A target tracking method and device based on multi-sensor fusion

Abstract

The application discloses a target tracking method and device based on multi-sensor fusion, and the method comprises the following steps: acquiring target region information to be tracked; performing registration processing on a camera, a laser radar and a millimeter wave radar; performing fusion tracking processing on the target by using the camera, the laser radar and the millimeter wave radar to obtain target position information. The application solves the problems of single system target tracking means, difficulty in realizing stable tracking of a target vehicle in an unstructured environment and how to effectively fuse multiple types of sensors.

Description

Technical Field

[0001] This invention relates to the fields of autonomous driving and target tracking, and specifically to a target tracking method and apparatus based on multi-sensor fusion. Background Technology

[0002] In autonomous driving scenarios, it is often necessary to track target vehicles or typical moving targets in space. Target tracking technology in structured scenarios is relatively mature. However, complex unstructured environments, such as those with heavy dust or dense fog caused by a moving vehicle ahead, pose significant challenges to vehicle target tracking.

[0003] For optical methods such as cameras, existing target tracking methods can achieve good performance on public datasets. However, in unstructured scenes, complex road environments lead to a lack of image stability. For example, rough terrain, changing lighting, and image jitter caused by partial occlusion by surrounding obstacles can severely affect tracking results. Furthermore, due to the limited field of view of the camera, the target can easily lose its view when the guiding vehicle turns, resulting in its loss of tracking. Simultaneously, perspective transformation is needed to convert the target's position in the image frame to the vehicle's local frame for motion planning. Inevitably, there are transformation biases during this process, making it difficult to provide the follower with an accurate and stable position. The guiding vehicle is the target vehicle to be tracked.

[0004] For 3D LiDAR, compared to cameras, the position of the guiding vehicle can be obtained directly from the point cloud space without perspective transformation, and the horizontal field of view is 360° without blind spots. Deep learning techniques, widely used in image processing, are also applied to 3D target detection, primarily addressing tracking problems in urban environments. However, in unstructured environments with dusty, bumpy roads, the performance of these methods deteriorates sharply due to the poor anti-interference capability of 3D LiDAR against dust. Furthermore, as distance increases, the point cloud becomes sparser, causing the detection accuracy to decrease rapidly with increasing distance.

[0005] Millimeter-wave radar exhibits strong resistance to interference from dust, rain, fog, and snow. Currently, it is widely used in automotive safety and adaptive cruise control. However, its drawback is its inability to distinguish between stationary vehicles and surrounding obstacles, which is crucial for target tracking initialization. When the lead vehicle starts from a stationary position, it requires other sensors to provide initial target tracking information. Furthermore, due to limited field of view during turns, it is also prone to losing track of the target.

[0006] In summary, single-system target tracking methods are insufficient for stable tracking of target vehicles in unstructured environments. How to effectively fuse multiple types of sensors is an urgent problem to be solved. Summary of the Invention

[0007] This invention primarily addresses the challenge of achieving stable tracking of target vehicles in unstructured environments using single-system target tracking methods, and the problem of effectively fusing multiple types of sensors. This invention discloses a target tracking method and apparatus based on multi-sensor fusion.

[0008] In a first aspect, this application discloses a target tracking method based on multi-sensor fusion, comprising:

[0009] S1, Obtain information about the target area to be tracked;

[0010] S2 performs registration processing on the target-assisted tracking position information and the millimeter-wave radar;

[0011] S3, using the camera, lidar and millimeter-wave radar to perform fusion tracking processing on the target to obtain the target position information;

[0012] The registration process for the camera, lidar, and millimeter-wave radar includes:

[0013] S21, perform system error registration processing on the camera, lidar and millimeter-wave radar;

[0014] S22, perform time registration and spatial registration processing on the camera, lidar and millimeter-wave radar;

[0015] S23, initialize spatial gate information; the spatial gate is a cube in space; the spatial gate information includes x-axis distance threshold X1, y-axis distance threshold Y1 and z-axis distance threshold Z1 in a three-dimensional rectangular coordinate system.

[0016] The process of using the camera, lidar, and millimeter-wave radar to perform fusion tracking of the target to obtain target location information includes:

[0017] S31, use a camera to acquire and process the image of the target area to be tracked to obtain the target area image information; use a lidar to detect and process the target area information to be tracked to obtain the target area point cloud data.

[0018] S32, perform joint monitoring and processing on the target area image information and target area point cloud data to obtain target auxiliary tracking position information;

[0019] S33, using the target auxiliary tracking position information as guidance information, the millimeter-wave radar is used to perform tracking filtering on the target to obtain target position sequence information; the target position sequence information includes several target positions;

[0020] S34, perform validity judgment on the target location sequence information to obtain a first judgment result; if the first judgment result is valid, use the target location sequence information as the target location information; if the first judgment result is invalid, execute S32.

[0021] The joint monitoring and processing of the target area image information and target area point cloud data to obtain target-assisted tracking location information includes:

[0022] S321, Perform image target detection processing on the target region image information to obtain the first target location region information;

[0023] S322, Perform target point cloud matching processing on the target area point cloud data to obtain the second target location area information;

[0024] S323, Spatial gate detection is performed on the first target location region information and the second target location region information to obtain target auxiliary tracking location information;

[0025] S324, perform consistency judgment on the target auxiliary tracking position information obtained at several time points to obtain a second judgment result; if the second judgment result is valid, execute S33; if the second judgment result is invalid, execute S31.

[0026] The step of performing spatial gate detection on the first target location region information and the second target location region information to obtain target-assisted tracking location information includes:

[0027] S3231, Obtain the overlapping area information of the first target location area information and the second target location area information; the overlapping area is a three-dimensional cube; the overlapping area information is the side length X2 in the x-axis direction, the side length Y2 in the y-axis direction and the side length Z2 in the z-axis direction of the three-dimensional cube;

[0028] S3232, perform joint discrimination processing on the overlapping region information and spatial gate information to obtain a third discrimination result;

[0029] If the third discrimination result is satisfied, the first target location area information and the second target location area information are fused and weighted to obtain the target auxiliary tracking location information;

[0030] If the third discrimination result is not satisfied, it is confirmed that the target region image information and the target region point cloud data do not contain the same target, and S321 is executed;

[0031] The expression for the joint discriminant processing is:

[0032] |X1-X2|≤a1,

[0033] |Y1–Y2|≤a2,

[0034] |Z1–Z2|≤a3,

[0035] |(X2-X1)(Y2-Y1)|≤a4,

[0036] |(Z2-Z1)(Y2-Y1)|≤a5,

[0037] Where a1 to a5 are all preset discrimination thresholds; when the overlapping region information satisfies the above five inequalities, the third discrimination result is confirmed to be satisfied; otherwise, the third discrimination result is confirmed to be unsatisfied.

[0038] The expression for the fusion weighting process is:

[0039]

[0040] Wherein, [x1,y1,z1] and [x2,y2,z2] are the center point coordinates of the first target location region information and the center point coordinates of the second target location region information, respectively. For target-aided tracking location information, α1 is the ratio of the area of ​​the overlapping region information to the area of ​​the first target location region information, and α2 is the ratio of the area of ​​the overlapping region information to the area of ​​the second target location region information.

[0041] The validity determination includes:

[0042] The tracking filtering process is implemented using Kalman filtering;

[0043] The measurement matrix of each target position in the target position sequence information is obtained during the corresponding Kalman filtering process;

[0044] The mean matrix is ​​obtained by averaging all the measurement matrices.

[0045] The difference matrix is ​​obtained by subtracting each measurement matrix from the mean matrix.

[0046] The mean difference matrix is ​​obtained by taking the mean of the difference matrices of all measurement matrices.

[0047] The difference mean matrix is ​​decomposed to obtain the feature matrix;

[0048] Extract the diagonal elements of the feature matrix to obtain the feature vector;

[0049] Linear fitting is performed on the elements and element index values ​​of the feature vector to obtain the difference discrimination polynomial;

[0050] The mean of all row vectors in the difference mean matrix is ​​calculated;

[0051] Substitute the row vector index corresponding to the maximum value among the mean values ​​of all row vectors into the difference discrimination polynomial to obtain the first discrimination value;

[0052] Determine whether the first discrimination value is greater than a preset trajectory threshold. If it is not greater, the first discrimination result is confirmed as valid; if it is greater, the first discrimination result is confirmed as invalid.

[0053] The consistency determination includes:

[0054] Acquire target-aided tracking position information of the same target obtained at several time points;

[0055] The target auxiliary tracking position information of the same target obtained at several time points is represented as an auxiliary tracking position matrix; the row vector of the auxiliary tracking position matrix is ​​the target auxiliary tracking position information obtained at one time point.

[0056] The auxiliary tracking position matrix is ​​processed to calculate a consistency value, and a consistency value is obtained.

[0057] The expression for calculating the consistency value is:

[0058]

[0059] Where, q ki The elements in the k-th row and i-th column of the auxiliary tracking position matrix are defined, where m and n are the row and column dimensions of the auxiliary tracking position matrix, respectively, and h is a consistency value.

[0060] Determine whether the consistency value is greater than the set consistency threshold. If it is not greater, the second discrimination result is confirmed as valid; if it is greater, the second discrimination result is confirmed as invalid.

[0061] A second aspect of the present invention discloses a target tracking device based on multi-sensor fusion, the device comprising:

[0062] Memory containing executable program code;

[0063] A processor coupled to the memory;

[0064] The processor calls the executable program code stored in the memory to execute the target tracking method based on multi-sensor fusion.

[0065] In a third aspect of this invention, a computer-storable medium is disclosed, wherein the computer-storable medium stores computer instructions, which, when invoked by a computer, are used to execute the target tracking method based on multi-sensor fusion.

[0066] In a fourth aspect of this invention, an information data processing terminal is disclosed, which is used to implement the target tracking method based on multi-sensor fusion.

[0067] The beneficial effects of this invention are as follows:

[0068] This invention addresses the challenge of achieving stable tracking of target vehicles in unstructured environments using a single target tracking method, and how to effectively fuse multiple types of sensors.

[0069] This invention first utilizes a camera and lidar to achieve preliminary target tracking, thereby obtaining target auxiliary tracking position information and overcoming the tracking problem of millimeter-wave radar when the guide vehicle is statically activated. After obtaining the target auxiliary tracking position information, a consistency discrimination model is established to perform consistency discrimination on the target auxiliary tracking position information obtained at several time points, ensuring the accuracy of the preliminary tracking results.

[0070] After completing the initial tracking, this invention uses the obtained target auxiliary tracking position information as the initial value and uses millimeter-wave radar to achieve high-precision tracking of the target. To ensure the effectiveness of the tracking results, this invention introduces an effectiveness discrimination algorithm, selects the system's measurement matrix as the effectiveness variable, and ensures the accuracy of the fused tracking results by judging the effectiveness of the measurement matrix. Attached Figure Description

[0071] Figure 1 This is a flowchart illustrating the implementation of the method of the present invention. Detailed Implementation

[0072] To better understand the content of this invention, an embodiment is provided here.

[0073] Figure 1 This is a flowchart illustrating the implementation of the method of the present invention.

[0074] To address the challenge of achieving stable tracking of target vehicles in unstructured environments using a single-system target tracking method, and to solve the problem of effectively fusing multiple types of sensors, this invention discloses a target tracking method and apparatus based on multi-sensor fusion.

[0075] In a first aspect, this application discloses a target tracking method based on multi-sensor fusion, comprising:

[0076] S1, Obtain information about the target area to be tracked;

[0077] S2 performs registration processing on the camera, lidar, and millimeter-wave radar;

[0078] S3, using the camera, lidar and millimeter-wave radar to perform fusion tracking processing on the target to obtain the target position information;

[0079] The registration process for the camera, lidar, and millimeter-wave radar includes:

[0080] S21, perform system error registration processing on the camera, lidar and millimeter-wave radar;

[0081] S22, perform time registration and spatial registration processing on the camera, lidar and millimeter-wave radar;

[0082] S23, Initialize spatial gate information; the spatial gate is a cube in space; the spatial gate information includes x-axis distance threshold X1, y-axis distance threshold Y1 and z-axis distance threshold Z1 in a three-dimensional rectangular coordinate system;

[0083] The process of using the camera, lidar, and millimeter-wave radar to perform fusion tracking of the target to obtain target location information includes:

[0084] S31, use a camera to acquire and process the image of the target area to be tracked to obtain the target area image information; use a lidar to detect and process the target area information to be tracked to obtain the target area point cloud data.

[0085] S32, perform joint monitoring and processing on the target area image information and target area point cloud data to obtain target auxiliary tracking position information;

[0086] S33, using the target auxiliary tracking position information as guidance information, the millimeter-wave radar is used to perform tracking filtering on the target to obtain target position sequence information; the target position sequence information includes several target positions;

[0087] S34, perform validity determination on the target location sequence information to obtain a first determination result; if the first determination result is valid, confirm that the target location sequence information is target location information; if the first determination result is invalid, execute S32;

[0088] The joint monitoring and processing of the target area image information and target area point cloud data to obtain target-assisted tracking location information includes:

[0089] S321, Perform image target detection processing on the target region image information to obtain the first target location region information;

[0090] S322, Perform target point cloud matching processing on the target area point cloud data to obtain the second target location area information;

[0091] S323, Spatial gate detection is performed on the first target location region information and the second target location region information to obtain target auxiliary tracking location information;

[0092] S324, perform consistency judgment on the target auxiliary tracking position information obtained at several time points to obtain a second judgment result; if the second judgment result is valid, execute S33; if the second judgment result is invalid, execute S31;

[0093] The step of performing spatial gate detection on the first target location region information and the second target location region information to obtain target-assisted tracking location information includes:

[0094] S3231, Obtain the overlapping area information of the first target location area information and the second target location area information; the overlapping area is a three-dimensional cube; the overlapping area information is the side length X2 in the x-axis direction, the side length Y2 in the y-axis direction and the side length Z2 in the z-axis direction of the three-dimensional cube;

[0095] S3232, perform joint discrimination processing on the overlapping region information and spatial gate information to obtain a third discrimination result;

[0096] If the third discrimination result is satisfied, the first target location area information and the second target location area information are fused and weighted to obtain the target auxiliary tracking location information;

[0097] If the third discrimination result is not satisfied, it is confirmed that the target region image information and the target region point cloud data do not contain the same target, and S321 is executed;

[0098] The expression for the joint discriminant processing is:

[0099] |X1-X2|≤a1,

[0100] |Y1–Y2|≤a2,

[0101] |Z1–Z2|≤a3,

[0102] |(X2-X1)(Y2-Y1)|≤a4,

[0103] |(Z2-Z1)(Y2-Y1)|≤a5,

[0104] Wherein, a1 to a5 are preset discrimination thresholds, and their values ​​can be 0.6, 0.5, 0.2, 0.4, and 0.8. When the overlapping region information satisfies all five inequalities mentioned above, the third discrimination result is confirmed to be satisfied; otherwise, the third discrimination result is confirmed to be unsatisfied.

[0105] The expression for the fusion weighting process is:

[0106]

[0107] Wherein, [x1,y1,z1] and [x2,y2,z2] are the center point coordinates of the first target location region information and the center point coordinates of the second target location region information, respectively. For the target auxiliary tracking position information, α1 is the ratio of the area of ​​the overlapping region information to the area of ​​the first target position region information, and α2 is the ratio of the area of ​​the overlapping region information to the area of ​​the second target position region information;

[0108] The validity determination includes:

[0109] The tracking filtering process is implemented using Kalman filtering;

[0110] The measurement matrix of each target position in the target position sequence information is obtained during the corresponding Kalman filtering process;

[0111] The mean matrix is ​​obtained by averaging all the measurement matrices.

[0112] The difference matrix is ​​obtained by subtracting each measurement matrix from the mean matrix.

[0113] The mean difference matrix is ​​obtained by taking the mean of the difference matrices of all measurement matrices.

[0114] The difference mean matrix is ​​decomposed to obtain the feature matrix;

[0115] Extract the diagonal elements of the feature matrix to obtain the feature vector;

[0116] Linear fitting is performed on the elements and element index values ​​of the feature vector to obtain the difference discrimination polynomial;

[0117] The mean of all row vectors in the difference mean matrix is ​​calculated;

[0118] Substitute the row vector index corresponding to the maximum value among the mean values ​​of all row vectors into the difference discrimination polynomial to obtain the first discrimination value;

[0119] Determine whether the first discrimination value is greater than a preset trajectory threshold. If it is not greater, the first discrimination result is confirmed as valid; if it is greater, the first discrimination result is confirmed as invalid.

[0120] The consistency determination includes:

[0121] Acquire target-aided tracking position information of the same target obtained at several time points;

[0122] The target auxiliary tracking position information of the same target obtained at several time points is represented as an auxiliary tracking position matrix; the row vector of the auxiliary tracking position matrix is ​​the target auxiliary tracking position information obtained at one time point.

[0123] The auxiliary tracking position matrix is ​​processed to calculate a consistency value, and a consistency value is obtained.

[0124] The expression for calculating the consistency value is:

[0125]

[0126] Where, q ki The elements in the k-th row and i-th column of the auxiliary tracking position matrix are defined, where m and n are the row and column dimensions of the auxiliary tracking position matrix, respectively, and h is a consistency value.

[0127] Determine whether the consistency value is greater than a set consistency threshold. If it is not greater, the second discrimination result is confirmed as valid; if it is greater, the second discrimination result is confirmed as invalid.

[0128] The method of using target-assisted tracking position information as guidance information and using millimeter-wave radar to track the target involves using the target-assisted tracking position information as the initial position information of the target in the Kalman filter.

[0129] The linear fitting process involves using the characteristic vector element index value Ix as the known independent variable and the characteristic vector element value as the known dependent variable. The curve to be approximated is constructed using the known independent and dependent variables, and the curve to be approximated is fitted using the function approximation method to obtain the difference discrimination polynomial f(Ix).

[0130] The curve fitting of the curve to be approximated using the function approximation method can employ the best uniform linear approximation method. The difference discriminant polynomial f(Ix) is expressed as:

[0131] f(Ix)=α P1 (Ix) P1 +α P1-1 (Ix) P1-1 +…+α2(Ix) 2 +α1(Ix)+α0,

[0132] Where P1 is the order of the difference discriminant polynomial f(Ix), α0, α1, α2, ..., α P1 The coefficients of the difference discrimination polynomial f(Ix);

[0133] The feature vector is represented as I a I a =[λ1,λ2,…,λ N1N1 is the number of elements contained in the feature vector;

[0134] The decomposition process is calculated using the following expression:

[0135] Y = UAV,

[0136] Where U is the left decomposition matrix, A is the characteristic matrix, V is the right decomposition matrix, U and V are both orthogonal matrices, A is a diagonal matrix, and Y is the difference mean matrix;

[0137] The elements in the i-th row and j-th column of the mean matrix obtained by the mean calculation process are the mean of the elements in the i-th row and j-th column of all measurement matrices.

[0138] The system error registration process involves acquiring the measurement system errors of the camera, lidar, and millimeter-wave radar respectively, and using the measurement system errors of each measuring device to perform error calibration on the collected data.

[0139] The time registration process is to unify the measurement data from different measuring devices onto the same time reference; the time registration process can employ methods such as extrapolation / extrapolation and Lagrange three-point interpolation.

[0140] The tracking filtering process involves using millimeter-wave radar to detect the target, obtaining the target point trajectory, and then using a Kalman filter algorithm to process the target point trajectory.

[0141] The spatial registration process involves transforming the coordinates of measurement data from different measuring devices to unify them under the same coordinate system.

[0142] The image target detection processing can be based on a template image of the target, performing template matching processing on the target region image information to obtain target location region information and target location center information; the template matching processing can be the MAD algorithm or the SSD algorithm.

[0143] The target point cloud matching process can employ point cloud feature extraction algorithms from LiDAR or deep learning algorithms for 3D point cloud target detection. Specifically, point cloud template data of the target can be acquired first, and then a global feature extraction algorithm based on three-dimensional Hough transform can be used to match the point cloud template data to obtain the second target location region information.

[0144] The target tracked in this method can be a vehicle target;

[0145] The camera, lidar, and millimeter-wave radar are all mounted on the same vehicle platform.

[0146] A second aspect of the present invention discloses a target tracking device based on multi-sensor fusion, the device comprising:

[0147] Memory containing executable program code;

[0148] A processor coupled to the memory;

[0149] The processor calls the executable program code stored in the memory to execute the target tracking method based on multi-sensor fusion.

[0150] In a third aspect of this invention, a computer-storable medium is disclosed, wherein the computer-storable medium stores computer instructions, which, when invoked by a computer, are used to execute the target tracking method based on multi-sensor fusion.

[0151] In a fourth aspect of this invention, an information data processing terminal is disclosed, which is used to implement the target tracking method based on multi-sensor fusion.

[0152] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A target tracking method based on multi-sensor fusion, characterized in that, include: S1, Obtain information about the target area to be tracked; S2 performs registration processing on the camera, lidar, and millimeter-wave radar; S3, using the camera, lidar, and millimeter-wave radar to perform fusion tracking processing on the target to obtain target position information, including: S31, use a camera to acquire and process the image of the target area to be tracked to obtain the target area image information; use a lidar to detect and process the target area information to be tracked to obtain the target area point cloud data. S32, perform joint monitoring and processing on the target area image information and target area point cloud data to obtain target auxiliary tracking position information; S33, using the target auxiliary tracking position information as guidance information, the target is tracked and filtered using millimeter-wave radar to obtain target position sequence information; the target position sequence information includes several target positions; the tracking filtering process is implemented using Kalman filtering; S34, perform validity determination on the target location sequence information to obtain a first determination result; if the first determination result is valid, use the target location sequence information as the target location information; if the first determination result is invalid, execute S32; The validity determination includes: The measurement matrix of each target position in the target position sequence information is obtained during the corresponding Kalman filtering process; The mean matrix is ​​obtained by averaging all the measurement matrices. The difference matrix is ​​obtained by subtracting each measurement matrix from the mean matrix. The mean difference matrix is ​​obtained by taking the mean of the difference matrices of all measurement matrices. The difference mean matrix is ​​decomposed to obtain the feature matrix; Extract the diagonal elements of the feature matrix to obtain the feature vector; Linear fitting is performed on the elements and element index values ​​of the feature vector to obtain the difference discrimination polynomial; The mean of all row vectors in the difference mean matrix is ​​calculated; Substitute the row vector index corresponding to the maximum value among the mean values ​​of all row vectors into the difference discrimination polynomial to obtain the first discrimination value; Determine whether the first discrimination value is greater than a preset trajectory threshold. If it is not greater, the first discrimination result is confirmed as valid; if it is greater, the first discrimination result is confirmed as invalid.

2. The target tracking method based on multi-sensor fusion as described in claim 1, characterized in that, The registration process for the camera, lidar, and millimeter-wave radar includes: S21, perform system error registration processing on the camera, lidar and millimeter-wave radar; S22, perform time registration and spatial registration processing on the camera, lidar and millimeter-wave radar; S23, initialize spatial gate information; the spatial gate is a cube in space; the spatial gate information includes x-axis distance threshold X1, y-axis distance threshold Y1 and z-axis distance threshold Z1 in a three-dimensional rectangular coordinate system.

3. The target tracking method based on multi-sensor fusion as described in claim 2, characterized in that, The joint monitoring and processing of the target area image information and target area point cloud data to obtain target-assisted tracking location information includes: S321, Perform image target detection processing on the target region image information to obtain the first target location region information; S322, Perform target point cloud matching processing on the target area point cloud data to obtain the second target location area information; S323, Spatial gate detection is performed on the first target location region information and the second target location region information to obtain target auxiliary tracking location information; S324, perform consistency judgment on the target auxiliary tracking position information obtained at several time points to obtain a second judgment result; if the second judgment result is valid, execute S33; if the second judgment result is invalid, execute S31.

4. The target tracking method based on multi-sensor fusion as described in claim 3, characterized in that, The step of performing spatial gate detection on the first target location region information and the second target location region information to obtain target-assisted tracking location information includes: S3231, Obtain the overlapping area information of the first target location area information and the second target location area information; the overlapping area is a three-dimensional cube; the overlapping area information is the side length X2 in the x-axis direction, the side length Y2 in the y-axis direction and the side length Z2 in the z-axis direction of the three-dimensional cube; S3232, perform joint discrimination processing on the overlapping region information and spatial gate information to obtain a third discrimination result; If the third discrimination result is satisfied, the first target location area information and the second target location area information are fused and weighted to obtain the target auxiliary tracking location information; If the third discrimination result is not satisfied, it is confirmed that the target region image information and the target region point cloud data do not contain the same target, and S321 is executed; The expression for the joint discriminant processing is: |X1-X2|≤a1, |Y1–Y2|≤a2, |Z1–Z2|≤a3, |(X2-X1)(Y2-Y1)|≤a4, |(Z2-Z1)(Y2-Y1)|≤a5, Where a1 to a5 are all preset discrimination thresholds; when the overlapping region information satisfies the above five inequalities, the third discrimination result is confirmed to be satisfied; otherwise, the third discrimination result is confirmed to be unsatisfied. The expression for the fusion weighting process is: Wherein, [x1,y1,z1] and [x2,y2,z2] are the center point coordinates of the first target location region information and the center point coordinates of the second target location region information, respectively. For target-aided tracking location information, α1 is the ratio of the area of ​​the overlapping region information to the area of ​​the first target location region information, and α2 is the ratio of the area of ​​the overlapping region information to the area of ​​the second target location region information.

5. A target tracking device based on multi-sensor fusion, characterized in that, The device includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the target tracking method based on multi-sensor fusion as described in any one of claims 1 to 4.

6. A computer-storable medium, characterized in that, The computer storage medium stores computer instructions, which, when invoked by the computer, are used to execute the target tracking method based on multi-sensor fusion as described in any one of claims 1 to 4.

7. An information data processing terminal, characterized in that, The information data processing terminal is used to implement the target tracking method based on multi-sensor fusion as described in any one of claims 1 to 4.

Images