CN116299473A - Method for detecting crossing target based on MIMO millimeter wave radar - Google Patents
Method for detecting crossing target based on MIMO millimeter wave radar Download PDFInfo
- Publication number
- CN116299473A CN116299473A CN202310339838.2A CN202310339838A CN116299473A CN 116299473 A CN116299473 A CN 116299473A CN 202310339838 A CN202310339838 A CN 202310339838A CN 116299473 A CN116299473 A CN 116299473A
- Authority
- CN
- China
- Prior art keywords
- point cloud
- scene
- algorithm
- target
- boundary
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 26
- 238000001514 detection method Methods 0.000 claims abstract description 36
- 230000003068 static effect Effects 0.000 claims abstract description 10
- 230000008030 elimination Effects 0.000 claims abstract description 5
- 238000003379 elimination reaction Methods 0.000 claims abstract description 5
- 238000001914 filtration Methods 0.000 claims description 7
- 238000007781 pre-processing Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 abstract description 7
- 239000011159 matrix material Substances 0.000 description 9
- 238000005259 measurement Methods 0.000 description 9
- 238000012545 processing Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 3
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- -1 O-F-O Chemical class 0.000 description 1
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- CLOMYZFHNHFSIQ-UHFFFAOYSA-N clonixin Chemical compound CC1=C(Cl)C=CC=C1NC1=NC=CC=C1C(O)=O CLOMYZFHNHFSIQ-UHFFFAOYSA-N 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000004080 punching Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
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
- 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
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
- G01S13/931—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of land vehicles
-
- 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
-
- 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
- G01S7/36—Means for anti-jamming, e.g. ECCM, i.e. electronic counter-counter measures
-
- 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
-
- 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
- G01S7/411—Identification of targets based on measurements of radar reflectivity
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- General Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention discloses a method for detecting a crossing target based on MIMO millimeter wave radar, which is applied to the field of target detection in an obstacle environment and aims at solving the problems that the whole process detection is not performed when the crossing target is detected and the multipath interference caused by the obstacle in a scene is not considered in the prior art; the method is based on an electromagnetic propagation mechanism of traversing a non-line-of-sight region of a target behind an obstacle and a line-of-sight region in front of a radar, and an echo model is established; the method comprises the steps of firstly acquiring 3D point cloud information of a scene according to radar echo data, then utilizing a reflection boundary estimation algorithm to complete estimation of a strong reflection surface boundary in the scene based on a static point cloud with zero speed in the scene, then utilizing a ghost eliminating algorithm to eliminate ghosts in the dynamic point cloud based on a dynamic point cloud with non-zero speed in the scene and the acquired reflection boundary information, and finally realizing tracking of the traversing target according to a dynamic point cloud center after ghost elimination.
Description
Technical Field
The invention belongs to the technical field of millimeter wave radar target detection, and particularly relates to a crossing target detection technology in an obstacle environment.
Background
With the rapid development of automatic driving technology, safety problems of drivers and pedestrians are becoming more and more interesting. The correct detection of moving targets such as pedestrians can provide early warning information for automatic driving automobiles or drivers, and effective measures are taken to avoid traffic accidents. However, in urban environments or other congested environments, the target is often obscured by obstacles and conventional line-of-sight detection methods are no longer suitable. Particularly in the case of a detection scenario where a vehicle is traversing a target behind it. In this scenario, when detecting the target with the radar, the parked vehicle not only makes it difficult to detect the traversing target in the non-line-of-sight region, but also makes the traversing target in the line-of-sight region vulnerable to interference by multipath signals.
Aiming at the target detection scene, related researches are also carried out by research institutions at home and abroad. In 2012, bartsch et al perceived the target with a density image and a frequency image based on 77GHz millimeter wave radar, but did not achieve good detection results in a traversing target scene (A.Bartsch, F.Fitzek, and r.h. rasshofer, "Pedestrian recognition using automotive radar sensors," Advances in Radio Science, vol.10, pp.45-55,2012.); in 2014, M.Heuer et al realized detection tracking of crossing targets based on 24GHz millimeter wave radar using a pre-detection tracking algorithm and a particle filtering algorithm (M.Heuer, A.Al-Hamadi, A.Rain, and M. Meinecke, "Detection and tracking approach using an automotive radar to increase active pedestrian safety," in 2014IEEE Intelligent Vehicles Symposium Proceedings,2014,pp.890-893.); in 2019, a.palffy et al achieved advanced detection of post-vehicle traversing targets based on fused millimeter wave radar and lidar sensors (A.Palffy, J.F.P.Kooij, and d.m. gavrila, "Occlusion aware sensor fusion for early crossing pedestrian detection," in 2019IEEE Intelligent Vehicles Symposium (IV), 2019, pp.1768-1774.).
However, none of the above related studies focused on the situation after punching out the non-line-of-sight region across the target. Multipath signals generated by obstructions may cause ghosting when the object is flushed from the non-line-of-sight region into the line-of-sight region. The ghost image may interfere with the detection of the real target, so that errors are generated in the detection and tracking results of the real target.
Disclosure of Invention
In order to solve the technical problems, the invention provides a method for detecting a crossing target based on a MIMO millimeter wave radar, which can realize effective detection and tracking of the whole movement process of the crossing target from a non-line-of-sight area to a line-of-sight area in an obstacle environment.
The invention adopts the technical scheme that: a crossing target detection method based on MIMO millimeter wave radar, the application scene includes: MIMO millimeter wave radar, parked vehicle, moving object moving from non-line-of-sight area in front of parked vehicle to line-of-sight area in front of radar;
the detection method comprises the following steps:
s1, transmitting a linear frequency modulation signal into an application scene by using a MIMO millimeter wave radar, and receiving a return wave signal;
s2, preprocessing echo signals to obtain 3D point cloud information of a scene;
s3, based on a static point cloud with zero speed in a scene, estimating a strong reflecting surface boundary in the scene by using a reflecting boundary estimation algorithm, and extracting line segment parameter information describing the reflecting surface boundary;
s4, eliminating ghosts in the dynamic point cloud by using a ghosting elimination algorithm based on the dynamic point cloud with non-zero speed in the scene and the line segment parameter information of the boundary of the reflecting surface obtained in the S3;
s5, the dynamic point cloud center after removing the ghost images in the S4 is tracked based on a nearest neighbor data association algorithm and a Kalman filtering tracking algorithm.
The invention has the beneficial effects that: the method can realize the accurate detection and tracking of the whole movement process of the blocked crossing target from the non-line-of-sight area to the line-of-sight area. The static point cloud is processed, so that scene information can be effectively obtained to assist in eliminating ghost interference signals in the dynamic point cloud; and combining scene information extracted from the static point cloud, and utilizing a ghost elimination algorithm, the interference of the ghost can be effectively eliminated. The actual measurement result shows that the method can achieve effective detection and tracking results on the crossing target after the vehicle is stopped.
Drawings
FIG. 1 is a schematic illustration of electromagnetic propagation across a target detection scene;
wherein, (a) is a top view of the target in the non-line-of-sight area, (b) is a top view of the target in the line-of-sight area, (c) is a 3D map of the target in the non-line-of-sight area, and (D) is a 3D map of the target in the line-of-sight area;
FIG. 2 is a flow chart of the proposed method of traversing object detection;
FIG. 3 is a flowchart of a reflection surface estimation algorithm;
FIG. 4 is a schematic view of the spatial neighborhood of the radar point cloud distribution features and the improved DBSCAN algorithm of the side vehicle surface;
wherein (a) is a radar point cloud distribution characteristic diagram of the side vehicle surface, and (b) is a space neighborhood schematic diagram for improving a DBSCAN algorithm;
FIG. 5 is a flow chart of a ghost rejection algorithm;
FIG. 6 is a diagram of the actual measurement scenario and the actual measurement data processing result;
wherein, (a) is an experimental scene graph, and (b) is a measured data processing result graph.
Detailed Description
The present invention will be further explained below with reference to the drawings in order to facilitate understanding of technical contents of the present invention to those skilled in the art.
The invention provides a crossing target detection method based on MIMO millimeter wave radarThe method, the detection scene is shown in fig. 1. In the scenario where the MIMO millimeter wave radar is located at point O, a vehicle is parked in front of the radar, and a target traverses from the front of the parked vehicle to the front of the radar, i.e., traverses from the non-line-of-sight region shown in fig. 1 (a) to the line-of-sight region shown in fig. 1 (b). The radar continuously emits electromagnetic waves into the scene to detect traversing objects in the scene. When the target is located in the non-line-of-sight region, as shown in FIG. 1 (a), the electromagnetic propagation path of the detected target mainly has a ground reflection path, and the side-vehicle diffraction path, without considering the combined path, the target detection path from the radar to the target and back to the radar mainly has a ground reflection path P as shown in FIG. 1 (c) 1 O-C-B-C-O and side car diffraction path P 2 O- & gt, A- & gt, B- & gt, A- & gt, O; (since the wavelength of the millimeter wave radar is short, the side diffraction echo is weak, and the side diffraction path P is ignored) 2 ) The method comprises the steps of carrying out a first treatment on the surface of the When the target is located in the line of sight area, as shown in fig. 1 (b), the direct-view path can realize the detection of the target, but the side-vehicle reflection path also has strong reflection echo, and there is strong interference on the detection of the target (specifically, the multipath signal will cause ghost in the subsequent radar signal processing, thereby affecting the detection and tracking of the real target). In this case, the electromagnetic propagation path from the radar to the target and back to the radar mainly includes a direct-view path P as shown in fig. 1 (d) without considering the combined path 3 O-F-O and by-pass reflection path P 4 :O→E→F→E→O。
As shown in fig. 2, the process flow of the method of the present invention comprises the steps of:
Assume that the parameters of the chirp signal transmitted by the radar are respectively: carrier frequency f 0 Frequency modulation slope mu, pulse width T, signal bandwidth B, signal amplitude A 0 . Then, the radar transmit signal is:
s(t)＝A 0 exp(j2πf 0 t+jπμt 2 )u(t)
wherein, the liquid crystal display device comprises a liquid crystal display device,
t is the time period of time, and the time period of the time period is,u (t) is a rectangular function:
when the transmitted signal is a large bandwidth signal, the extended targets such as human body and obstructions within the scene may be considered a collection of many scattering points. The echo model based on time delay can thus be expressed as:
wherein s (t-tau) represents a signal after s (t) time delay tau, l represents a first path of the detection target, i represents an i-th scattering point of a certain expansion target in the scene,
respectively representing the backscattering coefficient and the echo delay of the object,/->
And respectively representing the backscattering coefficient and the echo time delay of the parked vehicle, and zeta is noise. Thus, in the above equation, the first portion represents an echo signal traversing the target, and the second portion represents an echo signal of the parked vehicle.
Based on the electromagnetic propagation analysis described above, the radar passes through the path P when traversing a non-line-of-sight region where the target is located in front of the parked vehicle 1 Detecting the target, at this time, l is { l } 1 Echo model is rewritten as:
when the target is located in the line-of-sight area in front of the radar, the propagation path of the electromagnetic wave may be P 3 ，P 4 At this time, l ε { l 3 ,l 4 Echo model is rewritten as:
after the echo signal is subjected to the mixing filtering operation, an intermediate frequency signal is obtained and is denoted as y (t).
Further, the virtual array of the MIMO radar system is set to be Q uniform linear arrays. Then, after the intermediate frequency signals of all the receiving antennas are digitally sampled, an echo data matrix can be obtained:
where n represents the sampling point, P represents the P-th chirp signal within a frame, (p=1.,), P), Q represents the Q-th receive antenna (q=1,., Q). Further, y p,q (n) represents an echo sample signal of the p-th chirp signal of the q-th antenna in one frame.
The data preprocessing mainly comprises 2D-FFT, constant false alarm detection, arrival angle estimation and point cloud generation.
Firstly, performing fast Fourier transform along a slow time dimension to complete pulse compression, then performing fast Fourier transform along the fast time dimension to obtain a distance-Doppler spectrogram, then adopting a constant false alarm detection algorithm to extract strong scattering points passing through a detection threshold in a scene, and further solving distance and speed information of the scattering points passing through the detection threshold.
And estimating an arrival angle of the point target obtained after the constant false alarm detection along the antenna dimension, and obtaining the angle of the scattering point under a radar coordinate system. Based on the obtained distance and angle information, the two-dimensional space position of the scattering point is calculated through coordinate conversion, and finally, the radar 3D point cloud information can be obtained by combining the speed information of the scattering point.
And 3, based on the static point cloud with zero speed in the scene, estimating the boundary of the strong reflecting surface in the scene and extracting relevant parameters by using a reflecting boundary estimation algorithm.
As shown in fig. 3, the method specifically includes the following steps:
firstly, the improved DBSCAN algorithm is utilized to carry out clustering processing on the static point cloud. When electromagnetic waves propagate to the surface of a side car at a certain inclination angle, only discontinuous parts of the car, such as the corners of the car, the vicinity of the door handle part and the vicinity of the rearview mirror part, can strongly reflect electromagnetic signals, and fig. 4 (a) shows a set of point cloud distribution characteristics of the side face of the side car in the actual measurement environment; although the point cloud distribution characterizing the sides of the vehicle is discontinuous, they have strong linear distribution characteristics in spatial locations, so that the traditional DBSCAN (Density-Based Spatial Clustering of Applications with Noise, density-based clustering algorithm) algorithm is improved to adapt to the point cloud distribution characteristics of the vehicle. Compared with the traditional DBSCAN algorithm, except for the density parameter epsilon and the radius parameter Minpts, the improved DBSCAN algorithm is added with rho, theta and delta theta parameters to adjust the spatial neighborhood of the clusters so as to adapt to the distribution characteristics of static point clouds generated by the side vehicle reflection boundary; wherein ρ is the desired sector length, θ is the desired angle between the sector normal and the x-axis forward direction, and Δθ is the deviation of the desired angle; FIG. 4 (b) shows a spatial neighborhood of an improved DBSCAN algorithm; in the specific implementation process of the algorithm, only the point cloud in the spatial neighborhood formed by the three parameters is needed to be incorporated in the spatial neighborhood of the traditional DBSCAN algorithm.
After clustering the static point cloud, performing linear parameter fitting processing on each cluster obtained by clustering by using a Hough transformation algorithm; suppose that after the point cloud in a certain cluster is processed by the Hough transform algorithm, the extracted polar coordinate parameter information is that
The straight line corresponding to the parameter information is y=kx+b in the Cartesian coordinate system (in the two-dimensional space coordinate system), and the straight line parameters (k, b) and +.>
The relation of (2) is:
then, the coordinates of two ports of a fitting straight line are determined by taking the cloud coordinates of boundary points of each cluster as references, and the length of a line segment is calculated; and finally, setting a length reference standard based on the length information of the real automobile, and removing invalid line segments with overlong or overlong length.
In the above, the line segment information for characterizing the boundary of the reflecting surface may be obtained, and specifically includes a straight-line truncated parameter (k, b), (where k represents the slope of a straight line, b represents the intersection of the straight line and the y-axis), the length of the line segment, and the port coordinate information.
And 4, eliminating ghosts in the dynamic point cloud by using a ghosting eliminating algorithm based on the dynamic point cloud with non-zero speed in the scene and the scene information obtained in the step 3.
As shown in fig. 5, the method specifically includes: firstly, clustering dynamic point clouds by using a traditional DBSCAN algorithm, extracting the central coordinate of each cluster after clustering to be used for representing the cluster, and recording the central coordinate of a certain cluster as (x c ,y c )：
Wherein N is the number of point clouds in the cluster, (x) j ,y j ) Is the j-th point cloud in the cluster. And subsequently, further processing the obtained cluster center.
From the geometrical features of the ghost distribution, the position of the ghost and the real target caused by the side-car reflection (double-pass primary reflection) is mirror symmetric about the reflection boundary. Therefore, after plane distribution of the reflection boundary in the scene is obtained, based on the position information of the line segment representing the reflection boundary, a cluster center point pair which is in mirror symmetry with respect to the line segment can be searched, namely, the association matching of the cluster centers is realized.
Assuming that the continuous reflection boundary in the scene is obtained by step 3 as y=kx+b, then the center point (x c ,y c ) The mirror position about the boundary is (x mir ,y mir )：
Because the radar has measurement errors and the like, the ghost is not necessarily positioned at the mirror image position of the real target relative to the reflecting surface, and therefore, the midpoint (x 0 ,y 0 ) Point of mirror symmetry about a certain reflection boundary (x mir ,y mir ) Thereafter, a mirror image is provided at the position (x mir ,y mir ) Is a search area with a radius Δr, and the cluster center falling within the area is aligned with the original cluster center point (x 0 ,y 0 ) And matching, wherein DeltaR is an empirical value, and can be selected to be 0.5m. And traversing all cluster centers to finish the cluster center matching operation.
Carrying out ghost identification and elimination on the matched cluster centers, wherein the identification criteria are as follows: in a set of matching point pairs, cluster centers relatively far from the radar are marked as ghosts. After the ghost identification process, eliminating the cluster center marked as the ghost; unmarked and unmatched cluster centers remain.
And 5, based on the cluster center after removing the ghost images (the clustered dynamic point cloud center) in the step 4, tracking the traversing target by using a nearest neighbor data association algorithm and a Kalman filtering tracking algorithm.
The state quantity of the kalman filter is composed of the lateral position, the longitudinal position, the lateral velocity and the longitudinal velocity of the object, and is represented by x:
x＝[x,y,v x ,v y ] T
wherein [ (S)] T Representing the transpose.
Considering that the target motion is a uniform motion model, the state transition matrix is:
the measurement matrix is:
further, the Kalman filter formula is used for prediction and updating, and the Kalman equation set is as follows:
wherein t represents the current time, t-1 represents the last time, K represents Kalman filtering gain, P represents a state estimation error covariance matrix, z represents a measured value vector, H represents a measurement matrix, Q represents a process noise covariance matrix, and R represents a measurement noise covariance matrix;
the predicted value of the state quantity at the t moment and the predicted value of the state estimation error covariance matrix are respectively expressed, and x is t 、P t 、K t The updated value of the state quantity at the time t, the updated value of the state estimation error covariance matrix and the updated value of the Kalman filtering gain are respectively represented.
Fig. 6 (a) shows an experimental scene adopted, fig. 6 (b) shows a processing result of the method of the present invention, and in fig. 6 (b), a line segment of a reflection boundary on the right side of an automobile is an estimation result of an algorithm, and the remaining three boundaries are complement results. It can be seen that the boundary information of the reflecting surface of the parked vehicle in front of the radar can be effectively estimated; the ghosts caused by the side car reflected signals are effectively removed, and the tracking interference of the ghosts on the real target is effectively restrained; the target can be correctly detected and tracked throughout its movement from the non-line-of-sight region in front of the parked vehicle to the line-of-sight region in front of the radar. Actual measurement experiments prove the feasibility and effectiveness of the invention.
Those of ordinary skill in the art will recognize that the embodiments described herein are for the purpose of aiding the reader in understanding the principles of the present invention and should be understood that the scope of the invention is not limited to such specific statements and embodiments. Various modifications and variations of the present invention will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.
Claims (4)
1. The method for detecting the crossing target based on the MIMO millimeter wave radar is characterized in that the application scene comprises the following steps: MIMO millimeter wave radar, parked vehicle, moving object moving from non-line-of-sight area in front of parked vehicle to line-of-sight area in front of radar;
the detection method comprises the following steps:
s1, transmitting a linear frequency modulation signal into an application scene by using a MIMO millimeter wave radar, and receiving a return wave signal;
s2, preprocessing echo signals to obtain 3D point cloud information of a scene;
s3, based on a static point cloud with zero speed in a scene, estimating a strong reflecting surface boundary in the scene by using a reflecting boundary estimation algorithm, and extracting line segment parameter information describing the reflecting surface boundary;
s4, eliminating ghosts in the dynamic point cloud by using a ghosting eliminating algorithm based on the dynamic point cloud with non-zero speed in the scene and the line segment parameter information of the boundary of the reflecting surface obtained in the S3;
s5, the dynamic point cloud center after removing the ghost images in the S4 is tracked based on a nearest neighbor data association algorithm and a Kalman filtering tracking algorithm.
2. The method for detecting a crossing target based on the MIMO millimeter wave radar according to claim 1, wherein the step S3 specifically comprises the following sub-steps:
s31, clustering the static point cloud by using an improved DBSCAN algorithm;
s32, performing linear parameter fitting on each cluster obtained by clustering in the step S31 by using a Hough transform algorithm;
s33, determining coordinates of two ports of a fitting straight line by taking the cloud coordinates of boundary points of each cluster as a reference, and calculating the length of a line segment;
s34, setting a length reference standard based on the length information of the real automobile, and eliminating invalid line segments with overlong or excessively short lengths; thereby acquiring line segment information for characterizing the boundary of the reflecting surface.
3. The method for detecting a crossing target based on the MIMO millimeter wave radar according to claim 2, wherein the step S31 is specifically to incorporate a point cloud in a spatial neighborhood composed of three parameters of ρ, θ and Δθ in the spatial neighborhood of the conventional DBSCAN algorithm; where ρ is the desired sector length, θ is the angle between the desired sector normal and the positive x-axis, and Δθ is the deviation of the desired angle.
4. A method for detecting a crossing target based on a MIMO millimeter wave radar according to claim 3, wherein step S4 specifically comprises the following sub-steps:
s41, clustering dynamic point clouds by using a traditional DBSCAN algorithm, and extracting the center coordinates of each cluster after clustering to represent the cluster;
s42, searching a cluster center point pair which is mirror symmetry with respect to a line segment based on the position information of the line segment representing the reflection boundary, so as to realize the matching of cluster centers;
s43, carrying out ghost identification and elimination on the matched cluster centers.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310339838.2A CN116299473A (en) | 2023-03-31 | 2023-03-31 | Method for detecting crossing target based on MIMO millimeter wave radar |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310339838.2A CN116299473A (en) | 2023-03-31 | 2023-03-31 | Method for detecting crossing target based on MIMO millimeter wave radar |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116299473A true CN116299473A (en) | 2023-06-23 |
Family
ID=86828739
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310339838.2A Pending CN116299473A (en) | 2023-03-31 | 2023-03-31 | Method for detecting crossing target based on MIMO millimeter wave radar |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116299473A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117077407A (en) * | 2023-08-18 | 2023-11-17 | 北京华如科技股份有限公司 | Target detection method and device |
-
2023
- 2023-03-31 CN CN202310339838.2A patent/CN116299473A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117077407A (en) * | 2023-08-18 | 2023-11-17 | 北京华如科技股份有限公司 | Target detection method and device |
| CN117077407B (en) * | 2023-08-18 | 2024-01-30 | 北京华如科技股份有限公司 | Target detection method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11630197B2 (en) | Determining a motion state of a target object | |
| CN109634282B (en) | Autonomous vehicle, method and apparatus | |
| US11768286B2 (en) | Method of determining the yaw rate of a target vehicle | |
| US11508122B2 (en) | Bounding box estimation and object detection | |
| US11475678B2 (en) | Lane marker detection and lane instance recognition | |
| US9255988B2 (en) | Object fusion system of multiple radar imaging sensors | |
| CN112098990B (en) | Method for detecting and tracking medium and high speed vehicle by vehicle-mounted high-resolution millimeter wave radar | |
| CN110379178B (en) | Intelligent unmanned automobile parking method based on millimeter wave radar imaging | |
| US20090292468A1 (en) | Collision avoidance method and system using stereo vision and radar sensor fusion | |
| CN105197108B (en) | A kind of multiple target direction-finding system and its method based on automobile assistant driving system | |
| CN113850102B (en) | Vehicle-mounted vision detection method and system based on millimeter wave radar assistance | |
| Cui et al. | 3D detection and tracking for on-road vehicles with a monovision camera and dual low-cost 4D mmWave radars | |
| CN116299473A (en) | Method for detecting crossing target based on MIMO millimeter wave radar | |
| Blanc et al. | Track to track fusion method applied to road obstacle detection | |
| Gao et al. | DC-Loc: Accurate automotive radar based metric localization with explicit doppler compensation | |
| Meis et al. | Radar image acquisition and interpretation for automotive applications | |
| US20240159892A1 (en) | Non-line-of-sight radar apparatus | |
| Segawa et al. | Preceding vehicle detection using stereo images and non-scanning millimeter-wave radar | |
| Domhof et al. | Multi-sensor object tracking performance limits by the cramer-rao lower bound | |
| Hoffmann et al. | Filter-based segmentation of automotive sar images | |
| Xie et al. | On-bicycle vehicle tracking at traffic intersections using inexpensive low-density lidar | |
| Shen et al. | Darting-out target detection with NLOS signals for vehicle MIMO mmWave radar | |
| CN111123260B (en) | Method for identifying state of environmental object by using millimeter wave radar and visible light camera | |
| Gao et al. | Accurate automotive radar based metric localization with explicit doppler compensation | |
| CN117250595B (en) | False alarm suppression method for vehicle-mounted millimeter wave radar metal well lid target |
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 |