point cloud data after Doppler filtering based on a plurality of clustering algorithms and calculating the pose information of one or more targets;
and the prediction module is used for predicting the track of each target according to the pose information of the target and a Kalman filtering-based matching tracking algorithm.
7. A positioning device based on an enhanced radar wave signal, comprising the millimeter wave radar of claim 6.
8. An electronic device, comprising: one or more processors; storage means for storing one or more programs which, when executed by the one or more processors, cause the one or more processors to implement the positioning method based on enhanced radar wave signals according to any one of claims 1 to 5.
9. A computer-readable medium, on which a computer program is stored, wherein the computer program, when being executed by a processor, implements the enhanced radar wave signal-based positioning method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211430352.1A CN115494472B (en) | 2022-11-16 | 2022-11-16 | Positioning method based on enhanced radar wave signal, millimeter wave radar and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211430352.1A CN115494472B (en) | 2022-11-16 | 2022-11-16 | Positioning method based on enhanced radar wave signal, millimeter wave radar and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115494472A CN115494472A (en) | 2022-12-20 |
| CN115494472B true CN115494472B (en) | 2023-03-10 |
Family
ID=85116283
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211430352.1A Active CN115494472B (en) | 2022-11-16 | 2022-11-16 | Positioning method based on enhanced radar wave signal, millimeter wave radar and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115494472B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116879863B (en) * | 2023-09-09 | 2023-12-05 | 德心智能科技(常州)有限公司 | Multi-target measuring method and system for continuous wave 4D millimeter wave radar |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2593436A1 (en) * | 2006-11-09 | 2008-05-09 | Raytheon Canada Limited | Dual beam radar system |
| JP2018205174A (en) * | 2017-06-06 | 2018-12-27 | 株式会社東芝 | Radar device and radar signal processing method thereof |
| CN109766811A (en) * | 2018-12-31 | 2019-05-17 | 复旦大学 | The end-to-end detection and recognition methods of sea ship in a kind of satellite-borne SAR image |
| CN110363151A (en) * | 2019-07-16 | 2019-10-22 | 中国人民解放军海军航空大学 | Based on the controllable radar target detection method of binary channels convolutional neural networks false-alarm |
| CN112068120A (en) * | 2020-08-29 | 2020-12-11 | 西安电子工程研究所 | micro-Doppler time-frequency plane individual soldier squad identification method based on two-dimensional Fourier transform |
| CN112859033A (en) * | 2021-02-23 | 2021-05-28 | 加特兰微电子科技(上海)有限公司 | Target detection method, device and related equipment |
| CN113534120A (en) * | 2021-07-14 | 2021-10-22 | 浙江大学 | Multi-target constant false alarm rate detection method based on deep neural network |
| CN113938232A (en) * | 2020-07-13 | 2022-01-14 | 华为技术有限公司 | Communication method and communication device |
| EP3943970A1 (en) * | 2020-07-24 | 2022-01-26 | Aptiv Technologies Limited | Methods and systems for detection of objects in a vicinity of a vehicle |
| CN115308708A (en) * | 2022-08-03 | 2022-11-08 | 浙江中力机械股份有限公司 | Tray pose identification method and system based on laser radar |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11719805B2 (en) * | 2020-11-18 | 2023-08-08 | Infineon Technologies Ag | Radar based tracker using empirical mode decomposition (EMD) and invariant feature transform (IFT) |
-
2022
- 2022-11-16 CN CN202211430352.1A patent/CN115494472B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2593436A1 (en) * | 2006-11-09 | 2008-05-09 | Raytheon Canada Limited | Dual beam radar system |
| JP2018205174A (en) * | 2017-06-06 | 2018-12-27 | 株式会社東芝 | Radar device and radar signal processing method thereof |
| CN109766811A (en) * | 2018-12-31 | 2019-05-17 | 复旦大学 | The end-to-end detection and recognition methods of sea ship in a kind of satellite-borne SAR image |
| CN110363151A (en) * | 2019-07-16 | 2019-10-22 | 中国人民解放军海军航空大学 | Based on the controllable radar target detection method of binary channels convolutional neural networks false-alarm |
| CN113938232A (en) * | 2020-07-13 | 2022-01-14 | 华为技术有限公司 | Communication method and communication device |
| EP3943970A1 (en) * | 2020-07-24 | 2022-01-26 | Aptiv Technologies Limited | Methods and systems for detection of objects in a vicinity of a vehicle |
| CN112068120A (en) * | 2020-08-29 | 2020-12-11 | 西安电子工程研究所 | micro-Doppler time-frequency plane individual soldier squad identification method based on two-dimensional Fourier transform |
| CN112859033A (en) * | 2021-02-23 | 2021-05-28 | 加特兰微电子科技(上海)有限公司 | Target detection method, device and related equipment |
| CN113534120A (en) * | 2021-07-14 | 2021-10-22 | 浙江大学 | Multi-target constant false alarm rate detection method based on deep neural network |
| CN115308708A (en) * | 2022-08-03 | 2022-11-08 | 浙江中力机械股份有限公司 | Tray pose identification method and system based on laser radar |
Non-Patent Citations (2)
| Title |
|---|
| DL-CFAR: A Novel CFAR Target Detection Method Based on Deep Learning;Chia-Hung Lin 等;《2019 IEEE 90th Vehicular Technology Conference (VTC2019-Fall)》;20191107;第1-6页 * |
| 基于Faster_R-CNN网络的海面目标检测方法;潘美艳 等;《现代雷达》;20210630;第43卷(第6期);第19-26页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115494472A (en) | 2022-12-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20130217418A1 (en) | Hybrid method for high accuracy and cost-effective prediction of mobile device positions through mobile networks | |
| CN109324315B (en) | Space-time adaptive radar clutter suppression method based on double-layer block sparsity | |
| CN115494472B (en) | Positioning method based on enhanced radar wave signal, millimeter wave radar and device | |
| CN105866740B (en) | A kind of underwater sound Matched Field localization method based on compressed sensing | |
| CA2418786A1 (en) | High-precision 3-d location device | |
| CN108717174B (en) | Information theory-based passive cooperative positioning method for predicting rapid covariance interaction fusion | |
| CN112346030B (en) | Super-resolution direction-of-arrival estimation method for unmanned aerial vehicle group | |
| CN113267754B (en) | Three-dimensional grid-based terrain occlusion radar detection range calculation method | |
| CN107390229B (en) | A kind of processing method and its processing unit of anemometry laser radar signal | |
| CN114563784B (en) | Shipborne environment intrusion detection method and system based on double millimeter wave radar | |
| CN111964667B (en) | geomagnetic-INS (inertial navigation System) integrated navigation method based on particle filter algorithm | |
| CN115291207A (en) | Multi-target detection method for small rotor unmanned aerial vehicle based on MIMO radar | |
| CN108763158A (en) | Frequency difference combined calculation method and system when a kind of | |
| CN111401180A (en) | Neural network recognition model training method and device, server and storage medium | |
| CN116430126B (en) | Electromagnetic background cognition-based electromagnetic silence target detection method and device and computer equipment | |
| CN109799477B (en) | Millimeter wave Internet of vehicles oriented sequential vehicle fingerprint positioning method and device | |
| CN113514796B (en) | Passive positioning method, system and medium | |
| CN115131756A (en) | Target detection method and device | |
| Chervoniak et al. | Passive acoustic radar system for flying vehicle localization | |
| US10720949B1 (en) | Real-time time-difference-of-arrival (TDOA) estimation via multi-input cognitive signal processor | |
| CN108981707B (en) | Passive tracking multi-target method based on time difference measurement box particle PHD | |
| CN113439274A (en) | Identity recognition method, terminal device and computer storage medium | |
| US9800973B1 (en) | Sound source estimation based on simulated sound sensor array responses | |
| Ulmschneider | Cooperative multipath assisted positioning | |
| CN117592381B (en) | Atmospheric waveguide parameter inversion model training method, device, equipment and medium |
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 | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Gao Junfeng Inventor after: Zhao Jia Inventor after: Zhang Wenjia Inventor after: Zhang Bingyang Inventor after: Xiang Jie Inventor after: Huang Long Inventor after: Fu Junya Inventor after: Cao Shuqi Inventor before: Gao Junfeng Inventor before: Zhao Jia Inventor before: Zhang Bingyang Inventor before: Xiang Jie Inventor before: Huang Long Inventor before: Fu Junya Inventor before: Cao Shuqi |